category: literaturenote citekey: schrammfreightforwarderintermediaryrole2012 title: "Freight Forwarder's Intermediary Role in Multimodal Transport Chains" authors: "Schramm, Hans-Joachim" year: 2012 date: 2012-00-00 2012 doi: 10.1007/978-3-7908-2775-0 url: "https://doi.org/10.1007/978-3-7908-2775-0" zotero_key: U4V5DS84 zotero_storage: GA7XU8SP collections: transpordi_planeerimine folder: 001_artiklid firstAuthor: "Schramm, Hans-Joachim"

status: converted

Contributions to Management Science

Hans-Joachim Schramm

Freight Forwarder's Intermediary Role in Multimodal Transport Chains

A Social Network Approach

Contributions to Management Science

Hans-Joachim Schramm

Freight Forwarder's Intermediary Role in Multimodal Transport Chains

A Social Network Approach

Dr. Hans-Joachim Schramm Department of Global Business and Trade Institute for Transport and Logistics Management Nordbergstr. 15 1090 Vienna Austria hschramm@wu.ac.at

Additional material to this book can be downloaded from http://extra.springer.com.

ISSN 1431-1941 ISBN 978-3-7908-2774-3 e-ISBN 978-3-7908-2775-0 DOI 10.1007/978-3-7908-2775-0 Springer Heidelberg Dordrecht London New York

Library of Congress Control Number: 2011943881

© Springer-Verlag Berlin Heidelberg 2012

This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilm or in any other way, and storage in data banks. Duplication of this publication or parts thereof is permitted only under the provisions of the German Copyright Law of September 9, 1965, in its current version, and permission for use must always be obtained from Springer. Violations are liable to prosecution under the German Copyright Law.

The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

Printed on acid-free paper

Physica-Verlag is a brand of Springer Springer is part of Springer Science+Business Media (www.springer.com)

Foreword

This book sums up my research so far concerning the freight forwarding business in multimodal transport chains and was submitted successfully as a Doctoral Thesis at Dresden University of Technology. As research always comes along with at least some external support, acknowledgements are as follows.

First, I express my sincere gratitude to my supervisor Univ.-Prof. Dr. Sebastian Kummer and my local examiner Prof. Dr. Knut Haase for their patience as the finishing of my thesis took unusually long.

Second, thank you to Prof. Michael Browne, Prof. Dr. Herbert Kopfer, Prof. Dr. Herbert Kotzab, Prof. Dr. Wolfgang Stolzle, Prof. Dr. Michael Essig and Prof. Lauri ¨ Ojala being among those delivering useful comments during the various stages of my work.

Third, sincere thanks are given to my present and former colleagues from Dresden University of Technology, University of Applied Sciences bfi Vienna and WU Wirtschaftsuniversitat Wien, especially Dipl.-Verkwirtsch.Nicole G¨ ¨ ulle for raising a sample of Austrian freight forwarding companies engaged in air cargo operations as well as Dr. Dipl.-Verkwirtsch. Mirko Hoppe, MMag. Irene Sudy, Dipl.-Ing. Rainer P. Bogensperger and Dipl.-Buchhandelswirtin (FH) Astrid Fischer for suggestions for improvement while proof-reading the text.

Furthermore, I owe some libraries where I spend hours doing desk research on my topic a debt of gratitude, especially the Dresden State and University Library (SLUB), the University Library at WU Wirtschaftsuniversitat Wien and the ISL ¨ Information Centre at the Institute of Shipping Economics and Logistics at Bremen. Moreover, electronic resources providers like Karlsruhe Virtual Catalogue (KVK) for monographs and ProQuest, ScienceDirect, EBSCOhost, Springer or JSTOR for articles which allowed easy access to publications from all around the world. And last but not least subito (the document delivery service of research libraries in Germany, Austria and Switzerland) which saved my life many times when I was in need for really scarce sources elsewhere not available. Without them I could not have retrieved such a huge amount of references given in this work.

Moreover, I am in debt with Donald E. Knuth and Leslie Lamport for LATEX, MacKichan Software for Scientific Work, and Visio Corporation (as the original vi Foreword

developer of Microsoft Visio) which allowed me to write this thesis in perfect style than with any other word-processing system available. Referring to statistical analysis, SPSS was used, which recently got part of IBM's Business Analytics and Process Optimization organization.

Finally my thesis (and this book) is dedicated to my grandfather Dipl.-Ing. (FH) Hans Haupter who supported me from the very beginning of my academic carrier but could not see the end of my doctorate studies as he died some time ago and (of course) my little daughter Hanna which was growing up to now with her daddy writing hours and hours on a work that never seemed to get completed.

Hans-Joachim Schramm

| 1 | | Introduction |---|-------|-------- | | 1.1 | Freight | | 1.2 | Purpose | | 1.3 | 2 | Legal | and | | in | Multimodal | | 2.1 | The | | | of
View | | | 2.1.1 | | | | | | 2.1.2 | | | | | | 2.1.3 | | | 2.1.4 | | | | | 2.2 | Freight | | | 2.2.1 | | | 2.2.2 | | | | | | 2.2.3 | | | 2.2.4 | | | | | 2.3 | Summary | | | Perspective | 3 | New | | | and | the | | 3.1 | New | | | 3.1.1 | | | 3.1.2 | | | 3.1.3 | | | 3.1.4 | | | | | 1 | | | | ---------------------------------------|-----------------------------------------------------------------------------|-----|--|--|--| | Forwarding
as
a
Research
Topic
So
Far
| 2 | | | | | and
Research
Objectives
| 7 | | | | | Research
Methodology
and
Outline
| | | | | | | Functional
Perspective
of
Freight
Forwarding | | | | | | Transport
Chains | 13 | | | | | Freight
Forwarder
from
a
Legal
and
Functional
Point | | | | | | | 14 | | | | | Actual
Legal
Status
of
Forwarding
Business | | | | | | in
Germany
and
Austria
| 14 | | | | | Comparisons
with
Legal
Status
of
Forwarding | | | | | | Business
in
Other
Countries | 19 | | | | | Freight
Forwarders
from
a
Functional
Point
of
View
| 24 | | | | | Fields
of
Activity
and
Specialization | | | | | | in
Forwarding
Business | 30 | | | | | Forwarding
in
Multimodal
Transport
Chains
| 34 | | | | | Introductory
Notes
and
Basic
Definitions
| 34 | | | | | The
Ocean
Shipping
Industry
in
the
General | | | | | | Cargo
Segment
| 41 | | | | | The
Cargo
Segment
in
the
Airline
Industry
| 48 | | | | | Mapping
of
Inter-organizational
Interaction | | | | | | and
Cargo
Movement
| 54 | | | | | and
Concluding
Remarks
on
the
Functional | | | | | | | 72 | | | | | Institutional
Economics,
Intermediation
Theory, | | | | | | Intermediary
Perspective
| 77 | | | | | Institutional
Economics | 78 | | | | | Ancedents
of
the
New
Institutional
Economics
| 78 | | | | | Common
Assumptions
of
New
Institutional
Economics
| 80 | | | | | Building
Blocks
of
New
Institutional
Economics | 93 | | | | | Conclusions
and
Some
Comparisons
| 106 | | | | | | vii | | | |

viii Contents

	3.2	Market	Microstructure and Intermediation Theory	108
		3.2.1	Market Makers, Specialists, Dealers, Marketers
			and Merchants	111
		3.2.2	Matchmakers and Brokers	120
		3.2.3	Information Production, Guarantors and Certificators	128
		3.2.4	Delegated Expertise, Bargaining, Contracting
			and/or Monitoring	135
		3.2.5	Conclusions and Some Applications	144
	3.3	A	Taxonomy of Intermediaries in Multimodal Transport
		and	Logistic Chains	147
		3.3.1	Commercial Middlemen and Service Providers	148
		3.3.2	Financial Middlemen and Service Providers	151
		3.3.3	Transport Middlemen and Service Providers	152
		3.3.4	Logistics Middlemen and Service Providers	153
		3.3.5	Freight Forwarders as a Strange Hybrid	155
	3.4	Summary	and Some Criticism on the Intermediary Perspective	156

4	Models	of	Network Structure and the Network Perspective	161
	4.1	Network	Modelling in Operations Research
		and	Management Science	163
		4.1.1	Historical and Theoretical Background	163
		4.1.2	Applications	165
		4.1.3	Critical Comments	169
	4.2	Strategic	Networks: Focal Companies with Cooperative
			Relationships	170
		4.2.1	Theoretical Background	170
		4.2.2	Applications and Critical Comments	172
	4.3	Dynamic	Networks: Temporary Virtual Companies
		for	Specific Tasks	174
		4.3.1	Theoretical Background	174
		4.3.2	Applications and Critical Comments	175
	4.4		Entrepreneurial Networks: Specialized Intermediaries
			Economizing on Information Costs	176
		4.4.1	Theoretical Background	177
		4.4.2	Applications	180
		4.4.3	Critical Comments	183
	4.5	Industrial	Networks: Interdependent Business Relationships	184
		4.5.1	Theoretical Background	184
		4.5.2	Applications	188
		4.5.3	Concluding Comments	189
	4.6	Social	Networks: Network Structure, Positions and Relationships	190
		4.6.1	Theoretical Background	191
		4.6.2	Applications	198
		4.6.3	Concluding Comments	199
	4.7		Comparisons and Evaluation	200

Contents ix

5	A	Closer	Look on Contemporary Social Network Analysis	207
	5.1	Roots	of Contemporary Social Network Analysis	207
		5.1.1	Sociometric Analysis and Graph Theory	208
		5.1.2	US-American Community and Industrial Sociology	210
		5.1.3	Manchester School of Social Anthropology	211
		5.1.4	Methodological Breakthroughs	212
	5.2	Levels	of Analysis in Contemporary Social Network Analysis	213
	5.3	Analytical	Instruments of Contemporary Social
		Network	Analysis	214
		5.3.1	Actor's Degree, Density, Multiplexity,
			Heterogeneity and Connectedness	215
		5.3.2	Tie Strength, Durability, Directedness,
			Multiplexity and Connectedness	217
		5.3.3	Network Size, Density, Cohesion, Multiplexity
			and Connectedness	220
		5.3.4	Prominence: Centrality, Centralization, Prestige
			and Hierarchization	222
		5.3.5	Network Constraint and Brokerage Concepts	239
	5.4	Summary	and Concluding Remarks	247

6	Freight		Forwarder's Personal Network in Multimodal
		Transport	Chains: An Empirical Investigation	251
	6.1	Survey	Design	251
		6.1.1	A Generic Multimodal Transport Chain Model	252
		6.1.2	Levels of Analysis and Analytical Instruments	255
	6.2	Sample	Design and Preliminary Analysis	266
	6.3		Presentation of Empirical Results	269
		6.3.1	Sample Characteristics	269
		6.3.2	Personal Network Structure and Position	278
		6.3.3	Intermediary Roles	281
7	Final		Conclusions	285
	7.1		Summarizing Main Findings	285
		7.1.1	What Is the Intermediary Role of Freight
			Forwarders in Multimodal Transport Chains
			Including a Leg by Sea or Air?	285
		7.1.2	Which Body of Theory May Help to Explore
			Its Intermediary Role Theoretically?	286
		7.1.3	Which Kinds of Measures Are Needed to Catch
			Its Intermediary Role Empirically?	286
	7.2		Contribution to Extant Research	287
	7.3	Managerial	Implications	288
	7.4	Limitations	and Suggestions for Further Research	289
			References	293
	Appendix			351

List of Figures

Fig. 1.1	Dissertations in German about freight forwarding	3
Fig. 1.2	Dissertations in German about container shipping	5
Fig. 1.3	Dissertations in German about air cargo operations	6
Fig. 1.4	Structure of this Book	10
Fig. 2.1	Contract forms of affreightment	23
Fig. 2.2	Logistics channel versus logistic chain	35
Fig. 2.3	Forms of transport chains	37
Fig. 2.4	Three approaches to analyze transport chains	39
Fig. 2.5	Cargo Handling and Information Processing Revolution	40
Fig. 2.6	Fleet Structure and Commitment of Airlines	53
Fig. 2.7	Interorganisational Interaction and Cargo Movements
	in General Cargo	55
Fig. 2.8	Interorganisational Interaction and Cargo Movement
	in container trades	59
Fig. 2.9	Forms of container traffic	63
Fig. 2.10	Cargo and documentation flows in sea transport operations	64
Fig. 2.11	Interorganizational interaction and cargo movements
	at Airports	68
Fig. 2.12	Cargo flow and contractual relationships in air
	transport operations	72
Fig. 3.1	Illustrative commercial transactions	98
Fig. 3.2	Efficient governance structures	99
Fig. 3.3	A taxonomy of middlemen and service providers	147
Fig. 4.1	Four modes of organizing economic activity	171
Fig. 4.2	Physical goods movements in multimodal transport chains	182
Fig. 4.3	Information flows in multimodal transport chains	183
Fig. 4.4	Analytical scheme for industrial networks	187

| xii | List of Figures | | |-----|-----------------|--| | | | |

Fig. 5.1 Fig. 5.2	Roots of social network analysis Graphic representation of five different types of brokerage	209 243
Fig. 6.1 Fig. 6.2	Freight forwarder's personal network in export operations Freight forwarder's personal network in import operations	252 253

List of Tables

Table 2.1	Forwarder functions	29
Table 2.2	Forwarder fields of activity in Germany	33
Table 2.3	Environmental effects on freight forwarding business	73
Table 3.1	Contributions to normative principal-agent theory	102
Table 3.2	Principal-agent problems in supply chains	107
Table 3.3	Contributions about intermediaries: Part 1	109
Table 3.4	Contributions about intermediaries: Part 2	110
Table 4.1	Models of network structure	201
Table 5.1	Concepts of social network structure and level of analysis	214
Table 5.2	Analytical instruments for social network structure
	per level of analysis	215
Table 5.3	Measurement concepts for prominence	238
Table 5.4	Cross-classification of centrality measures	238
Table 5.5	Network flow typology of centrality measures	248
Table 5.6	Measures for social capital	249
Table 6.1	Comparison of multimodal transport chains with a leg
	by sea or air	254
Table 6.2	Sampling frame	267
Table 6.3	Survey respondents	268
Table 6.4	Network size and density in export	278
Table 6.5	Network size and density in import	279
Table 6.6	Centrality measures for complete network in export	280
Table 6.7	Centrality measures for complete network in import	280
Table 6.8	Brokerage scores and degree of vertical integration in export	282
Table 6.9	Brokerage scores and degree of vertical integration in import	282

List of Symbols (in order of appearance)

ith point (or actor) in a network
p_i
i, j, k, l
                  (Index for) points or actors in a network with i \neq j \neq k \neq l
N
                  Total number of points (or actors) in a network
                  mth link (or relation) in a network
q_m
                  (Index for) lines (or relations) in a network
m
M
                  Total number of links (or relations) in a network
                  Strength of a link between p_i and p_i, i \neq j
Z_{ii}
                  Strength of an adjacent between p_i and p_i, i \neq j
a_{ii}
                  Distance of a path between p_i and p_i, i \neq j
d_{ii}
                  Geodesic (or shortest path) between p_i and p_j; i \neq j
g_{ii}
h
                  Home country
f
                  Foreign country or country abroad
x
                  (Index for) country at home (h), foreign or abroad (f)
TM_{\star}
                  Transport middleman with x \in \{h, f\}
                  Transport service provider with x \in \{h, f\}
TP_{\nu}
LP_{\star}
                  Logistics service provider x \in \{h, f\}
P_{\rm r}
                  Primary sector firm with x \in \{h, f\}
I_{r}
                  Industrial sector firm with x \in \{h, f\}
                  Distributor with x \in \{h, f\}
D_r
R_{r}
                  Retailer with x \in \{h, f\}
C_{\rm r}
                  Customer with x \in \{h, f\}
d(p_i)
                  Actor's degree of p_i
                  Actor's outdegree of p_i
d_0(p_i)
d_1(p_i)
                  Actor's indegree of p_i
                  Actor's density of p_i
AD(p_i)
z_{ij}(m)
                  Strength of the mth link between p_i and p_i, i \neq j
AM(p_i(m))
                  Multiplexity of actor's contacts of p_i

(Index for) oth distinct subgroup in a network 0 0 Total number of distinct subgroups in a network Strength of a link between $p_i$ and $pj$ of the same subgroup o, $z{ii}(o)$ $i \neq j$ Heterogeneity of contacts of $p_i$ as a member of subgroup o $AH(pi(o))$ NSNetwork size Ego's network size of an ego-centric network $NS{ego}$ NDNetwork density $ND_{alter}$ Alter's network density in an ego-centric network ND'Standardized measure of network density NCNetwork cohesion NMNetwork multiplexity $C_D(p_k)$ Degree-based absolute measure of point centrality of $p_k$ $C'_D(p_k)$ Degree-based relative measure of point centrality of $p_k$ Total number of geodesics between $p_i$ and $pi$ , $i \neq j$ $G{ii}$ $G_{ii}(p_k)$ Total number of geodesics between $p_i$ and $p_j$ that contain $p_k$ , $i \neq j \neq k$ $C_B(p_k)$ Betweeness-based absolute measure of point centrality of $pk$ $C'{R}(p_{k})$ Betweeness-based relative measure of point centrality of $p_k$ $C_C(p_k)^{-1}$ Closeness-based absolute measure of point centrality of $p_k$ $C'_C(p_k)$ Closeness-based relative measure of point centrality of $p_k$ $C_R(p_k)$ Rank-based measure of centrality or eigenvector centrality of $p_k$ C $(N \times 1)$ vector of rank-based centrality measures for each actor Z $(N \times N)$ matrix of links between actors in a network λ Constant, largest positive eigenvalue of Z $C_R(p^)$ Maximum value of all $C_R(p_k)$ X (Index for) measures of point centrality based on degree (D), closeness (C), betweeness (B) or rank (R) $C_X(p_k)$ Point centrality measures of $p_k$ for $X \in {D, C, B, R}$ $C_X(p^)$ Maximum value of all $C_X(p_k)$ for $X \in {D, C, B, R}$ Measure of centralization for $X \in {D, C, B, R}$ $C_X$ Degree-based measure of graph centrality $C_D$ $C_D(p^)$ Maximum value of all $C_D(p_k)$ Betweenness-based measure of graph centrality $C_B$ $C'_R(p^)$ Maximum value of all $C'{R}(p{k})$ Closeness-based measure of graph centrality $C_C$ $C'_C(p^)$ Maximum value of all $C'_C(p_k)$ $C_R(p^)$ Maximum value of all $C_R(p_k)$ $\overline{C}_D$ Average of actor's degrees $\overline{C}'_D$

Average standardized degree Variance of actor's degrees

$\overline{C}_C$	Avarage of actor's alegeness
$\frac{C_C}{C'}$	Average of actor's closeness
$ \overline{C}{C}^{'} $ $ S{C}^{2} $ $ \overline{C}{B} $ $ \overline{C}{B}^{'} $ $ S{B}^{2} $ $ \overline{C}{R} $	Average standardized closeness Variance of actor's closeness
$\frac{S_{\bar{C}}}{C}$
$\frac{C_B}{C'}$	Average of actor's betweeness
$C_B$	Average standardized betweeness
$\frac{S_B^2}{\overline{S}}$	Variance of actor's betweeness
	Average of rank-based point centrality
$S_R^2$	Variance of rank-based point centrality
$P_D(p_k)$	Degree-based prestige of $p_k$
$P_D'(p_k)$	Standardized degree prestige of $p_k$
$I_k$	Number actors in the influence domain of $p_k$
$P_P(p_k)$	Proximity-based prestige of $p_k$
$P_R(p_k)$	Rank-based prestige of $p_k$
P	$(N \times 1)$ vector of rank-based prestige for each actor
$\Phi$	$(N \times N)$ identity matrix
$P_Y$	Network hierarchization with $Y \in {D, C, R}$
Y	(Index for) prestige measures based on degree $(D)$ ,
	proximity $(P)$ or rank $(R)$
$\frac{P_Y(p^*)}{\overline{z}}$	Maximum value of all $P_Y(p_k)$
$\overline{P}_{Y}$	Average of prestige measures with $Y \in {D, P, R}$
$S_{P_Y}^2$	Variance of prestige measures with $Y \in {D, P, R}$
$NS_k$	Effective network size of an actor $k$
$p_{kj}$	Proportion of $k$ 's network time and energy invested in a contact with $j \neq i \neq k$
$v_{ji}$	Marginal strength of j's relations with contact $i \neq j \neq k$
$p_{ki}$	Proportion of $k$ 's network time and energy invested in a contact
	with $i \neq j \neq k$
$p_{ji}$	Proportion of j 's network time and energy invested in a contact with $i \neq j \neq k$
$c_{ki}$	Contact-specific constraint of $k$ engaged in a relationship with $i \neq j \neq k$
$NC_k$	Network constraint index of an actor $p_k$
$T_k$	Total raw brokerage score of an intermediary $p_k$
$w_{1k}$	Individual raw within-group brokerage score of $p_k$ of the local broker type
$w_{0k}$	Individual raw within-group brokerage score of $p_k$ of the itinerant broker type
$b_{01k}$	Individual raw between-group brokerage score of $p_k$ of the gatekeeper type
$b_{10k}$	Individual raw between-group brokerage score of $p_k$

of the representative type

$b_{0k}$	Individual raw between-group brokerage score of $p_k$ of the liaison type
o(ij)	Common membership of $p_i$ , $p_j$ and/or $p_k$ in a subgroup $o$ according to brokerage type with $i \neq j \neq k$
$z_{ij}\left(o\right)$	Strength of a link between $p_i$ and $p_j$ of the same subgroup $o$ with $i \neq j$
$w_1(ij)$	Within-group brokerage between $p_i$ and $p_j$ of the local broker type with $i \neq j$
$w_0(ij)$	Within-group brokerage between $p_i$ and $p_j$ of the itinerant broker type with $i \neq j$
$b_{01}(ij)$	Between-group brokerage between $p_i$ and $p_j$ of the gatekeeper type with $i \neq j$
$b_{10}(ij)$	Between-group brokerage between $p_i$ and $p_j$ of the representative type with $i \neq j$
$b_0(ij)$	Between-group brokerage between $p_i$ and $p_j$ of the liaison type with $i \neq j$
$p_{ij}(p_k)$	Number of two-step paths with $p_k$ between $p_i$ and $p_j$ with $i \neq j \neq k$
$T_{\nu}^*$	Total partial brokerage score of an intermediary $p_k$
$Tk^$ $w{1k}^$	Individual partial within-group brokerage score of $p_k$ of the local broker type
$w_{0k}^{*}$	Individual partial within-group brokerage score of $p_k$ of the itinerant broker type
$b_{01k}^{*}$	Individual partial between-group brokerage score of $p_k$ of the gatekeeper type
$b_{10k}^{*}$	Individual partial between-group brokerage score of $p_k$ of the representative type
$b_{0k}^*$	Individual partial between-group brokerage score of $p_k$ of the liaison type
$w_k$	Total within-group raw brokerage score of an intermediary $p_k$
$w_k^*$	Total within-group partial brokerage score of an intermediary $p_k$
$b_k$	Total between-group raw brokerage score of an intermediary $p_k$
$b_k^*$	Total between-group partial brokerage score of an intermediary $p_k$
$W_1$	Global within-group raw brokerage score of the local broker type
$W_0$	Global within-group raw brokerage score of the itinerant broker type
$B_{01}$	Global between-group raw brokerage score of the gatekeeper type
$B_{10}$	Global between-group raw brokerage score of the representative type
$B_0$	Global between-group raw brokerage score of the liaison type
W	Global within-group raw brokerage score
B	Global between-group raw brokerage score

$W_1^*$	Global within-group partial brokerage score of the local broker type
$W_0^*$	Global within-group partial brokerage score of the itinerant broker type
$B_{01}^*$	Global between-group partial brokerage score of the gatekeeper type
$B_{10}^*$	Global between-group partial brokerage score of the representative type
$B_0^$ $W^$	Global between-group partial brokerage score of the liaison type Global within-group partial brokerage score
B*	Global between-group partial brokerage score
$GD_{hf}$	Transport distance between country $h$ and $f$
$GD'_{hf}$	Transport time between country $h$ and $f$
	• • •
$CD_{hf}$	Cultural distance index between country $h$ and $f$
	based on average scores (Index for) cultural dimensions of power distance ( PDI ),
С	uncertainty avoidance ( UAI ), individualism ( IDV ) and masculinity ( MAS )
$CI_{ch}$	Values of cultural dimensions at home
	with $c \in {PDI, UAI, IDV, MAS}$
$CI_{cf}$	Values of cultural dimensions abroad
c)	with $c \in {PDI, UAI, IDV, MAS}$
$CV_c$	Overall variance in the sample of each cultural dimension
_	with $c \in {PDI, UAI, IDV, MAS}$
$CD'_{hf}$	Cultural distance index between country $h$ and $f$
nj	based on Euclidean concept
e	(Index for) country indexes of economic freedom ( IEF ),
	freedom of the world (EFW), corruption perception (CPI) and
	globalization (GI)
$ED_{hf}$	Economic distance index between country $h$ and $f$
	based on average scores
$EI_{eh}$	Values of country indexes at home
	with $e \in {IEF, EFW, CPI, GI}$
$EI_{ef}$	Values of country indexes abroad
,	with $e \in {IEF, EFW, CPI, GI}$
$EV_e$	Overall variance in the sample of each country index
•	with $e \in {IEF, EFW, CPI, GI}$
$ED'_{hf}$	Economic distance index between country $h$ and $f$ based on
",	Euclidean concept
$VI_k$	Degree of vertical integration of an intermediary k
	·

List of Abbreviations

3PL Third-Party Logistics Provider 4PL Fourth-Party Logistics Provider

ABB Allgemeine Beforderungsbedingungen ¨

AB-SVV Allgemeine Bedingungen des Schweizerischen

Spediteurs-Verbandes

ACL Atlantic Container Lines

ACT Associated Container Transportation ADHGB Allgemeines Deutsches Handelsgesetzbuch ADSp Allgemeine Deutsche Spediteursbedingungen

AGBSp Allgemeine Geschaftsbedingungen im ¨ osterreichischen ¨

Speditionsgewerbe

ALFA Automatisiertes Luftfrachtabwicklungsverfahren AOSp Allgemeine ¨ Osterreichische Spediteursbedingungen ¨

APM Arnold Peter Møller

APL-NOL American President Lines - Neptun Orient Lines

ATLAS Automatisiertes Tarif- und Lokales Zollabwicklungssystem

AWB Air waybill B/L Bill of Lading

BAX Burlington Air Express BGBl Bundesgesetzblatt

BGL Bundesverband G¨uterkraftverkehr, Logistik und Entsorgung

BIFA British National Freight Association BLG Bremer Lagerhaus-Gesellschaft

BPAWG Business Process Analysis Working Group BSL Bundesverband Spedition und Logistik

CAGE Cultural, Adminstrative, Geographical and Economic Distance

CBL Combined Bill of Lading

xxii List of Abbreviations

CCS Cargo community system CdC Code de Commerce

CEFACT Centre for the Facilitation of Procedures and Practices

for Administration, Commerce and Transport

CFR Cost and freight (INCOTERM 2000/2010)

CIF Cost, insurance and freight (INCOTERM 2000/2010) CINA Commission Internationale pour la Navigation Arienne CIP Carriage and insurance paid to (INCOTERM 2000/2010)

CMI Comit Maritime International

COD Cash on Delivery

CoFR Code of Federal Regulations COSCO China Ocean Shipping Company CPI Corruption Perception Index

CPM Critical Path Method CPS Cargo Portal System

CPT Carriage paid to (INCOTERM 2000/2010)

CTO Combined transport operator D/C Documentary collection

DAF Delivered at frontier (INCOTERM 2000)

DAKOSY Daten-Kommunikationssystem

DAP Delivered at place (INCOTERM 2010) DAT Delivered at terminal (INCOTERM 2010)

DB Deutsche Bahn

DBGB Deutsches B¨urgerliches Gesetzbuch DBGBl Deutsches Bundesgesetzblatt DBH Datenbank Bremische Hafen ¨

DDP Delivered duty paid (INCOTERM 2000/2010) DDU Delivered duty unpaid (INCOTERM 2000) DES Delivered ex ship (INCOTERM 2000)

DEQ Delivered ex quay duty paid (INCOTERM 2000)

DHL Dalsey-Hillblom-Lynn

DIN Deutsches Institut f¨usr Normung DPWN Deutsche Post World Net DRGBl Deutsches Reichsgesetzblatt

DSLV Deutscher Speditions- und Logistikverband

DVB Deutsche Verkehrsbank DVZ Deutsche Verkehrszeitung EACP European Air Cargo Programme

EC European Community

ECAA European Common Aviation Area ECT Europe Combined Terminals

List of Abbreviations xxiii

EDI Electronic Data Interchange EEC European Economic Community EFTA European Free Trade Association EFW Economic Freedom of the World Index EHK Euler Hermes Kreditversicherungs AG

EMC Export management company ETC Export trading company ETI Enabling Trade Index EU European Union

EXW Ex works (INCOTERM 2000/2010)

FAS Free alongside ship (INCOTERM 2000/2010)

FBL FIATA Bill of Lading

FCA Free carrier (INCOTERM 2000/2010)

FCL Full-container-load

FCR FIATA Forwarders Certificate of Receipt FCT FIATA Forwarders Certificate of Transport

FedEx Federal Express

FIATA Fdration Internationale des Associations

de Transitaires et Assimils

FOB Free on board (INCOTERM 2000/2010)

FTL Full-truck-load

GAP German Airfreight Partners

GATT General Agreement on Tariffs and Trade GBlO Gesetzblatt der Republik ¨ Osterreich ¨

GDP Gross domestic product

GERT Graphical Evaluation and Review Technique

GF-X Global Freight Exchange GI Globalization Index GNP Gross national product GSA General sales agent

GSSA General sales and service agent GT&C General terms and conditions GTC General trading company HGB Handelsgesetzbuch HMM Hyundai Merchant Marine HPH Hutchison Port Holding HVertG Handelsvertretergesetz

IAPH International Association of Ports and Harbours

IATA International Air Transport Association IATA-CASS IATA Cargo Accounts Settlement System IBM International Business Machines Corporation xxiv List of Abbreviations

ICAO International Civil Aviation Organization ICC International Chamber of Commerce

ICH IATA Clearing House

IDV Individualism versus collectivism IEC Immediate effects centrality IEF Index of Economic Freedom IFF International freight forwarder

IGLU Interessensgemeinschaft Luftfracht GmbH IMD Institute for Management Development IMP Industrial Marketing and Purchasing INCOTERMs International Commercial Terms IOT Industrial Organization Theory

IOU I owe you

ISO International Standardization Organization

ITF Intermediation Theory of the Firm ITZ Internationale Transportzeitung KLM Koninklijke Luchtvaart Maatschappij

km Kilometre kn Knots

KVK Karlsruher Virtueller Katalog

L/C Letter of Credit

LACES London Airport Cargo Electronic Data Processing System

LCL Less-than-container-load LLP Lead logistics provider LPI Logistics Performance Index

LTL Less-than-truckload

LTO Long-term versus short-term orientation

MaklerG Maklergesetz

MAS Masculinity versus femininity MEC Mediative effects centrality

MISC Malaysian International Shipping Corp.

MLS Multiple listing service

MMT Market Microstructure Theory MOL Mitsui Osaka Shoshen Kaisha Line

mph Miles per hour

MPM Meta Potential Method MS Management Science

MSC Mediterranean Shipping Company MTO Multimodal transport operator NAFTA North-American Free Trade Area

List of Abbreviations xxv

NCBFAA National Customs Brokers and Forwarders Association

of the United States

NCE (Neo)classical Economics NES New Economic Sociology NET Network Exchange Theory NIE New Institutional Economics

nm Nautical miles

NSAB Nordisk Speditørforbunds Alminnelige Bestemmelser

NSF Nordiskt Speditor F ¨ orbund ¨ NVO Non-vessel operator NVOC Non-vessel owning carrier

NVOCC Non-vessel operating common carrier

NYK Nippon Yushen Kaisha

OCL Overseas Container Limited OeKB Osterreichische Kontrollbank AG ¨ OEM Original Equipment Manufacturer

OG Obligationenrecht

OHGB ¨ Osterreichisches Handelsgesetzbuch ¨

OIE Old Institutional Economics OOCL Oriental Overseas Container Line

OR Operations Research

OTI Ocean transport intermediary

P&O Peninsular and Oriental Steam Navigation Company

PAT Principal Agent Theory

PDI Power distance

PERT Programm Evaluation and Review Technique

PRT Property Rights Theory PSA Port of Singapore Authority

RO/RO Roll-on/roll-off

SABRE Sales And Business Reservations Electronically

SAS Scandinavian Airlines System SET Social Exchange Theory SLUB Staats- und Landesbibliothek

sm Statute miles

SNA Social Network Analysis

SOFIA Systme d'ordinateur du Frt International Arien

STC Standard Trading Conditions

SWB Sea waybill

SWIFT Society of Worldwide Interbank Financial Telecommunication

TACT The Air Cargo Tariff TCT Transaction Cost Theory xxvi List of Abbreviations

TEC Total effects centrality TEU Twenty foot equivalent unit

TLF Fdration des Entreprises de Transport et Logistique de France

TMWG Techniques and Methodologies Working Group

TNT Thomas Nationwide Transport

TPG The Post Groep

TRAXON Tracking and Tracing Online UAI Uncertainty avoidance

UASC United Arab Shipping Company UGB Unternehmensgesetzbuch

ULD Unit load device

UN United Nations Organization

UN/ECE United Nations Economic Commission for Europe UNCTAD United Nations Conference on Trade and Development

UPS United Parcel Service US United States (of America) VAN Value Added Network

VAT Value Added Tax VBGL Vertragsbedingungen f¨ur den G¨uterkraftverkehrs- und

Logistikunternehmer

VBSp Verein Bremer Spediteure VDS Verein Deutscher Spediteure VHSp Verein Hamburger Spediteure WACO World Cargo Organization WIFI Wirtschaftsforderungsinstitut ¨ WIN Worldwide Independent Network WKO Wirtschaftskammer Osterreich ¨ WTO World Trade Organization

Chapter 1 Introduction

"A work of this kind is really never finished; one only calls it finished because one has done all that is possible in the time and the circumstances."

(Johann Wolfgang von Goethe (1970 [1786–1788]), Italian Journey, 16/03/1787, translated by W.H. Auden and E. Meyer)

In this book, the business of international freight forwarding is examined both from a theoretical and empirical point of view with a special emphasis on multimodal transport chains including sea or air transport operations - a truly multi-billion dollar business, where international freight forwarders get an ever increasing share of the pie and generate highly attractive returns on capital.1 There, the freight forwarder is always considered as "The Architect of Transport," but this intermediary role seems not to be well theoretically reflected in research up to now. Therefore, relevant theoretical concepts from economic theory and economic sociology are employed to draw both an intermediary and a network perspective of freight forwarding in order to provide a better understanding of this kind of transportation business. Furthermore, the position of freight forwarder such an inherent network structure of multimodal transport chains is explored by mapping relationship patterns in a stylized model framework which is applied to a questionaire based sample raised among freight forwarders in multimodal transport chains with a leg by sea or air in Germany (especially from Hamburg, Bremen, and Bremerhaven) as well as Austria in 2003.

Writing a book in English about freight forwarding seemed to be very uncommon for most of my German and Austrian colleagues (at least at the beginning), but for me as the author, there are some striking reasons to do this I want to declare briefly. First, since my time as an research and teaching assistant at Dresden University of Technology I got in contact with many colleagues in the field of university education outside German speaking countries being very interested in my focus of research work and so I decided to offer my contribution to a greater audience. Second, over

¹See e.g. MergeGlobal (2008a,b,c).

2 1 Introduction

the years my assignments came along with more and more lecturing as well as research work in English. Augmenting this experience, I saw that being a citizen of the European Union, there is a need for a more "unifying" approach in writing about German and Austrian freight forwarding issues. Last but no least, some sections of this work are heavily influenced by Anglo-Saxon scholarship, especially in the case of theoretical treatments in New Institutional Economics, Intermediation Theory of the Firm and Social Network Analysis. Writing about these topics in German, there is always a danger to get "lost in translation," because at least some of the wording used cannot be translated in a straightforward way without alternating their original meaning. Therefore, German speaking authors nowadays often "fall in love" to write German texts with a considerable high percentage of Anglicisms and ending up then with a work being neither German nor English but something in between. I definitely did not want to end up with such a piece of work, so I decided to write it down in English as a contemporary "lingua franca."²

In the remaining sections, findings from an initial literature review about freight forwarding, container and air cargo operation issues as well as purpose, objectives, methodology, and the structure of this work are outlined.

1.1 Freight Forwarding as a Research Topic So Far

In order to motivate purpose, objectives, methodology, and outline of this work, a rough overview about the status quo of research in freight forwarding (as well as container shipping and air cargo operations) as a research topic will be given first to show clearly to what extent it is theoretically reflected in scientific literature so far. In the following, a focus is set on dissertations (or works on doctoral or PhD level) originating from Germany, Austria and Switzerland before a brief outlook about Anglo-Saxonian literature is given.

Reviewing freight forwarding related literature in German, a considerable number of 97 dissertations from Germany, Austria and Switzerland about the freight forwarding business with mainly non-legal focus could be identified as shown in Fig. 1.1. ³ Whereas these dissertations cover a widespread variety of topics from the

²Following Welch et al. (2001, p. 195), "English has evolved as the most commonly used corporate language, reflecting its more general position as a type of lingua franca of international business." According to Encyclopædia Britannica (2002), the term "lingua franca" originally means frankish language and is an "auxiliary or compromise language used between groups having no other language in common." It was first applied to a jargon or pidgin based on southern French and Italian, developed in the Middle Ages by crusaders and traders in the eastern Mediterranean.

³This review of literature from German speaking countries is mainly based on a bibliography of forwarder-related scientific literature by M¨uller (1990), a review survey of dissertations in the field of logistics from 1990 to 1997 by Large and Stolzle (1999), the annual bibliographic notes ¨ in the Jahrbuch der Logistik from 1990 to 2007 as well as library catalogues of monographs from Deutsche Bibliothek Frankfurt-on-Main, Deutsche B¨ucherei Leipzig, Staatsbibliothek zu Berlin – Preussischer Kulturbesitz, Osterreichische and Schweizerische Nationalbibliothek. Fur- ¨

Goerres (1909)* Heinrich (1913) Silberstein (1914)			Zöllner (1990) Thaler (1990) Kleer (1991) Niegel (1991) Severin (1992) Rendez (1992) Gollnick (1992) Freichel (1992) Hebert (1992) Fuhrmann (1993) Otto (1993) Simon (1993) Stahl (1995a)* Heierli (1995) Lublow (1995)
Reinmüller (1919) Kürbs (1920)*			Falk (1995) Wittenbrink (1995)
Vadnai (1921)			Jacob (1996)
Vonnegut (1921)			Niebuer (1996)
Thost (1921)			Steffen (1996)
Meyer (1922)			Waldenberger (1997)
Pontow (1922)	Vogel (1947)		Rumpf (1997)
Vogler (1924) Scharlibbe (1925)*	Lehmann (1949)	Schott (1971)	Richter (1997) Wlcek (1998)
Höppel (1925)	Kirchner (1950) Ullmann (1950)	Ringshausen (1971) Lösch (1972)	Wolf (1999)
Schneider (1925)	Ehret (1951)	Diehl (1973)	Becker (1999)*
Tiebel (1925)	Kobelnig (1952)	Freese (1975)	Engelsleben (1999)
Grass (1927)	Fritz (1955)	Matthäi (1979)	Erdmann (1999)
Dietrich (1929)	Meyer (1957)	Seiler (1981)	Nöcker (2002)
Hansen (1928)	Schmitz (1957)	Offergeld (1984)	Pankratz (2002)
Blank (1929)	Bader (1959)	Krass (1984)	Uhlmann (2004)
Böllinger (1930)	Scharl (1963)	Quintero (1985)*	Schmidthöfer (2004)
Weisel (1934)	Donndorf (1963)	Eckstein (1985)	Schmitz (2005)
Dietz (1935)	Keller (1966)	Seiffert (1986)	Helf (2005)
Stern (1938)*	Gretzschel (1968)*	Schumacher (1987a)	Lohre (2005) Weddewer (2007)
Eggli (1942) Lehrner (1942)	Lamprecht (1969)	Dehn (1987)*	Krajewska (2007)
Palmhofer (1944)	Wecker (1969) Kloetz (1969)*	Drechsler (1988) Welker (1988)	Rieck(2008)
1909 to 1945: 26	1946 to 1969: 17	1970 to 1989: 16	1990 to 2008: 38

Fig. 1.1 Dissertations in German about freight forwarding

freight forwarder business, 14 of them are worth to highlight with an asterix here, because they explicitly coped with the international freight forwarding business. In the very first contribution, Goerres (1909) mainly discussed in depth tariff structure

ther, a snowball sampling method was applied to all monographs physically available for the author in order to find other works not cited or filed at the above mentioned sources.

4 1 Introduction

and paperwork needed for cross-border railway transports. K¨urbs (1920) observed freight forwarding operations between France and Germany, whereas Scharlibbe (1925) and Stern (1938) described thoroughly the freight forwarder's business at seaports. Ullmann (1950) and Kirchner (1950) gave an overview on the freight forwarding business in Austria after World War II, together with a detailed description of historical, legal, functional and organizational issues. Furthermore, Gretzschel (1968), Kloetz (1969) and Quintero (1985) focused on organizational topics in export-import operations of international freight forwarding in socialist countries. Moreover, Dehn (1987) assessed chances of freight forwarding companies offering distribution services in air cargo as well as sea freight operations. Thaler (1990) discussed the economic impacts on European freight forwarding companies borne by the establishment of the European Common Market, followed by Simon (1993) and Stahl (1995a) further investigating freight forwarder transport networks in groupage service on a European level. Finally, Becker (1999) analyzed strategic alliances among freight forwarders in the air cargo industry.

Looking at the literature outside German-speaking countries, freight forwarding seems to be of less scientific interest. Works at a doctoral or PhD level in English language found coping directly with forwarding issues or freight forwarderrelated third-party logistics since 1980 are e.g. Okeafor (1982), Westfall (1987) and Rahman (1989) in the United States, Stone (1998) and Ford (2001) in the United Kingdom as well as Hertz (1993), Andersson (1995, 1997), Ludvigsen (2000), Berglund (2000) and Stefansson (2004) in Sweden or Lommelen (2004) in Belgium.4 Referring to peer reviewed journal articles, early contributions are e.g. Davies (1981) about exporter and freight forwarder relationships in the United Kingdom, Pope and Thomchick (1985) or Sherwood and Burns (1992) about freight forwarders and "non-vessel operating common carriers" (NVOCCs) in the United States. Furthermore, Ozsomer et al. (1993) presented an expert system tool that helps to evaluate and select international freight forwarders and Paul R. Murphy, James M. Daley and Douglas R. Dalenberg published a series of surveys about export management and international freight forwarding issues resulting in contributions like Murphy et al. (1992) and Murphy and Daley (1994, 1995, 1996a,b, 2000, 2001). Yet, other recent contributions are Lu and Dinwoodie (2002) about international freight forwarding services in the People's Republic of China, Lemoine and Dagnæs (2003) dealing with the dynamics of internationalization and globalization of European freight forwarding and logistics service providers, Bowen and Leinbach (2004) examining the air freight forwarding industry with a special focus on South-East Asia as well as Lai and Cheng (2004) with a survey of the freight forwarding business in Hong Kong, Markides and Holweg (2006) about diversification of international freight forwarders in the United Kingdom

⁴See Stock (1988/1989), Stock and Luhrsen (1993), Stock (2001) and Stock and Broadus (2006) for a review and bibliographic notes for dissertations filed at United Microfilms International from 1970 to 2004 or Gubi et al. (2003) with a review of dissertations written in the Scandinavian countries from 1990 to 2001.

			Hinricher (1993)
	Baumöller (1970)		Geuther (1990)
	Wyremba (1970)		Lieb (1990)
	Egert (1970)		Cao (1991)
	Schuh (1971)		Mira (1993)
	Käselau (1971)	Scheibe (1980)	Opatz (1994)
	Walter (1971)	Philipp (1980)	Bortfeld (1995)
	Zinnecker (1972)*	Jehle (1981)	Gürsel (1996)
	Mönch (1972)	Schmitt (1981)	Exler (1996)
	Baumgarten (1972)	Jarke (1981)	Ordemann (1996)
	Ramcke (1973)	CardonneMolina (1981)	Malchow (1996)
	Huch (1973)	Ernst (1982)	Kasprzak (1997)
	Kranz (1973)	Lutze (1982)	Ponta (1997)
	Detrez (1973)	Krieger (1984)*	Alicke (1999)
	Nguyen (1973)	Lasser (1984)	Pumpe (2000)
	Broers (1974)	Moral Garcia (1984)	Hautau (2002)
	Monzel (1974)	Richter (1985)	Kramer (2004)
	Meyer (1975)	Behrens (1986)	Reise (2005)
	Bartscher (1976)	Mester (1986)	Ninnemann (2005)
	Kühnbaum (1976)	Kunzmann (1988)	Hein (2007)
	Bopp (1976)	Shalaby (1988)	Schönknecht (2007)
	Trettin (1977)	Schott (1989)	Hildebrand (2008)
Mathiesen (1969)	Knepper (1978)	Sinnig (1989)	Meier (2008)
Rackwitz (1969)	Gerckens (1978)	Grabellus (1989)	Lauf (2008)
1946 to 1969: 2	1970 to 1979: 23	1980 to 1989: 19	1990 to 2008: 24

Fig. 1.2 Dissertations in German about container shipping

and Neiberger (2008) describing spatial and organizational patterns of European air cargo freight forwarding. Another stream of literature closely related to the freight forwarding business is the booming field of third- and fourth-party logistics with numerous contributions since the 1990s well documented in Razzaque and Sheng (1998), Maloni and Carter (2006), Selviaridis and Spring (2007) and Marasco (2008).

In addition to freight forwarding related contributions, dissertations from German speaking countries related to container shipping and air freight operations were surveyed, too, again with a focus on merely economic issues.⁵ In Figs. 1.2 and 1.3, all dissertations originating from Germany, Austria or Switzerland found are summarized with a total of 105 contributions since 1931.⁶

⁵This implies, that dissertations coping solemnly with topics from engineering, polity or law were excluded. As the scope of this work is international freight forwarding, a more rigorous review including works at a doctoral or PhD level of non-German country origin was not conducted but Stock (1988/1989), Stock and Luhrsen (1993), Stock (2001) and Stock and Broadus (2006) list a total of 104 contributions filed at United Microfilms International from 1970 to 2004 covering about the same topics in the field of water and air cargo transportation as in the case of dissertations of German country origin.

⁶Notably, Zinnecker (1972) and Krieger (1984) were dealing both with air cargo operations as well as container shipping, but they are counted once.

6 1 Introduction

Fig. 1.3 Dissertations in German about air cargo operations

The first dissertations in German language dealing explicitly with container shipping issues can be dated back to 1969, shortly after the first regular intercontinental container shipping services were introduced by Sea-Land Service.⁷ In the 1970s and 1980s, contributions mushroomed, covering various aspects of container shipping. Most of them were of merely general interest, describing and discussing en detail container shipping operations or observing the impacts of containerization on shipping companies and/or seaports. Others, like Baumoller (1970), Schuh ¨ (1971), Kaselau (1971), Bartscher (1976), Schmitt (1981) or Krieger (1984) focused ¨ on pricing or controlling issues. Furthermore, authors like Ramcke (1973), Huch (1973), Jehle (1981), Jarke (1981), Ernst (1982), Kunzmann (1988), Schott (1989), Cao (1991), Bortfeld (1995), G¨ursel (1996), Ordemann (1996), Kasprzak (1997), Ponta (1997), Alicke (1999), Pumpe (2000), Hein (2007) or Meier (2008) employed mathematical approaches to improve container transport operations.8

From the time before 1945, only one dissertation in German language dedicated to air cargo transport operations could be identified. But other authors like W¨ust (1927), Stahlberg (1931), Maetz (1937) or Ziegler (1938) raised both passenger and cargo transport issues. After 1945, a steadily increasing stream of contributions from Wagner (1952) to Linz (2008) either commenting actual developments in the air cargo industry or treating special problems in air cargo transport operations

⁷See Sect. 2.2.2.1 with further comments. Moreover at that date, maritime economics was a rather new emerging research field with isolated works before the 1960s, as Goss (2002) noted.

⁸See Sect. 4.1.2 for a more thoroughly treatment of this issue.

can be observed.⁹ Among these various special treatments, Seidelmann (1971) or Zinnecker (1972) discussing air cargo unitization issues, Bey (1972), Krebs (1974), Arndt and Kinkeldei (1978), Terhorst (1992) or Frye (2003) surveying ground handling operations and Schneider (1993), Bachmeier (1999) or Gutthal (1999) focusing on integrated air cargo transport chains can be highlighted.

To conclude, the total amount of dissertations about freight forwarding (as well as container shipping and air cargo operations) originating from German speaking countries count up to considerable volume of scientific research work. But especially contributions in the field of multimodal transport chains with a leg by sea or air seems to be rare with only a few exceptions like Scharlibbe (1925), Stern (1938), Gretzschel (1968), Kloetz (1969), Quintero (1985) or Dehn (1987) about international freight forwarding in maritime shipping and Becker (1999) or (again) Dehn (1987) in air cargo operations. Referring to other contributions about international freight forwarding outside German speaking countries, especially Okeafor (1982), Westfall (1987), Rahman (1989), Hertz (1993), Ludvigsen (2000), Ford (2001) and Lommelen (2004) can be mentioned.

1.2 Purpose and Research Objectives

Seeing the rather underdeveloped field of contributions about international freight forwarding in multimodal transport chains with a leg by sea and air, the purpose of this work is to explore international freight forwarding beyond a merely descriptive approach and look on the topic from both a strict theoretical as well as empirical point of view. Accordingly, the research question(s) in short are as follows:

1. What is the intermediary role of freight forwarders in multimodal transport chains including a leg by sea or air?
2. Which body of theory may help to explore its intermediary role theoretically?
3. Which kinds of measures are needed to catch its intermediary role empirically?

The term "role" in the context of this work is derived from its well defined meaning in role theory as covered by anthropologists, sociologists and psychologists, but it can be merely regarded as a classificatory concept than a theory. After first explicit treatments in the 1930s by Mead (1934), Moreno (1934) and Linton (1936), a language of technical terms was developed during the 1950s to 1960s.10

Probably the most employed basic definition of role and its relationship to position and status stem from the anthropologist Ralph Linton, who defined a status as "[:::] a position in a particular pattern. [:::] It represents his position in relation

⁹In addition to this, there is a bulk of dissertations dealing with the airline industry as a whole. But in contributions like Berendt (1961), Wolf (1967) or Buchwald (1973), air cargo transport operations are of minor interest.

¹⁰Cf. Zukin and DiMaggio (1990, p. 11–20), Dahrendorf (1960, p. 12–14), Thomas and Biddle (1966a), Turner (1985) or Biddle (1986).

8 1 Introduction

to the total society" (Linton (1936, p. 113)).¹¹ Furthermore, a status can be either ascribed or achieved with the first type assigned to individuals from birth without reference to their innate differences or abilities and the second to be acquired through competition and effort.12 To define the term role, he simply asserted that it "[:::] presents the dynamic aspect of a status. The individual is socially assigned to a status and occupies it with relation to other statuses [:::] Role and status are ˆ quite inseparable, and the distinction is only of academic interest. There are no roles without statuses or statuses without r ˆ oles" (Linton (1936, p. 114)). ˆ ¹³ As a status always comes along with a collection of rights and duties (or obligations) associated with a particular position, a role determines an individual's behaviour as a social pattern expected by others (or the society in general).14 While sharing Linton's (1936) ideas of role and status definition, Merton (1957, p. 110) pointed to the fact that each person in society not only occupies multiple statuses with "each social status involves a single associated role, but an array of roles." He named it role-set and defined it as "the work complement of role-relationships in which persons are involved by virtue of occupying a particular social status" (ibid). Accordingly, Merton (1957, p. 111) designated a status-set to be a "complement of distinct statuses of a person, each of these in turn having its own role-set." Later on, sociologists like Gross et al. (1958, p. 48), Dahrendorf (1960, p. 45–46) or Thomas and Biddle (1966b, p. 28–29) suggested that position rather than status is a better term because "status connotes the idea of differential ranking among a set of person or social locations, whereas the more neutral term, position, does not" (Gross et al. (1958, p. 48)). Moreover, Gross et al. (1958, p. 48–69), Dahrendorf (1960, p. 46– 52) or Thomas and Biddle (1966b, p. 29–31) showed, how these initial definitions can be further enlarged to a language for role analysis. But without elaborating on further details, it is useful to restate these basic relationships between role, position and status briefly as follows:15

• A position is defined by the location of a person in a system of social relationships.

¹¹See Gross et al. (1958, p. 11–48) or Dahrendorf (1960, p. 40–45) for broad discussion about definitions of the terms role, position and status from other authors.

¹²See Linton (1936, p. 115–131) and other authors like Gross et al. (1958, p. 48–58), Dahrendorf (1960, p. 35–40) or Davis (1966).

¹³Writing "role" instead of "role" seems to be a subtle indication that the word itself originates ˆ from French, see e.g. Dahrendorf (1960, p. 13–17) or Thomas and Biddle (1966a, p. 5–7) for further discussion.

¹⁴Gross et al. (1858:58–60) worked out two dimensions of a single expectation, namely direction and intensity with the first reduced either a statement for or against something and the second a continuum from completely permissive (may or may not do) through the preferential (preferably should do) to the mandatory (absolutely must do). Further, Dahrendorf (1960, p. 18–34) added that with deviance from such expectations, positive or negative sanctions must follow in order to secure their compliance.

¹⁵Referring to Biddle (1986), this set of definitions is in line with the structuralist perspective of role theory.

A role of a person is a set of expectations of others about its behaviour, connected with a particular position.
A role-set is the sum of different roles associated with a particular position.
A position-set is the sum of different positions associated with a person.
A status implies the ranking of a person's position among a set of persons or locations.

Additionally, Blau and Scott (1963, p. 195) and Evan (1965) regarded these role theoretic considerations to be valid on an organisational level, too. As organizations consist of a set of persons with common position called membership, one can speak of an organization-set, referring to the various other organizations an particular organization is related. These relationships are mediated by "(a) the role-sets of its boundary personnel, (b) the flow of information, (c) the flow of products or services, and (d) the flow of personnel" (Evan (1965, p. B220)). As in every organization at least some positions are connected with a liaison role vis-`a-vis other organizations, the occupants of these roles can be regarded as representatives for their organization. This is especially true in the case of the freight forwarding business, a people-driven service industry where most personnel is in direct contact with their companies' clients in the name and on behalf of their company. Therefore throughout this work, the term "role-set" subsumes both role- as well as organization-set as a freight forwarder can be regarded as an individual and a legal entity at the same time without getting different outcomes in this context.

1.3 Research Methodology and Outline

Being a rather exploratory study emphasizing the quest for theories matching to realworld observations, an appropriate research approach for this piece of work may be best described as abductive reasoning (Kovacs and Spens (2005) or Spens and ´ Kovacs (2006)). Departing from mainstream research processes being either strict ´ deductive or inductive, the abductive research approach implies a creative interactive process (Taylor et al. (2002)) of matching or systematic combining theories to realworld observations made (Dubois and Gadde (2002)) in order to find a matching framework or to extend existing theory (Andreewsky and Bourcier (2000)) for further treatment. Following this research approach, the book is structured in seven Chapters as follows (see Fig. 1.4 for reference):

Starting with Chapters 2, first legal status and functional views of freight forwarder operations are outlined with a special focus set on Germany and Austria. After that, a closer look at multimodal transport chains including a leg by sea or air is made including technological development, functional aspects, institutional environment and market structure. Then inter-organizational interaction and cargo movement in such multimodal transport chains are described thoroughly by identifying actors and showing how they work together. In this context, a freight forwarder can be regarded as some special intermediary, organizing goods movements and 10 1 Introduction

Fig. 1.4 Structure of this Book

providing other related services along a chain of transport and logistics operations. These real-world observations from actual freight forwarding industry practice lead to a legal and functional perspective of freight forwarding which serve as a status quo or point of reference from which more elaborated scientific treatments are conducted.

In Chapter 3, a conceptual framework of a freight forwarder's intermediary perspective based on New Institutional Economics and Intermediation Theory of the Firm will be developed. First, the three basic concepts of New Institutional Economics, namely (1) Property Rights Theory (discussing forms of contractual relationships and contract enforcement mechanisms), (2) Transaction Cost Theory (addressing bounded rationality, opportunism, transaction costs and their impact on governance structure) and (3) Principal–Agent Theory (focusing on information asymmetry and their impact on agency costs and contractual design) are reviewed as a starting point for further discussion. Then recent developments in the Intermediation Theory of the Firm are addressed where a special focus on several distinct intermediary roles like being (1) a matchmaker (broker) in dyadic relationships, (2) a market maker (specialist, dealer, merchant or marketer) in triadic relationships, (3) an information producer, guarantor and/or certificator and (4) an agent employed for delegated expertise, bargaining, contracting and/or monitoring tasks is offered. After that, a taxonomy of middlemen and service providers in multimodal transport chains is presented, including players for commercial, financial, transport and logistics operations as well as some forms of infomediaries and information service providers. In this framework, the freight forwarders' role-set may be described best as a somehow strange hybrid providing intermediation as well as financial, transport and logistics support to shippers, consignees or consignors. Furthermore, some critical comments on the intermediary perspective are stated that demand a network perspective to catch the inherent network structure of freight forwarding in multimodal transport chains.

Therefore in Chapter 4, these multimodal transport chains in question are considered as specific forms of network structure. There are six distinct models of network structure, which are regarded fruitful in the context of this work: (1) Network Models in Operations Research and Management Science mainly coping with problems of network design and flow, (2) Strategic Networks focusing on cooperative relationships and strategic alliances around a central actor, (3) Dynamic Networks as business communities or temporal virtual companies for specific tasks, (4) Entrepreneurial Networks consisting of specialized intermediaries in a hierarchical information structure, (5) Industrial Networks with power-dependant exchange relationships in a framework of actors, activities and resources and (6) Social Networks with personal interaction, weak and strong ties, structural holes, social capital and embeddedness issues. Comparing these models of network structure and their underlying body of theory with initially made observations as documented in Chapter 2, it can be concluded, that a freight forwarder's intermediary role in multimodal transport chains can at best be described as a focal company in a social network structure.

As social network analysis is a wide prosperous field of research, its roots and levels of analysis are briefly outlined in Chapter 5 before relevant analytical concepts for both personal network structure (e.g. network size and density) and intermediary position (e.g. centrality and prestige) as well as relationships (e.g. brokerage capacity and network constraint) are thoroughly discussed. In addition to this, suggestions are made how to apply these measures to empirical data in order to explore social network structures as well as social capital.

12 1 Introduction

In Chapter 6, a stylized model framework of freight forwarder's personal network in multimodal transport chains is then set up from an ego-centric social network point of view in order to show empirical evidence for the freight forwarder's intermediary role in multimodal transport chains in the case of a small sample of freight forwarder's personal networks in Germany (especially from Hamburg and Bremen-Bremerhaven) as well as Austria.

Finally, in Chapter 7, both theoretical and empirical results as well as contribution to extant research are summarized and some managerial implications are given along with recommendations for further research.

Chapter 2 Legal and Functional Perspective of Freight Forwarding in Multimodal Transport Chains

Following authors like Palmhofer (1944, p. 8–9), Ullmann (1950, p. 18), Murr (1979, p. 1), Matthai (1979, p. 4), Seiler (1981, p. 33) or Gass (1991, p. 15), freight ¨ forwarding as a special business can be traced to the Medieval Age. Up to this time, a merchant principally transported cargo overland by own means or vessels chartered directly from their owners. Of course, since the times of the Roman Empire, usually merchandise cargo in transit was always accompanied by some agent who was obliged to take care for the shipment and to act on behalf of the owning merchant, but such transport operations can be considered merely as being done on own account.1

Several trade practices coming up in the late Medieval Age can be regarded as the beginning of the freight forwarding business. First, stand-alone foreign subsidiaries of trading companies called "Accumenda," "Implicita" or "Faktoren" started business on own account, combining commercial trade with freight forwarding.2 In the sixteenth century, several cities in Central Europe raised regulations, upon each cargo of an extraneous merchant subject to a "Stapelrecht" crossing their territory had to be unloaded there and offered for sale to domestic merchants.³ As they were not always able to buy all the merchandise offered to them, they took it on commission in order to sell them later or send them to somewhere else under their name.4 Both tasks are nowadays part of the freight forwarding business: acting as a

¹See Stamm (1908, p. 1–2), Goerres (1909, pp. 7–8), Securius and Boning (1941, pp. 10–11), ¨ Palmhofer (1944, p. 80–86), Ullmann (1950, p. 13–18), Hald (1951, p. 11), Matthai (1979, p. 4) or ¨ Gass (1991, pp. 14–15) with further reference.

²See Goerres (1909, pp. 7–8), Vadnai (1921, pp. 29–30), Securius and Boning (1941, pp. 10–11), ¨ Ullmann (1950, p. 18), Hald (1951, pp. 11–12) or Gass (1991, pp. 15–16) with further reference.

³Cf. Goerres (1909, pp. 9–11), Vadnai (1921, p. 30a-31a), Securius and Boning (1941, p. 10), ¨ Palmhofer (1944, p. 8–9), Ullmann (1950, pp. 19–22), Hald (1951, p. 12), Matthai (1979, pp. 4–6), ¨ Seiler (1981, p. 33) or Jaeger and Laudel (2003, p. 70).

⁴See e.g. Wolters (1973, p. 31) or Seiler (1981, pp. 33–34).

commission agent or being a despatcher of cargo shipments on behalf and account of a shipper.

In addition to this, another regulation closely connected to the "Stapelrecht" was the "Umschlagsrecht" standing for the allowance of cargo handling operations for domestic carriers.⁵ Specialized agents called "G¨uterbestatter" or "G¨ ¨ uterfertiger" were organizing and supervising these handling operations as well as checking the goods being unloaded.6 They can be regarded as the real ancestors of foreign freight forwarders, because step by step, they expanded their operations to expedition of cargo and customs brokerage.⁷ With the increasing importance of railway operations becoming the dominant inland mode of transport during the nineteenth century, freight forwarders added new fields of activity like cargo consolidation for shippers and cartage service on behalf of railway companies.8 All these developments resulted in the forwarding business as we know it today.

2.1 The Freight Forwarder from a Legal and Functional Point of View

Freight forwarders are always described along their legal status and/or their functions, so in the following first the legal framework of forwarding business in Germany, as well as in Austria are briefly described and differences to the forwarding business in other countries will be highlighted. Then freight forwarder functions as well as general fields of activity and specialization in forwarding business are outlined as a second, functional point of view.

2.1.1 Actual Legal Status of Forwarding Business in Germany and Austria

Forwarder law was codified for Germany as a nation-state with the German Commercial Code (HGB) of 1897 (DRGBl 219) and for Austria with the Austrian

⁵Cf. Vadnai (1921, p. 30a–31), Securius and Boning (1941, pp. 10–11), Palmhofer (1944, p. 9), ¨ Ullmann (1950, p. 19–23), Hald (1951, pp. 12–13) or Matthai (1979, p. 6). According to Goerres ¨ (1909, p. 10) or Vadnai (1921, p. 31), the "Stapelrecht" can be regarded as a privilege of commercial trade and the "Umschlagsrecht" as a privilege of cargo movement.

⁶Cf. Goerres (1909, pp. 11–13), Vadnai (1921, p. 31a–32), Palmhofer (1944, pp. 9–10), Ullmann (1950, p. 23–24), Matthai (1979, pp. 6–7), Seiler (1981, pp. 34–35) or Jaeger and Laudel (2003, ¨ p. 70).

⁷Cf. Vadnai (1921, p. 31a–33a), Palmhofer (1944, pp. 89–92), D'Amato and D'Amato (1977), Seiler (1981, pp. 34–35) or Gass (1991, pp. 16–17).

⁸Cf. Goerres (1909, pp. 21–49), Vadnai (1921, p. 34), Ullmann (1950, pp. 28–49) and Wolters (1973, p. 27–28).

Commercial Code (UGB) already established in 1862. As one of many causes of the occupation of Austria by the German Reich in 1938, the content of the original UGB of 1862 was fully synchronized with the HGB with DRGBI 1938 I 1428 (GBIÖ 1938/520) and DRGBI 1938 I 1999 (GBIÖ 1939/86). After the end of the German Reich in 1945, the original UGB of 1862 was never set into force again so that today, the actual UGB parallels the actual HGB in many ways. 10

Now, in the fourth book of the HGB as well as the fourth book of the UGB, the legal statuses of:

Forwarding (§§ 453–466 HGB and §§ 407–415 UGB)
Transportation (§§ 407–452d HGB and §§ 425–452 UGB)
Warehousing (§§ 467–475h HGB and §§ 416–424 UGB)
Commissioning (§§ 383–406 HGB and §§ 383–406 UGB)

as special forms of commercial business are explicitly addressed. $^{11}$ In Austria, the term "Spediteur" used for freight forwarder is explicitly defined by law in $\S$ 407 $IIGB^{\cdot 12}$

"§ 407

(1) A freight forwarder is someone who undertakes by profession arrangements for dispatching goods shipments through [inland] carriers or carriers of seagoing vessels on account of a third person (the consignor) in his name.
(2) If this section contains no provisions for the rights and duties of a freight forwarder, provisions applicable to a commission agent, especially the provisions in §§ 388 to 390 [UGB] about reception storekeeping, and insurance of goods have to be applied."

Or, in other words, a freight forwarder is in first instance (1) a professional, who (2) dispatches goods for transport in his name and (3) signs contracts of

^9

Stamm (1908, pp. 3–4), Goerres (1909, pp. 17–18), Palmhofer (1944, pp. 93–94), Ullmann (1950, pp. 31–32) or Gass (1999, pp. 17–20) trace German forwarding law back to the first general German commercial code called "Allgemeines Deutsches Handelsgesetzbuch" (ADHGB) of 1861 and remarked that the provisions there about freight forwarding were not altered when the HGB was set into force in 1897. But lacking political unity until the establishment of the German Reich in 1871, the ADHGB was adopted only by a fraction of all states in the German Union of that time so that it cannot be considered to be a proper commercial the concerning the whole German territory. For ease of exposition, in the following, the abbreviations HGB ("Deutsches Handelsgesetzbuch") standing for German Commercial Code and UGB ("Österreichisches Handelsgesetzbuch"), recently renamed in "Unternehmensgesetzbuch" with "Handelsrecht-Änderungsgesetz" (BGBI. 1 No. 120/2005) standing for Austrian Commercial Code are used.

^10

Cf. Kirchner (1950, p. 5), Hausmaninger (2000, p. 233) or Straube (2003, pp. 1280–1281).

^11

For a more detailed discussion of the actual UGB see Straube (2003, pp. 1280–1574) or Krejci (2005, p. 368–419) with further references. In addition to this, the fifth book of HGB as well as the fifth book of UGB, contains provisions of national sea trade law in §§ 474–905 UGB, §§ 476–905 HGB and supplement to § 664 HGB.

¹²Translation by the author.

affreightment with carriers by land or sea. 13 In addition to this, he may act as a commission agent on behalf of a consignor and/or does other ancillary tasks closely connected to the carriage of goods like temporary storage or insurance provision.

With the German Transport Law Reform Act of 1998 (DBGBl.1998 I 1588. revised with DBGB1.1999 I 42), the fourth book of the HGB as well as other sources of German transport law were subject to a fundamental revision.¹⁴ Doing this, the former legal definition of a freight forwarder identical to the actual § 407 UGB was abandoned in favour of a more elaborate listing of obligations and duties constituting a freight forwarding contract in §§ 453-454 HGB:15

"§ 453: Forwarding contract

(1) By virtue of the forwarding contract the forwarder is obliged to arrange for the dispatch
(2) The sender is obliged to pay the agreed remuneration.
(3) The provisions of this chapter only apply if dispatching goods is part of the operation of a commercial enterprise. [...]
- § 454: Arranging for the dispatch of the goods
(1) The duty to arrange for the dispatch of the goods includes organization of the carriage, in particular (a) determination of the means and route of transport, (b) choice of performing enterprises, conclusion of carriage, warehousing and forwarding contracts required for the dispatch, as well as providing information and giving instructions to the performing enterprises, and (c) securing the sender's claims for compensation.
(2) The duties of the forwarder also include the provision of other agreed services relating to carriage such as insuring and packaging the goods, labelling them and clearing them through customs. However, it is only if the agreement calls for it that the forwarder shall be obliged to arrange for the conclusion of contracts for these services.
(3) The forwarder concludes the required contracts in his own name or, if he is authorized to do so, in the name of the sender.
(4) In fulfilling his obligations the forwarder shall act in the interest of the sender and carry out his instructions."

Comparing this actual legal status of forwarding business in Germany with the § 407 UGB, the actual §§ 453–454 HGB places organization of a carriage of goods in the foreground followed by an explicit description, whereas obligatory main tasks as well as ancillary tasks have to be executed upon request. Doing this, the freight forwarder acts normally in his name on behalf and account of a consignor, like a commission agent.

In addition to this "core business" of freight forwarding, a freight forwarder is, according to § 458 HGB or § 412 UGB, allowed to act as carrier "im Selbsteintritt"

^13

Carriers by air are not explicitly mentioned here, because at the time of initial codification in the nineteenth century, no-one could think of transportation by air at all. But nowadays, § 407 UGB is regarded to be applicable for national air transport operations, too.

¹⁴A full description of the actual provision in the HGB after 1998 are beyond the scope of this work. See Herber (1998), De la Motte (1998a), Gass (1999), Schauer (2001), Müglich (2002), Jaeger and Laudel (2003), Wieske (2003, pp. 39-65) or Lorenz (2005, pp. 85-148) for an overview and some commentary notes on the actual §§ 407-475h HGB.

¹⁵Translation by Deutsche Gesellschaft für Transportrecht e.V., see http://www.transportrecht.org/ hgb_4_synopse.html.

that means performing carriage of goods by own means and on own account with the rights and duties of a carrier as codified in §§ 453–466 HGB and §§ 407–415 UGB, too. Furthermore, § 459 HGB and § 413(1) UGB allows forwarding at fixed costs. Doing this, the freight forwarder has, as far as the carriage is concerned, again the rights and duties of a carrier by land or sea. Last but not least, a freight forwarder is entitled by § 460 HGB and § 413(2) UGB to arrange a dispatch of goods from several consignors on the basis of a contract for a collective consignment concluded on his account called groupage service. ¹⁶ Again, as far as the carriage as a collective consignment is concerned, he has then the rights and duties of a carrier by land or sea. These provisions in §§ 458–460 HGB and §§ 412–413 UGB give freight forwarders permission to provide cargo consolidation and transport services on own account. Further important aspects in forwarder law are provisions for subsequent forwarding contracts which are codified in § 465 HGB as well as § 411(1) UGB. There, the consequences of a procurement of tasks out of a freight forwarder's forwarding contract with a consignor to another freight forwarder are regulated. ¹⁷

Another important legal source in forwarding business are General Terms and Conditions (GT&Cs) issued by freight forwarding associations, sometimes together with other federations representing shippers. ¹⁸ In first instance, they want to clarify rights and duties of freight forwarders and consignors as well as to give more elaborate provisions for everyday forwarding business, especially concerning ancillary services like customs brokerage or third-party logistics not explicitly addressed by law. Furthermore, they usually reduce the maximum liability raised by law and sometimes establish a special system of forwarding insurance. Normally, these GT&Cs are considered as trade customs, at least after some notification. ¹⁹

A well-known and widely accepted GT&C in the German forwarding business are the "Allgemeine Deutsche Spediteursbedingungen" (ADSp). They were first formulated in 1927 by a mixed consortia with forwarding as well as shippers' associations and underwent several revisions with the actual version called ADSp 2003 effective as of 01/01/2003. They are applicable to

"[...] all contracts for the transportation of goods, irrespective of whether they concern freight forwarding, carriage, warehousing or other services common to the forwarding trade;

^16

According to Nickenig (1996, p. 92), the first groupage service dates back to 1877, where German railway companies introduced the "Reformtarif" with a differential tariff structure divided in express, part loads and wagon loads which allowed freight forwarders to consolidate small shipments for wagon loads.

^17

See Krejci (2005, pp. 378–379) for more details.

^18

Cf. Mayer (1933, pp. 20–21), Ullmann (1950, pp. 64–66) or Gass (1991, p. 37).

^19

For further discussion, see e.g. Krejci (2005, pp. 369–370) for Austria and De la Motte (1998b) or Lorenz (2005, pp. 149–151) for Germany.

^20

See Valder (2002), Jaeger and Laudel (2003, pp. 73–95) or Lorenz (2005, pp. 149–175) for the ADSp 2003. Already in 1908, the Verein Deutscher Spediteure (VDS) passed the "Allgemeine Beförderungsbedingungen" (ABB), but these GT&C was in the following merely regarded as a non-obligatory recommendation by their member companies, see Esch (1923).

these also include logistics services commonly provided by freight forwarders in connection with the carriage or storage of goods."21

Furthermore, the ADSp 2003 takes precedence in case of deviating trade customs or legal provisions and a freight forwarder is only responsible for arranging the necessary contracts required for the performance of these services, unless there are no other legal provisions mandatory.²² Exceptions to this are "contracts that deal exclusively with packaging, the carriage of removal goods and their storage. crane lifting, assembly jobs or heavy lift and high volume transports, except for normal transhipment services of the freight forwarder, the carriage and storage of goods to be towed or salvaged"²³ and contracts with private consumers²⁴, because provisions for this kind of services are given either by other GT&Cs or by HGB. The ADSp 2003 adds special provisions for e.g. transmission of instructions, customs clearance, packaging, marking, handling, warehousing or payment issues to the regulations of forwarding business in the HGB. Moreover, it lowers the liability of a freight forwarder in concordance of § 431 HGB and gives some new provisions about insurance issues discussed later on in the case of Austria.

In addition to the ADSp 2003, the "Vertragsbedingungen für den Güterkraftverkehrs- und Logistikunternehmer" (VBGL) are issued by the "Bundesverband Güterkraftverkehr, Logistik und Entsorgung" (BGL) with the latest version called VBGL 2003, effective as of 27/01/2003.²⁵ The VBGL 2003 contains virtually all provisions of the ADSp 2003 and adds additional provisions for the carriage of goods as well as some new ones for transport of disposal, carriage on basis of a charter party and other more or less uncommon logistics services for goods not covered by HGB.²⁶ To some extent, the VGBL 2003 can be considered to be a good alternative to the ADSp 2003, especially if a freight forwarder is more engaged in carriage of goods on own account and/or provides third-party logistics services.

In Austria, the "Allgemeinen Geschäftsbedingungen im Österreichischen Speditionsgewerbe" (AGBSp) were first issued in 1930²⁷, succeeded by the "Allgemeinen Österreichischen Spediteursbedingungen" (AÖSp) in 1947, with the actual version effective as of 01/01/2002.²⁸ Generally, the AÖSp take precedence over local and

^21

See Art. 2.1. ADSp 2003, translation by Verein Hamburger Spediteure e.V., see http://www. vhsp.de.

^22

Art. 2.2. and 2.5. ADSp 2003.

^23

Art. 2.3. ADSp 2003.

^24

Art. 2.4. ADSp 2003.

^25

See Schindler (1998, 2003) or Jaeger and Laudel (2003, pp. 108-114) for a brief overview and some comments.

^26

Strictly speaking, the VBGL 2003 deviates from the provisions in the ADSp 2003 in two ways: There is a uniform liability for all tasks done by a freight forwarder and the transport insurance is only covered after a written order.

^27

Cf. Mayer (1933, pp. 20-21) or Stern (1938, p. 54).

^28

In the mean time, the AGBSp were abandoned in favour of the ADSp in 1939, see Ullmann (1950, p. 119-123). From time to time, the AÖSp were subject to some minor revisions, the last

municipal trade customs and they are applicable to all activities performed by a freight forwarder in forwarding, carriage of goods, warehousing, commission or any other business related to.²⁹ Conversely, they are not applicable if a freight forwarder is acting solely as an agent of a carrier in accordance with special conditions, works as a carrier on basis of a surface transport contract of the Austrian Federal Railways or performs removal operations (including warehousing) on a national scale or from/to Austria, provided that the forwarding activities are done under § 407 UGB. 30 Furthermore, the AÖSp contains additional provisions to the legal framework of the non-compulsory UGB about orders, communication, inspection, packaging, warehousing, customs clearance and payment issues. But the most important topic is, that the AÖSp limits liability of a freight forwarder otherwise being unlimited under UGB and introduces an one-for-all risk insurance policy which covers all damage to goods and loss of property occurring under a forwarding contract or contract of carriage concluded with reference to the AÖSp. This kind of a combined forwarding and transport insurance is unique in forwarding business all over the world. In Germany before 2003, likely a similar one-for-all risk insurance policy system existed for a very long time. But with the German Transport Law Reform Act of 1998 giving the provision of the fourth book of HGB a status of compulsory law, this system of a combined forwarding and transport insurance was questioned more and more.³¹ At last, it was abandoned after about 75 years in favour of an obligatory liability insurance for the freight forwarder and additional coverage through a cargo insurance upon request.

2.1.2 Comparisons with Legal Status of Forwarding Business in Other Countries

Looking at the legal status of forwarding business in other countries like Switzerland, France, Denmark, Norway, Sweden, Finland, United Kingdom and the United States considerable deviations from the provisions found in the HGB and UGB can be remarked.³²

In Switzerland, the commercial code is called the "Obligationenrecht" (OG) established in 1911. In Art. 439 OG, the freight forwarder is legally defined as

revision in 2002 was because of Austria joining the European Monetary Union. See Spera (2011, p. 76–81) for a brief overview.

²⁹ See § 2a) and 2c) AÖSp.

³⁰ See § 2b) AÖSp.

^31

Cf. Valder (2002), Wieske (2002), Häusser and Abele (2003), Heuer (2003) or Lorenz (2005, p. 176–192).

^32

For more international comparisons of forwarding law and/or definitions of freight forwarders in different countries see Matthäi (1979, pp. 28–38), Thaler (1990, pp. 92–96), Gass (1991, pp. 27–30, 67–72) or Jaeger and Laudel (2003, p. 71).

a special kind of commission agent and so subject to Art. 424–438 OG.³³ But concerning the carriage of goods, provisions for carriers in Art. 440–447 OG with a full liability system are applied to him. Therefore, local swiss freight forwarder associations joined together and issued the "Allgemeinen Bedingungen des Schweizerischen Spediteurs-Verbandes" (AB-SSV) in 1922 (actual version as of 01/07/2005 under the title of AB SPEDLOGSWISS) in order to reduce their liability.³⁴

Under the French "Code de Commerce" (CdC) of 1808, a forwarder can either act as a commission agent of the consignor for ancillary services to the carriage of goods called "transitaire," or like a contracting carrier called 'commissionaire de transport' as legally defined in Art. L 132–1 to 132–3 CdC.³⁵ Furthermore, the "Fed´ eration ´ des Entreprises de Transport et Logistique de France" (TLF) issued a model GT&C for both forms of operations called "Conditions gen´ erales de ventes TLF" with the ´ latest version as of 01/10/2001 which also limits liability of all kind of forwarders to actual fault.³⁶ According to Gass (1991, pp. 48–49), a French forwarding company has to decide in advance which field of activity it wants to serve due to a restrictive licensing system.

In the four Scandinavian countries Denmark, Norway, Sweden and Finland, up to now there does not exist any codified law concerning forwarding business. But already in 1919, a first version of the General Conditions of the Nordic Association of Freight Forwarders ("Nordisk Speditørforbunds Alminnelige Bestemmelser" (NSAB)) was issued by the "Nordiskt Speditor F ¨ orbund" (NSF), the latest version ¨ is NASB 2000 effective as of 01/06/1998.³⁷ In this GT&C, a distinction is drawn between a freight forwarder acting on own account called "Speditor" or as a ¨ commission agent on behalf of a consignor called "Ekspeditor." Following ¨ - 1 NASB 2000, a freight forwarder has a liability like a carrier if he performs the carriage of goods by own means or he has expressly or impliedly accepted liability of a carrier like issuing a transport document in his name or quoting own prices for transport.

³³See Vadnai (1921, p. 6), Kirchner (1950, p. 6), Tschudi (1975, p. 41) or Seiler (1981, pp. 21–22). Further, Guhl et al. (2000, pp. 588–603) or Foglar (2009) for a brief overview of legal provisions in the actual OR regarding commission agents, carriers and freight forwarders.

³⁴Cf. Tschudi (1975, pp. 34–36), Seiler (1981, pp. 27–30), Foglar (2009) or http://www. spedlogswiss.ch.

³⁵Cf. Ullmann (1950, pp. 30–31), Matthai (1979, p. 36), Chevalier (1989), Thaler (1990, p. 94), ¨ Tetley (1992), M¨uller-Feldhammer (1996, pp. 105–116), Ramberg (1998) or Victor-Granzer (2005).

³⁶Notably, as a commission agent, the forwarder is only liable for actual fault, but as a contracting carrier, his liability is extended to actual and third party fault in case of loss, damage or late delivery of cargo, see L 132–4 to 132–6 CdC.

³⁷See Gass (1991, p. 70), Wegdell (1996, pp. 171–239) or http://nordicfreight.org/engsida/model. htm.

In the United Kingdom as well as in other case law countries, there exists no codified law for forwarding business, either.³⁸ Freight forwarders in the German and Austrian sense usually act under the Law of Agency mainly as forwarding agents on behalf of their consignors and/or carriers by land, sea or air and therefore they sign contracts of affreightment in the name of their principal, indicating this at least with the formulation "on behalf of our principals."39 Running their business in this manner, they are called "forwarding agents" or "freight forwarders."40 In addition to this, a freight forwarder in the United Kingdom can be regarded as a "forwarder as principal," if he actually performs the carriage of goods by own means.⁴¹ Regarding GT&Cs, the British International Freight Association (BIFA) issued the "Standard Trading Conditions" (STCs) applicable to all kinds of forwarding business "as rules of the game."⁴²

In the United States, foreign freight forwarders emerged first in the early nineteenth century mainly acting at sea ports as agents on behalf of shipowners and later on expanding their business to cargo consolidation for inland shippers.43 After decades of a laissez-faire policy, first legal regulations about foreign freight forwarding business were included in the US-American maritime commercial law in 1942.⁴⁴ After 1950, all freight forwarding companies had to be registered and from 1961 to 1996 they were even subject to a licensing system, restricting them mainly to a despatching and/or cargo consolidation function in water-born trade.45 Therefore, affreightment of carriers including despatch of goods for carriage on a national scale are normally done by "transportation brokers," "shippers" agents'

³⁸Cf. Gass (1991, pp. 62–65), Benson et al. (1994, pp. 305–307), Triebel et al. (1995, pp. 153–159) or Ramberg (1998).

³⁹Cf. Gass (1991, p. 23) or Triebel et al. (1995, pp. 157–159).

⁴⁰Cf. Matthai (1979, pp. 37–38), Thaler (1990, pp. 93–94) or Triebel et al. (1995, p. 157). ¨

⁴¹Matthai (1979, p. 37) proposed to call a freight forwarder then "private carrier." According to ¨ Benson et al. (1994, pp. 305–306), a "private carrier" is in opposition to a "common carrier" with the first acting only on basis of private "Conditions of Carriage" between them and their customers and the second providing carriage of cargo on a regular basis with full liability only limited to a few exceptions like Act of God, Act of the Queen's enemies, inherent vice, fault or fraud of the consignor.

⁴²BIFA (2000, p. 1). By the way, terms like "freight forwarder" or "forwarding agent" are never mentioned in this GT&C. Instead of this, the term "company" as a BIFA member trading under the STC is used.

⁴³Cf. Barton and McGehee (1942), D'Amato and D'Amato (1977), Murr (1979, pp. 3–10), Rahman (1989, pp. 15–18), Dempsey (2000) or Kehagiaras (2004).

⁴⁴C.f. Murr (1979, pp. 11–13, 23–55).

⁴⁵See Barton and McGehee (1942), D'Amato and D'Amato (1977), Murr (1979, pp. 23–55), Brown (1984b), Crum (1985), Morash (1986, 1987), Hill (1988), Sherwood and Burns (1992), Johnson and Schneider (1995a), Wood et al. (1995, p. 116), Wood and Johnson (1996, pp. 264– 267), Lambert and Stock (1999, pp. 180–185), Taylor and Jackson (2000) or Dempsey (2000). Further see the actual provisions in the "Code of Federal Regulations" (CoFR), especially 49 CoFR, Chap. 139, Sect. 13901–12908 about registration of motor carriers, water carriers, brokers and freight forwarders.

or "shipper associations." In the case of cross-border cargo movements, "ocean transportation intermediaries" (OTIs) or other forms of "non vessel operation common carriers" (NVOCCs) are usually employed for export shipments whereas "customs-house brokers" work on the import side.⁴⁶ But as subsequent surveys performed by Murphy et al. (1992) or Murphy and Daley (1995, 2000) showed, more and more international freight forwarders in the United States provide all these specialized services in a one-stop shopping manner.47 The National Customs Brokers and Forwarders Association of the United States (NCBFAA) issued GT&Cs called "Terms and Conditions of Service," which seem to be a widely acknowledged industry standard contract practice employable by regular members of NCBFAA vis-`a-vis their clients after some notification.⁴⁸

Last, but not least, the International Federation of Forwarding Agents Associations ('Fed´ eration Internationale des Associations de Transitaire et Assimil ´ es' ´ (FIATA)) adopted "FIATA Model Rules for Freight Forwarding Services" in 1996.⁴⁹ Again, basic provisions for everyday business as well as a limitation of forwarder's liability are given and two different concepts of forwarding business are defined. First, a freight forwarder can act as a "forwarder as agent" on behalf of his principal like the "Spediteur" in Germany or Austria, the "transitaire" in France, the "Ekspeditor" in the Scandinavian Countries and the "forwarding agent" or ¨ "freight forwarder" in the United Kingdom.50 Second, he can act as a "forwarder as principal," procuring or even performing carriage of goods in his name and on own account with his rights, duties and liability being the same as a normal carrier by land, sea or air. If he performs the carriage of goods by own means, he is usually called a "performing carrier," otherwise he is regarded as a "contracting carrier," still assuming carrier liability despite another carrier is actually shipping the goods. Being a performing carrier, he is called "Spediteur im Selbsteintritt" in Germany and Austria, "Speditor" in the Scandinavian countries and "forwarder as principal" ¨ in the United Kingdom. Being a model rule, they can be regarded as a blue print for

⁴⁶Cf. Thomchick et al. (2004, p. 44) or Passas and Jones (2007). Both are still subject to a strict licensing system according to 46 CoFR, Chap. 36, Sect. 1718 (ocean transportation intermediaries) and 19 CoFR, Chap. 4, Sect. 1641 (customs-house brokers). See Hill (1988), Sherwood and Burns (1992), Wood et al. (1995, pp. 115–116), Lambert and Stock (1999, pp. 205–207), Clott (2000), Thorby (2001a) or Kehagiaras (2004) for more information about US-based OTIs and NVOCCs.

⁴⁷According to their survey from 2000, 81.0% of their responding international freight forwarders act as customs-house brokers and 57.0% as NVOCCs with half of the respondents doing both. Further, 64.7% of their revenues are from sea freight and 34.4% from air freight operations.

⁴⁸See e.g. http://www.cargolaw.com/guides usingterms.html.

⁴⁹See Ramberg (1998, 2000) with further comments.

⁵⁰This holds likely in the same manner for customs brokerage services, where a freight forwarder acts as an agent, when he clears goods at customs house on behalf of his principal on his account. Doing this special service, he is named e.g. "Zollspediteur" or "Grenzspediteur" in Germany and Austria, "transitaire agree en douance" or "commissionaire agr ´ ee en douance" in France, ´ "clearance forwarder" in the United Kingdom and "customs-house broker" in the United States, see Gass (1991, p. 59a-67).

Fig. 2.1 Contract forms of affreightment

GT&Cs for freight forwarding associations as well as individual forwarders all over the world.

In order to clarify these different concepts of forwarding business, it is worth to have a closer look at different contract forms of affreightment as shown in Fig. 2.1. The first and foremost simple form of affreightment is a contract of carriage directly concluded between a shipper and his carrier, including the case of a freight forwarder acting as a "performing carrier." The other four forms can be subsumed under the label "indirect conclusions of affreightment," because there is always some third-party involved. In case of a carrier holding a contract of affreightment with a shipper but acting as a "contracting carrier," he procures the actual carriage of goods to another carrier, called the "performing carrier." Doing this, he is usually called "freighter," "non-vessel operator" (NVO) or "non-vessel owning carrier" (NVOC) bearing full liability concerning the carriage of goods. ⁵¹ If a freight forwarder is enacted by a shipper to dispatch goods for carriage on basis

^51

See Klippel (2000) or Branch (2000, pp. 94–95). In the United States, NVOs are usually called NVOCCs or OTIs which are terms originating from the US-American maritime commercial law, see Sect. 2.1.2.

of a forwarding contract, he concludes affreightment with the carrier for carriage of goods on behalf of his principal as a beneficiary person.⁵² Next form of indirect conclusion occurs if the third-party acts as an agent for the carrier based on an agency contract.⁵³ Then, he is usually called "shipping agent" giving quotation to the consignor and concluding contracts of affreightment in the name of his principal on a regular basis, with the carrier paying him for his work.⁵⁴ Last but not least, a "transport broker" is an independent third-party matching shippers and carriers in his name and on own account.⁵⁵ He gets a payment called "commission" or "courtage," if a contract of affreightment is concluded.

Looking on the legal status of forwarding business and comparing it with these contract forms of affreightment, it is quite obvious, that the last two contract forms of affreightment described above are usually not concluded by a freight forwarder. As it will be shown later on in Sect. 2.2.4, these shipping agents and transport brokers are specialized intermediaries often employed in sea and/or air transport cargo operations standing between the freight forwarder or his consignor and a carrier.

2.1.3 Freight Forwarders from a Functional Point of View

A freight forwarder is always considered as "The Architect of Transport." In order to do a right job, he seems to have some specialized knowledge or means to organize

^52

Cf. Mittendorf and Oelfke (1974, pp. 24-25).

^53

Cf. Mittendorf and Oelfke (1974, pp. 232–234), Trappe (1995), Puttfarken (1997, pp. 248–250) or Herber (1999, pp. 168-171).

^54

In Germany and Austria, a shipping agent is legally defined as an agent according to §§ 84–92 HGB or the Austrian "Handelsvertretergesetz" (HVertG) of 1993. In common law countries, agents are acting on basis of the Law of Agency, but with (86/653/EEC), the United Kingdom and Ireland have to enact now a uniform law of agency on the national level, see Triebel et al. (1995, pp. 166-178). Conversely, the third party might also be an agent acting on behalf of a shipper or consignor based on an agency contract. But this seems actually to be a very rare practice, only Spera (2011, pp. 81-82) referred to such a "Transportagent" in the case of Austria and Brown (1984b), Morash (1986) or Johnson and Schneider (1995b) described "shippers agents" or "shipping associations" as intermediaries working for shippers in railway piggybacking as well as other transport operations in the United States on a national scale.

^55

Cf. Svendsen (1958, pp. 298–306), Mittendorf and Oelfke (1974, pp. 228–232), Puttfarken (1997, p. 248), Branch (1998, p. 212-217) or Herber (1999, pp. 169-171). Other terms are "loading broker," "discharging broker" or "air freight broker." Legal provisions for this business of brokerage are §§ 93-104 HGB for Germany and the "Maklergesetz" (MaklerG) of 1996 in the case of Austria. According to Triebel et al. (1995, pp. 179-180) or Trappe (1995, p. 321), it is worth to remark, that in common law countries, a "broker" is not the same as a "Makler" defined by German or Austrian law. But in the context of this work, such a distinction is not drawn so that in the following, the term "broker" is used to describe a "Makler" in the German or Austrian sense. ⁵⁶Cf. FIATA (1975, p. 4), Schumacher (1987a, p. 131), Thaler (1990, pp. 83–84), Danzas Lotse (2002, p. 146), Lorenz (2005, p. 57) or Dischinger et al. (2005, p. 23).

carriage of goods by land, waterways (including sea transport and inland navigation) and air supporting national as well as cross-border cargo movements. In addition to this, he provides merely ancillary services like warehousing, handling, packaging and transportation of goods. According to a brochure of the FIATA in 1975, a freight forwarder principally:57

1. Gives advice to customers on the quickest and most economical means of transportation (consultancy function).
2. Gives advice to customers on packing problems (packaging function).
3. Cares for customs clearance (clearance function).
4. Secures compliance with foreign trade regulations and Letter of Credit instructions (documentary function).
5. Makes choice of the most suitable carrier and the conclusion of the contract of carriage (affreightment function).
6. Provides groupage service as a particular contribution to economy (consolidation function).
7. Gives insurance coverage during transportation (insurance function).
8. Gives advice to customers on warehousing and distribution (logistics function).
9. Provides carriers' as well as forwarders' documents (fiduciary function).
10. Provides supervision of the movements of goods (supervision function).

Reviewing authors defining freight forwarder functions like Metzger (1922, pp. 89–99), Schlichting (1931, pp. 11–58), Mayer (1933, pp. 12–14), Stern (1938, pp. 28–35), Ohling (1950, pp. 216–218), Rosenthal (1950, pp. 116–121), Ullmann (1950, pp. 71–107), Kirchner (1950, pp. 11–45), Smith (1974, p. 178–179), Mittendorf and Oelfke (1974, pp. 25–29), D'Amato and D'Amato (1977), Murr (1979, pp. 20–21, 293–297), Seiler (1981, pp. 40–49), Matthai (1985, pp. 24–26), ¨ Dehn (1987, pp. 22–23), Schumacher (1987a, p. 131–139), Thaler (1990, pp. 87– 88), Sherwood and Burns (1992), Murphy and Daley (1995), Johnson and Wood (1996, p. 410–412), Branch (2000, pp. 417–418), Lucke et al. (2001, pp. 265–267), Pfohl (2003, p. 285–286), Lorenz (2005, pp. 61–63, 492–494) and Dischinger et al. (2005, pp. 23–24) two typical functions can be added, where a freight forwarder

11. Credits freights, fees, and duties payable at once and bills it later on or collects and submits money on behalf of the consignor (quasi-banking function) or
12. Acts as a carrier on own account (transport function).

In the following, these twelve functions of a freight forwarder as found in the literature are briefly commentated one after another before 25 freight forwarder definitions from the above mentioned authors are compared:

Having usually a direct contact to shippers, a freight forwarder serves first a consultancy function (1) in quoting freight rates, making routing recommendations and advising in export and/or import regulations as well as helping in choosing appropriate terms of sales and/or payment.

⁵⁷Cf. FIATA (1975), the words in brackets are added by the author.

The packaging function (2) of a freight forwarder mainly consists of securing proper export packing and marking of shipment as well as organizing, stuffing and stripping of containers. Sometimes, neutralising, relabeling and/or repacking of shipments are needed, if e.g. a consignor wants to disguise their origin. All these packaging functions are either organized by the freight forwarder or even conducted by him on own account.

In case of cross-border shipments, the clearance function (3) of a freight forwarder acting as an agent for a consignor or consignee is very important to speed up cargo movements at frontiers. Despite the fact that in the member states of the European Union today almost everyone is allowed to deal with customs authorities, a freight forwarder is usually regarded to have more expertise than normal shippers. Therefore, he is often asked for dealing with customs like submitting export declarations at the country of departure, opening and closing transit procedures and/or performing import procedures in connection with customs clearance at the country of destination on behalf of a consignor or consignee.58 In other countries like the United States, the clearance function is mainly served by special intermediaries called "customs broker" or "customs-house broker" due to strict licensing requirements.⁵⁹

The documentary function (4) comes along with preparing and processing documentary paperwork for a shipment in concordance with customs and consular requirements or concerning terms of sales and payment as described in an underlying contract closed between consignor and consignee. Prominent examples of this kind of paperwork are export licenses or embargo permits needed for export procedures at the country of departure and import licenses, delivery permits, commercial invoices, consular invoices, inspections certificates, certificates or declarations of origin needed for import procedures at the country of destination.60 Furthermore, a freight forwarder can issue own documents like movement certificates or a Forwarding Agent Certificate of Receipt (FCR).

The affreightment function (5) seems to be the most important one of a freight forwarder. It includes usually booking cargo space or chartering vessels as well as despatching of shipments, including arrangements for inland transportation and handling operations at place of departure, at transhipment points and/or place of destination on behalf and account of a consignor.61 The handling of cargo includes

⁵⁸Following DSLV (2005, p. 2), freight forwarding companies in Germany are heavily engaged in customs brokerage, covering 70% of all import and 90% of all export movements. Furthermore, 75% of all EU transit procedures are issued by them, too. See Kummer et al. (2010, pp. 204–226) for a brief overview of German and Austrian external trade law, the European Customs Code and export, import and customs procedures therein regulated in the case of Germany and Austria.

⁵⁹See Murr (1979, pp. 261–266) or the provisions in 19 CoFR, Chap. 4, Sect. 1641 for the customs brokerage licensing system in the United States.

⁶⁰See Kummer et al. (2010, pp. 193–200) for a brief description of the paperwork needed for cross-border shipments and further references.

⁶¹Of course, arrangement of loading of cargo at the point of departure and unloading it at the point of destination by a freight forwarder is usually restricted to some industrial high-and-heavy goods

loading, unloading, reloading and temporary storage of goods, if needed. A freight forwarder can do these handling operations either by own means (serving a logistics function) or procure it to some specialized handling company. In addition to this, some freight forwarders act as shipping agents for carriers or loading brokers.

As a consolidation function (6), the freight forwarder arranges consolidation of cargo in order to lower freight expenses as well as to save money spent on documentary issues by sending shipments in groupage service or collecting smaller shipments for one consignee to consolidate them at the point of departure and sending them in one shipment. At the point of destination, a freight forwarder in groupage service breaks bulk consolidated cargo and distributes shipments to their consignors.

In concluding a forwarding contract, there is always some forwarding or liability insurance coverage provided up to a certain limit, which is called the insurance function (7) of a freight forwarder. To cover additional interest on a shipment in transit, a cargo insurance can be easily arranged by him upon request. Usually, he holds some open insurance policy contract with a cargo insurance company which allows him to conclude insurance contracts at reasonable premiums and issue certificates on behalf of the insurer.

In FIATA (1975), the logistics function (8) of a freight forwarder was argued only to be consultant in warehousing and distribution matters. But considering the ongoing development in the field of third- and fourth-party logistics today, freight forwarding companies serve a more active role: loading, unloading, reloading and temporary storage of cargo as well as warehousing for a longer time period are logistics services directly conducted by them.62 Sometimes, a freight forwarder offers more complex ones like running consignment stocks or bonded warehouses at his premises. In connection with these handling and warehousing operations, a freight forwarder often performs additional services, like quality inspections, sample drawing, cleaning, drying, pest control of goods and even assembly of products on behalf of his customer.63 Therefore, the logistics function of a freight forwarder is expanded to both consultancy with or without actual service provision.64

or removal operations, where a consignor or consignee is not able to load or unload shipments by own means.

⁶²See Sect. 3.3.4 for a more detailed treatment.

⁶³Cf. Pfohl (2003, pp. 285–287).

⁶⁴This is in line with FIATA (2004) where freight forwarding and logistic services together were defined as "[...] services of any kind relating to the carriage (performed by single mode or multimodal transport means), consolidation, storage, handling, packing or distribution of the Goods as well as ancillary and advisory services in connection therewith, including but not limited to customs and fiscal matters, declaring the Goods for official purposes, procuring insurance of the Goods and collecting or procuring payment or documents relating to the Goods. Freight Forwarding Services also include logistical services with modern information and communication technology in connection with the carriage, handling or storage of the Goods, and de facto total supply chain management. These services can be tailored to meet the flexible application of the services provided."

The fiduciary function (9) of a freight forwarder is closely connected with the dispatch or delivery of goods to a beneficiary person. Concerning the dispatch of goods, Bills of Lading (B/Ls) or other documents of title issued by carriers are compiled, verified and submitted to banks for Letter of Credit (L/C) or Documentary Collection (D/C) transactions. In the same way, freight forwarders can issue own documents of title like the "FIATA Forwarders Certificate of Transport" (FCT), or the "FIATA Multimodal Transport Bill of Lading" (FBL).⁶⁵ On delivery or in connection with warehousing, a freight forwarder releases goods subject to such transactions upon request and submission of documents of title to the beneficiary.

Under the supervision function (10) several operations of a freight forwarder can be subsumed, including tracking and tracing of shipments and documents, arranging inspection of cargo if needed or notification of arrival if special arrangements have to be made in the country of destination. Furthermore, a freight forwarder is obliged to inform his customer as soon as possible in case of irregularities occurring due to loss, damage or late delivery of cargo. In addition to this, cargo claims and freight rate billings usually have to be filed on behalf of a consignor.

Often forgotten is the quasi-banking function (11) of a freight forwarder. In first instance, he pays freight as well as handling charges and ancillary fees e.g. for documents prepared on behalf of a consignor as well as credits the Value Added Tax (VAT) or other duties payable at once to customs house in order to speed up processes.⁶⁶ Moreover, in Cash on Delivery (COD) transactions he collects and submits money for freight costs and/or goods payment from the consignor when he is authorized by the consignor.

The transport function (12) is mainly served in inland road transportation, when a freight forwarder acts on own account as a performing carrier or provides cartage services for pre- and onward-carriage of shipments e.g. for railway companies.⁶⁷ In other transport modes, freight forwarders transporting goods by own means are very rare, so that booking cargo space or chartering whole transport units from carriers on own account for resale as a contracting carrier is a more common feature.

Of course, this list of freight forwarders' functions is by far not complete, other functions not belonging to the core of freight forwarding business may be buying and selling goods as a commercial agent, providing other banking services like providing loans, del credere or factoring, acting as travel agencies or car rental companies only to name a few.⁶⁸

⁶⁵See Jaeger and Laudel (2003, pp. 245–251), Wieske (2003, pp. 123–26), Lorenz (2005, pp. 70– 73), Kummer et al. (2010, pp. 322–323), Spera (2011, pp. 83–97) or http://www.vhspnpenaltynz@. spediteure.de/fiata.htm for a brief description of these FIATA forwarder documents of title.

⁶⁶Cf. Vadnai (1921, pp. 59–60), Eller (1996) or Lorenz (2004, pp. 308–315).

⁶⁷In former times, cartage services for railway companies were at the core of business for many freight forwarders. But today, they play a diminishing role, because of more and more railway companies like the Deutsche Bahn AG gave up the provision of rail-based groupage services.

⁶⁸Cf. Vadnai (1921, pp. 59–63), Schlichting (1931, pp. 56–58), Mayer (1933, p. 13), Ullmann (1950, pp. 106–107), Tschudi (1975, pp. 13–15) or Seiler (1981, p. 85).

Table 2.1 Forwarder functions

Table 2.1 Forwarder functions Authors		Forwarder functions as described by FIATA (1975)									added	Functions
	1	2	3	4	5	6	7	8	9	10	11	12
Metzger (1922:89–99)	C	–	C	C	C	–	A	–	C	–	A	A
Schlichting (1931:11–58)	A	A	C	C	C	C	C	C	A	A	C	C
Mayer (1933:12–14)	C	C	A	A	C	C	A	C	A	–	A	A
Stem (1938:28–35)	X	X	X	–	X	X	X	X	X	–	X	X
Ohling (1950:216–218)	X	–	X	X	X	–	X	–	X	X	X	–
Rosenthal (1950:116–121)	–	–	X	X	X	X	X	–	X	X	X	–
Ullmann (1950:71–107)	C	C	A	A	C	C	A	A	C	A	C	–
Kirchner (1950:11–45)	C	–	A	A	C	C	A	A	A	–	A	A
Smith (1974:178–179)	C	A	C	C	C	A	A	A	–	–	A	A
Mittendorf & Oelfke	A	A	A	A	C	C	A	C	A	A	A	C
(1974:25–29)
D'Amato and D'Amato (1977)	X	X	X	X	X	X	X	–	X	X	X	–
Murr (1979:20–21, 293–297)	X	X	X	X	X	X	X	–	–	X	X	–
Seiler (1981:40–86)	C	C	A	C	C	C	A	C	C	C	A	A
Matthai (1985:24–26) ¨	–	A	A	A	C	A	A	C	C	C	–	A
Dehn (1987:22–23)	A	A	A	A	C	C	A	–	–	–	A	–
Schumacher (1987a:131–139)	C	C	C	C	C	A	A	C	C	C	A	A
Thaler (1990:87–88)	C	A	A	A	C	A	A	A	C	C	A	A
Sherwood and Burns (1992)	X	–	–	X	X	–	X	–	X	X	–	–
Murphy and Daley (1995)	X	X	X	X	X	X	X	X	X	X	X	–
Johnson and Wood	X	–	X	X	X	X	X	–	X	X	–	–
(1996:410–412)
Branch (2000:417–418)	X	X	X	X	X	X	X	X	–	X	X	–
Lucke et al. (2001:265–267)	X	X	X	X	X	X	X	X	X	X	X	X
Pfohl (2003:285–286)	C	A	A	C	C	A	A	A	C	A	A	–
Lorenz (2005:61–63, 492–494)	X	X	X	X	X	X	X	X	X	X	X	X
Dischinger et al. (2005:23–24)	X	X	X	X	X	X	X	X	X	X	X	X

For purpose of comparison, these twelve forwarder functions as described above are compiled in Table 2.1. Some of the authors distinguished between core functions of a freight forwarder (denoted by 'C') and ancillary services provided (denoted by 'A'), whereas other authors made no distinction (denoted by 'X') or did not mention some functions at all (denoted by '-'). According to these 25 authors, who explicitly define functions of a freight forwarder, dispatching of cargo (5) seems to be the foremost prominent core function, followed by consultancy in transportation matters (1), consolidation of cargo for groupage services (6) and serving a fiduciary function (9). According to ancillary functions of a freight forwarder, providing insurance coverage (7), acting as a performing or contracting carrier (12) and caring for customs clearance (3) are most mentioned. Astonishingly, over the whole time frame of over 80 years of freight forwarding business being observed, definitions of basic freight forwarder functions did not seem to be altered, with the authors from the 1920s to the 1950s laying only more stress on knowledge of trade usances and tariff issues and less on the topic of logistics as it is done today.69 Notably, most Anglo-Saxon authors taken into account did not think of a freight forwarder to serve a transport function (12) at all, whereas most of the authors from German-speaking countries indicate this function at least to be ancillary. Reason for this is a different legal status of forwarding business in most Anglo-Saxon countries restricting freight forwarders to organize cargo movements but not to perform transport operations on own account as already shown in Sect. 2.1.1.

2.1.4 Fields of Activity and Specialization in Forwarding Business

Fields of activity of forwarding companies are subject to some specialization along several dimensions like:70

Mode of transport predominantly employed.
Type and scope of forwarding operations.
Geographical scope of operations.
Location of forwarding premises.
Direction of cargo movements and/or
Cargo type and/or shipper industry served.

Considering different modes of transport predominantly served, a freight forwarder is usually named "road haulage agent" in road transport, "railway agent" or "railway forwarder" in railway transport, "loading agent" or "shipping agent" in connection with inland navigation or sea transports and "air freight agent" or "air freight forwarder" in transports by air.71

Concerning the type and scope of operations, a freight forwarder can be engaged in single shipments of general cargo or bulky goods usually called "full-truck-load" (FTL) or "full-container-load" (FCL) as well as "less-than-truck-load" (LTL) or "less-than-container-load" (LCL) in groupage service as a "cargo consolidator."72 Furthermore, he can act as a non-vessel operator (NVO), organize intermodal

⁶⁹At this time, the freight forwarding business as well as the rest of the transportation industry was subject to a strict regulatory tariff framework, see Ohling (1950, pp. 200–201), Ullmann (1950, pp. 75–85).

⁷⁰Cf. Fischer et al. (1930, p. 281), Schlichting (1931, pp. 3–4), Mayer (1933, p. 26), Securius and Boning (1941, p. 13), Ullmann (1950, p. 73), Kirchner (1950, pp. 8–10), Hald (1951, pp. 15–16), ¨ Rossger (1960, pp. 48–50), M ¨ ittendorf and Oelfke (1974, p. 27), Seiler (1981, pp. 87–89), Matthai¨ (1985, pp. 27–6), Schumacher (1987a, p. 141), Chevalier (1989), Thaler (1990, pp. 130–143), Zollner (1990, pp. 44–67), Diederich (2000), Lucke et al. (2001, p. 267) or Pfohl (2003, p. 288). ¨

⁷¹Cf. Securius and Boning (1941, p. 12), Seiler (1981, p. 87), Matth ¨ ai (1985, p. 27), Chevalier ¨ (1989), Gass (1991, p. 64), Triebel et al. (1995, pp. 157–158), Lucke et al. (2001, p. 267), Jaeger and Laudel (2003, pp. 66–68) or Lorenz (2005, pp. 63–66).

⁷²Cf. Mayer (1933, p. 26), Seiler (1981, p. 88) or Matthai (1985, p. 27) or Lorenz (2005, p. 66–67). ¨

transportation of goods as a "combined transport operator" (CTO) or door-to-door shipments as a "multimodal transport operator" (MTO).73 In addition to this, he can carry out customs formalities for exporters and importers as a "customs broker" or "customs agent"⁷⁴ or provide additional logistics services on a long-term regular basis like warehousing as a "warehouse operator."⁷⁵

The location of a freight forwarder's premises are usually closely connected with the mode of transport predominantely employed, because he has to be at the points of transhipment, like a "port agent" at seaports or an "airport agent" at airports.76 Furthermore, he can be a "border agent" or "transit agent" at frontier crossing points.

If the direction of cargo movement is taken into account, a freight forwarder can be labelled as a "loading agent," "export agent" or "FOB-agent" despatching outgoing shipments at the place or country of departure, a "receiving agent," "landing agent," "import agent" or "CIF-agent" despatching incoming shipments at the place or country of destination or being again a "transit agent" at transhipment points somewhere in between.⁷⁷

Concerning the geographical scope of operations, freight forwarders can either act on a national or an international scale with the last labelled as "international freight forwarder" (IFF).78 In addition to this, he can specialize himself on distinct transport routes, destination countries or regions like Eastern Europe, Middle East or Latin America where special expertise and/or equipment needed.

Some freight forwarders are focusing on special cargo types and/or shipper industries.⁷⁹ The most prominent example of such a specialization is the "removal agent" who basically cares for packaging, transportation and warehousing of household goods.⁸⁰ Usually, these forwarding companies extent their operations to distribution services for new furniture, white and brown goods81 as well as transportation of object d'art, computers or other electronic devices and even highand-heavy goods. Other specialization of freight forwarders are fairs and exhibition

⁷³Cf. Matthai (1991), Wolf (2000a,b) or Sect. ¨ 2.2.4.2.

⁷⁴Cf. Chevalier (1989), Wood and Johnson (1996, p. 522).

⁷⁵Cf. Lorenz (2005, p. 67).

⁷⁶Cf. Schlichting (1931, p. 3), Securius and Boning (1941, p. 12), Matth ¨ ai (1985, p. 27), Chevalier ¨ (1989), Lucke et al. (2001, p. 267) or Lorenz (2005, p. 65).

⁷⁷Cf. Schlichting (1931, pp. 3–4), Securius and Boning (1941, pp. 60–69), M ¨ ittendorf and Oelfke (1974, p. 241), Matthai (1985, p. 27), M¨ ¨ uglich (2002, pp. 6–7) or Jaeger and Laudel (2003, pp. 263– 267). The abbreviations "FOB" and "CIF" are standing for the terms of sales "free-on-board" and "cost-insurance-freight" as defined in the "International Commercial Terms" (INCOTERMs) by the "International Chamber of Commerce" (ICC).

⁷⁸Cf. Seiler (1981, pp. 88–89), Matthai (1985, p. 27) or Lorenz (2005, pp. 63–65). ¨

⁷⁹Cf. Fischer et al. (1930, pp. 301–303), Mayer (1933, p. 26), Matthai (1985, p. 27), Lucke et al. ¨ (2001, p. 267).

⁸⁰Cf. Fischer et al. (1930, pp. 303–306), Mayer (1933, pp. 40–41), Charisse (2000) or Lorenz ´ (2005, p. 67).

⁸¹The cargo segment of "white goods" consists of washing machines, stoves, teller machines or other kitchen appliances whereas "brown goods" are television sets or other multimedia devices.

services or project logistics, where they virtually perform all operations requested by their customer, including even mounting of appliances at the point of destination.82 Furthermore, freight forwarding companies specialize themselves on special types of cargo like hazardous materials, food stuff, textiles and garments, mineral oils, coal, wood, cotton, valuables or live animals.

In the last decades, at least two new fields of activity not explicitly mentioned above added to the traditional forwarding business: an emerging segment of courier, express and parcel services and contract or third-party logistics.⁸³ The first grew from freight forwarder's traditional groupage service with a focus on small, standardized shipments.84 The second seems to be a quite natural extension to existing transport, handling and warehousing operations of freight forwarders leading to contract or third-party logistics provision as a "third-party logistics provider" (3PL), a "fourth-party logistics provider" (4PL) or a "lead logistics provider" (LLP).⁸⁵

In Table 2.2, some empirical evidence is shown about fields of activity and specialization of freight forwarding companies in Germany.86 The majority of respondents to the last two surveys conducted regularly in a five years frequency by the former German Bundesverband Spedition und Lagerei (BSL) – now Deutscher Speditions- und Logistikverband (DSLV) – regard international freight forwarding and affreightment of hauliers in road transport as a main field of their activity, followed by outbound logistics, short distance road transport, inbound logistics services, customs brokerage, groupage service and long distance road transport on own account.⁸⁷ Furthermore, about two out of five lay their focus on affreightment of hauliers on road transport, one third on international freight forwarding, and about a

⁸²Cf. Obergfell and Senghas (2000).

⁸³With the stepwise deregulation of postal markets, privatization and international expansion of some national post offices, this small shipment transport segment is now often named courier, express and postal services with parcel services being a part of the express services, see http:// www.m-r-u.de.

⁸⁴Cf. Schumacher (1987a) or Lorenz (2005, pp. 537–602).

⁸⁵See Sect. 3.3.4.

⁸⁶Similar studies were conducted by the Wirtschaftskammer Osterreich, Fachverband der Spedi- ¨ teure, in 1993 and 1998 with the last survey reaching a very low response rate so not more than some figures were made officially. According to the first survey, WIFI (1993, p. 12), Austrian freight forwarders are mostly engaged in international freight forwarding (75%), followed by warehousing (55%), customs brokerage (53%), cartage service (49%), national freight forwarding (44%), long distance road transport by own means (37%), handling of hazardous materials (36%), combined transports (30%), air cargo forwarding (27%), removal agency and/or transport of new furniture (22%), parcel and express service (22%), groupage service (22%) and freight forwarding in inland navigation (8%). Apart from long distance road transport operations by own means, the fields of activity of them seem be quite similar in comparison to the German figures from 2000 and 2005.

⁸⁷Both surveys can be regarded as representative for the German freight forwarding business as a whole as BSL (2000) reported a sample size of 3,300 member companies with a response rate of approximately 50% and DSLV (2005) a sample size of 2,900 with a response rate of approximately 33%.

Table 2.2 Forwarder fields of activity in Germany (Source: BSL (2000), DSLV (2005))

Forwarder fields of activity	Usually offered	(% of total sample)	Focus laid on (% of total sample)
	2000	2005	2000	2005
Groupage service	47	50	26	26
Parcel or express service	23	20	11	7
Long distance road transport on own account	47	47	26	26
Affreightment of Hauliers in road transport	66	73	28	40
Short distance road transport	56	53	24	24
Affreightment of Railway companies	16	13	5	4
International freight forwarding	68	69	33	34
Sea Cargo forwarding	34	38	18	21
Air Cargo forwarding	34	30	16	14
Forwarding in Inland navigation	10	10	4	3
Customs brokerage	47	51	17	19
Handling of Hazardous materials	40	28	10	10
Transportation of new furniture	8	7	3	3
Removal agency (including Warehousing)	10	9	5	5
Forwarding for objects and facilities	13	21	9	9
Warehousing for distribution	37	33	16	14
Warehousing for bulky goods	10	10	4	4
Warehousing for Hazardous materials	8	9	3	3
Inbound logistics services	47	52	21	25
Outbound logistics services	57	59	25	31
Reverse logistics services	9	11	3	4

quarter on groupage service, long distance road transport on own account and short distance road transport with in- and/or outbound logistics services getting more and more important. Remarkably, about one third of all respondents in 2005 agreed to be a forwarding company engaged in shipments by sea or air with 21% of them focusing on sea cargo and 14% on air cargo.⁸⁸ According to BSL (2000, p. 11) as well as DSLV (2005, p. 10), each respondent serves on average seven different fields of activity as shown in Table 2.2, with a third of small forwarding companies with less than ten employees serving not more than three and the majority of big forwarding companies with more than 100 employees serving at least nine fields of activity.

⁸⁸Of course, some of these freight forwarders are engaged in both fields and some of the respondents mentioning international freight forwarding as a their focus may subsume haulage by road, sea and air under this term, so that the real amount of freight forwarders engaged in such field of international operations seems to be rather underestimated as e.g. Dehn (1987, pp. 17–20) concluded. The same might be true in the case of Austrian freight forwarders, where 75% of them are engaged in international freight forwarding and 27% in air cargo forwarding with sea cargo operations not asked for, see WIFI (1993, p. 12).

2.2 Freight Forwarding in Multimodal Transport Chains

After describing the freight forwarder business in general, now the focus is set on freight forwarding operations in multimodal transport chains. After some introductory notes and basic definitions, two fields of multimodal transport – the general cargo segment in ocean shipping and the cargo segment of the airline industry – are thoroughly outlined along functional aspects, institutional environment, technological developments, and market structure. Then inter-organizational relationships in transport or logistic chains and cargo movements at sea- and airports are mapped in order to clarify, where the freight forwarder's business is located there.

2.2.1 Introductory Notes and Basic Definitions

Looking at transport as well as logistics related literature, a cacophony with ever new definitions and redefinitions of already existing terms can be found. In current research literature, often well defined terms like "logistical channel," "transport chain," "logistic chain" or "supply chain" occur, standing for different point of views on transport and logistics processes.

The first term originates in the US-American marketing channel literature, where a "logistical channel" (sometimes also called "physical distribution channel") serves a special marketing function, namely the movement of physical goods in space (transportation or handling) and time (storage or warehousing) or closely related processes of transformation (e.g. packaging, sorting or labelling).89 The other channels (as shown left hand side of Fig. 2.2) are often subsumed under the label "commercial channel," "distribution channel" or "primary channel" having all in common, that they can be regarded as means to trade physical goods as well as services, including negotiation, ordering, ownership change, financing, risking, promotion and information diffusion.⁹⁰ In addition to this, Johnson and Wood (1996,

⁸⁹Cf. Morash (1986, pp. 91–95), Kleer (1991, pp. 9–15), Bowersox and Cooper (1992, pp. 96–98), Rosenbloom (1999, pp. 100–101) or Pfohl (2003, pp. 221–224).

⁹⁰Originally, Vaile et al. (1952, pp. 113–129) considered that a marketing channel consists of eight different forms of flows, which can be grouped in (1) forward flows of negotiation, ownership and physical possession, (2) backward flows of ordering and payment and (3) two-way flows of information, financing and risking. Later on, this classification was step by step updated with renaming physical possession flows in physical distribution or logistics flows and highlighting promotional tasks as the most important component of information flows, see e.g. Stern and El-Ansary (1992, pp. 11–13) or Rosenbloom (1991, pp. 12–15, 1995, pp. 147–149, 1999, pp. 107– 108). Another development was, that these flows in a marketing channel were considered to be channels, too, see e.g. Stern and El-Ansary (1992, p. 14). This notion was adapted by most authors later on like Gompf (1994, p. 11), Johnson and Wood (1996, pp. 30–35) or Trost (1999, pp. 66– 67), so that in contemporary marketing channel literature, always different channels are discussed instead of different flows. Recently, Taylor and Jackson (2000) employed this marketing channel methodology to describe intermodal channels on a national scale.

Fig. 2.2 Logistics channel versus logistic chain

pp. 400–401) claimed, that in international marketing operations a "documentation channel" has to be added due to excessive documentary needs for cross-border shipments to meet terms of sale and payment fixed in the underlying contracts or trade regulations and restrictions imposed by e.g. customs authorities or other governmental agencies at home as well as abroad.

For the term "transport chain," a common basic source of definition is the German DIN 30 781, Part 1:91

"A transport chain is a sequence of technical and organizational processes tied together, where persons or goods are moved from a source to an end. The transport chain has to be regarded as a system.[...] The organizational connection is reached by coordination of information and control systems for legal and commercial issues. The transport chain as a system holds close relationships to other neighbouring systems like manufacturing and consumption of goods."

With logistics services getting more and more important, the term "logistic chain" was introduced, with the transport chain as the basic subcomponent (see right

⁹¹Translation by the author. DIN is an abbreviation for "Deutsches Institut f¨ur Normung," the German Institute for Standardization which develops standards and technical rules. See Gompf (1994, p. 7), Michaletz (1994, pp. 37–39), Isermann (1995, pp. 604–605), Koberlein (1997, p. 192) ¨ or Wolf (2000d) for critical comments on this useful but rather wide definition.

hand side of Fig. 2.2).⁹² Around this provision of physical handling or movement of shipments and transport-related services, additional logistics activities as well as other value added services for the ordering, promotion and information channels are summing up to a "logistic chain." In addition to this, recent research literature employed the term "supply chain" in connection with transport or logistic chains, too.⁹³ This should indicate a far more intensified integration of all business processes along a logistic or a transport chain from source to sink and even beyond them. But for the purpose of this work, the focus is merely restricted on transport chains and less on logistic or even supply chains because third-party logistics and supply chain management operations of freight forwarders are considered to be of minor importance in this work.

Another approach followed in this context to distinguish transport from logistic chains is to align it with the twelve functions of a freight forwarder already discussed in Sect. 2.1.3. Doing this, the consolidation and the logistics function of a freight forwarder may fall in both categories, depending on duration and/or purpose of action. In the case of the logistics function, a distinction between temporary storage and warehousing has to be drawn: The first is regarded as being a short-term, transport-related storage of shipments while the second can be characterized as merely long-term and not directly transport-related.94 Referring to the consolidation function, a (de)consolidation of cargo can occur either in order to perform groupage service to lower transport costs or as a dedicated logistics service on behalf of a shipper or consignee.⁹⁵ A transport chain can be further split in two groups of business processes with actual handling, manipulation or movement of shipments done on one side and transport-related services performed on the other.⁹⁶

In addition to this, several forms of transport chains can be distinguished as shown in Fig. 2.3. ⁹⁷ First, a transport chain can include some transhipment points between source and sink, where a change in the mean of transport occurs leading to a multi-stage transport chain.98 Furthermore, typical subsequent stages of multi-stage transport chains are usually called "pre-carriage," "main haul" and

⁹²See Michaletz (1994, pp. 37–39) or Isermann (1995) for a similar description and Gompf (1994, pp. 101–105) for further discussion concering the air cargo industry.

⁹³See Grandjot (2002, p. 126) or Sch¨uller (2003, pp. 89–156) with the last one developing a model framework for airline-driven supply chains.

⁹⁴Notably, this distinction is exactly drawn in the legal provisions of the HGB or UGB as well as their corresponding GT&Cs as discussed in Sect. 2.1.1.

⁹⁵The very difference between these two types of (de)consolidation is, that the first can be regarded as being transport-related and is done on freight forwarder's account and the second is not necessarily transport-related but performed on behalf and account of a shipper or consignee.

⁹⁶Thinking of storage as a movement of shipments in time without moving them physically, there is no need to add this term explicitly.

⁹⁷See Michaletz (1994, p. 39), Trost (1999, p. 70), Wolf (2000d), Pfohl (2003, pp. 164–166) or Kummer (2006, pp. 47–50) for a similar classification.

⁹⁸Cf. Pirath (1934, p. 160), Wolf (2000b) or Pfohl (2003, p. 164).

Fig. 2.3 Forms of transport chains

"onward-carriage."99 If there is only a direct transport from source to sink and therefore no change in means of transport at any point in the transport chain takes place, it is usually called a single-stage transport chain.¹⁰⁰ Next, referring to the unit of handling at a point of transhipment, a direct handling of cargo can be done, called "break bulk" or "transloading."¹⁰¹ If several pieces of cargo are collected together to form a unit load (like a container, swap body or road trailer) and this unit load is transhipped in a whole, then this kind of transport operation is usually called an 'intermodal transport', if it comes along with a change in the mode of transport employed.102 According to the type of unit load used, several criteria might be applicable. One of them is dependency, referring to whether a unit load needs in ever instance additional handling facilities like a forklift truck or is at least to some extent self-sustained and therefore able to get moved without them.¹⁰³

⁹⁹Cf. Wolf (2000d) or Pfohl (2003, p. 164).

¹⁰⁰Cf. Pirath (1934, p. 160), Wolf (2000b), Pfohl (2003, p. 164) or Kummer (2006, p. 47).

¹⁰¹Cf. Pirath (1934, p. 160), Beplat (1970, pp. 29–30), Jennings and Holcomb (1996).

¹⁰²Cf. Eurostat (2003, p. 103) or Kummer (2006, p. 48).

¹⁰³See Koberlein (1997, p. 98), Seidelmann (2000b), Lucke et al. (2001, pp. 287–293) and Pfohl ¨ (2003, p. 173–178) with the last one calling them shortly "piggy-back" and "container." According e.g. to M¨uller-Feldhammer (1996, pp. 24–25), Aberle (2003, p. 22) or Lorenz (2005, pp. 606– 609) another distinction can be drawn between "accompanied" and "unaccompanied" transports,

In addition to these distinctions as shown, transport chains generally can be classified in uni-, inter- and multimodal ones, referring to the number of modes employed for transportation between source and sink. Since a lot of very different definitions for these three terms are stated in research literature, authors like Jennings and Holcomb (1996), Taylor and Jackson (2000), Jones et al. (2000) or Bontekoning et al. (2004) came to the conclusion, that a fundamental interpretation of these terms does currently not exist. But from a semantic point of view, the distinction between these three types seems to be quite straightforward: Unimodal transport chains are restricted to one single mode of transport, intermodal ones employ two (or more) and multimodal ones at least two so that intermodal transports can be regarded as a particular type of multimodal transports.104 Furthermore, as UN/ECE (2001, p. 18) or Eurostat (2003, p. 103) regard 'combined transport' as an "intermodal transport, where the major part of a journey is by rail, inland waterways or sea and any initial and/or final legs carried out by road are as short as possible," a strict rank ordering can be established: "Combined transport" is some sort of 'intermodal transport' and both are part of "multimodal transport."¹⁰⁵ Moreover, the term "multimodal" is at the same time legally well defined and more thoroughly discussed later on in Sect. 2.2.4.2, which first appeared in the "UN Convention on International Multimodal Transport of Goods" initiated by the United Nations (UN) in 1980.¹⁰⁶ To conclude this discussion and for ease of exposition, in the following 'intermodal' is employed to indicate that unit loads are transferred between two or more modes of transport and "multimodal" is used either as a legal term or it stands for a not explicitly defined number of modes employed along a transport or logistic chain, where either unit loads or cargo are the unit of handling while changing the mode of transport.

According to Beplat (1970, pp. 17–25), principally two approaches to the analysis of transport chains can be distinguished: a functional and an institutional

with the last one as a more common practice in multimodal transports. Further, unit loads can be distinguished according to handling technologies employed during transhipment, like lift-on/liftoff, roll-on/roll-off, swim-in/swim-out, or truck-to-truck (see M¨uller-Feldhammer (1996, pp. 22– 27), Trost (1999, p. 70) or Seidelmann (2000b)).

¹⁰⁴See Slack (1998, p. 263), Wolf (2000d), Lucke et al. (2001, pp. 251–254), UN/ECE (2001, p. 16–17) or Eurostat (2003, p. 103) for quite similar definitions. Furthermore, it is worth to mention that an alternative term for "unimodal" might be also "intramodal," stressing on the fact that changes in means of transport are done within one mode of transport. But this term is not used in literature.

¹⁰⁵Especially among German-speaking authors, the term "combined transport" is often unnecessarily restricted to mainly road-rail transport operations, see e.g. Linden (1966, p. 822) or Aberle (2003, p. 21–22) and short statements by Slack (1998, pp. 263–264) or M¨uller-Feldhammer (1996, pp. 5–6). But in this piece of work, it is used in the original sense defined as a transport of a unit load combining at least two means of transport (or better say a passive mean of transport carried by an active one) and employing one or more modes of transport, see Beplat (1970, p. 31), Koberlein ¨ (1997, pp. 97–98). Seidelmann (2000b) or Eurostat (2003, p. 103).

¹⁰⁶Cf. M¨uller-Feldhammer (1996, p. 4), Wolf (2000d) or Eurostat (2003, p. 103).

Fig. 2.4 Three approaches to analyze transport chains

one.¹⁰⁷ In the functional approach, a focus is set on cargo flows along a transport chain composed of subsequent business processes like transport or handling of shipments. This can be visualized by a graph, consisting of nodes standing e.g. for cargo handling at points of departure, destination or transhipment and edges representing transport operations in between as shown at the top of Fig. 2.4.¹⁰⁸ If a transport chain is realized by more than one individual or institution, it can be

^107

More recent similar description can be found in Michaletz (1994, pp. 39–40), Trost (1999, pp. 67–68) or Wolf (2000d). Further this is also valid in the case of a logistic chain, see Beplat (1970, p. 18).

^108

Originally, Beplat (1970, pp. 18–20) did not think of handling processes at the point of departure or destination being part of a transport chain. Therefore, he got to the conclusion, that direct transports cannot be analyzed in this way. Assuming, that these two endpoints are part of a transport chain allows to define the smallest form of a functional transport chain to be loading a shipment at a source, unloading it at a sink with a transport from source to sink in between, see Köberlein (1997, p. 192) with a similar explanation.

analyzed following an institutional approach. 109 There, individuals or institutions being part of the transport chain and their contractual relationships are then the objects under observation. A visualization of such an institutional transport chain is again a graph, but with individuals or institutions as nodes and their contractual relationships between them represented as edges (see the graph in the middle of Fig. 2.4). For the purpose of this work, a combined approach is chosen to map transport as well as logistic chains as depicted at the bottom of Fig. 2.4. There, nodes are standing for persons or institutions being part of a transport or logistic chain which were connected together by two types of edges: (1) big arrows as directed transport links indicating physical movement of shipments by some carriers from one point to another and (2) bidirectional edges representing some kind of contractual relationships as well as exchange of documents and information between them

As it will be shown in the following, two technological developments in the last 50 years had a tremendous impact on both transport industries as shown in Fig. 2.5. The first, called "cargo handling revolution," comes along with a systematic unitization and/or containerization of cargo starting in the late 1950s and leading

Fig. 2.5 Cargo Handling and Information Processing Revolution (Source: Ojala (1991, p. 13) with extensions)

¹⁰⁹ Beplat (1970, p. 19) claimed, that these persons or institutions must be independent from each other. But thinking of the mirad of cooperative agreements and dependencies occuring in contemporary transport as well as logistic chains, this notion of independency seems to be rather too restrictive and therefore it was dropped.

to tremendous improved transport as well as handling operations.¹¹⁰ Related to information processing along a transport or logistic chain, a second revolutionary development started in the 1970s with an ever faster adoption of new developed information and communication technologies having decisive impacts on transport documentation and information flows.

2.2.2 The Ocean Shipping Industry in the General Cargo Segment

2.2.2.1 Technological Development and Functional Aspects

Seagoing transport is a very old transport business and was performed for centuries by tramp ships employed by merchants on own account. The modern ocean shipping industry dates back to the 1870s, when improving steamship technology first allowed to offer scheduled liner services.¹¹¹ These scheduled liner services were run by shipping companies employing fleets of multi-deck vessels with own cargo handling gear. Although these multi-deck vessels were quite flexible carrying all kinds of general cargo, ship-to-shore handling operations were time and labour intensive.

A first approach to improve handling productivity at seaports was palletization of shipments beginning in the 1940s.¹¹² After their initial usage to improve productivity in lift-on/lift-off ship-to-shore handling, special purpose-built ships with side doors for truck-to-truck operations were designed later on. But this approach of assembling seagoing cargo on a unit load allowing mechanized handling with fork lift trucks through side doors was not so successful, because it was mainly limited to a quay-to-quay or shed-to-shed service for many reasons.¹¹³

¹¹⁰See e.g. Pfohl (2003, pp. 154–162) for a more general description of unitization and containerization.

¹¹¹Cf. Fischer et al. (1930, pp. 218–223), Philipp (1980, pp. 21–23), Richter (1985, p. 4), Stopford (2000, pp. 338–340) or Fleming (2002, p. 370). According to Stopford (2000, p. 343) a liner service is "a fleet of ships, with a common ownership or management, which provide a fixed service, at regular intervals, between named ports, and offer transport to any goods in the catchment area served by those ports and ready for transit by their sailing dates. A fixed itinerary, inclusion in a regular service, and the obligation to accept cargo from all comers and to sail, whether filled or not, on the date fixed by a published schedule are what distinguish the liner from a tramp."

¹¹²Cf. Kirschnick (1969, pp. 133–134), Rath (1973, p. 8), Burg (1975, pp. 45–57) or Stopford (2000, p. 4).

¹¹³Following Seidelmann (1969, p. 22) or Burg (1975, pp. 45–57). To Kaselau (1971, pp. 10–11), ¨ Zinnecker (1972, pp. 64–65), Berg (1975, pp. 53–57), major problem fields of this truck-to-truck handling were protection of goods, through transports under bond and a lack of standardization in unit load dimensions. In addition to this, pallets needed multiple mechanical handling at each point of interchange like from inland carrier to shed, from shed to quay, from quay to ship and

During the 1960s, the ocean shipping industry underwent a tremendous change, well known under the term "the container revolution."¹¹⁴ Long before that time, non-standardized containers were already used for regular shipments in railway operations with early experiments already in the 1830s in the United Kingdom and France, "liftvans" in removal business around the 1900s as well as containerized military goods transports of the Allied Forces during World War I and II.¹¹⁵ But the first large scale container transport operations as part of commercial transport activity can be dated back to the late 1950s where e.g. Sea-Land Service, a former trucking company set up by Malcolm MacLean, started to transport wheel-less trailers between Newark and Houston in 1956.¹¹⁶ The first regular inter-continental container service started in 1966 on the North Atlantic, again operated by Sea-Land Service.117

From that time on, a rapid adoption of the container technology occurred in the ocean transport industry, which resulted in an overwhelming dominance of containerized shipments, especially in the segment of seagoing general cargo.118 At least two contributing factors led to this development:119

• At the advent of containerization, a world-wide accepted ISO-container standard applied to dimensions, strength, gross weight and other specifications for seagoing containers (called ISO-1) as well as inland containers (called ISO-2) was

onboard a ship, too. So it can be regarded only as a temporary solution between traditional break bulk operations and containerization of shipments in the case of the ocean shipping industry.

¹¹⁴The term "container" can be tracked back to "continere" (Lat.) which simply means to enclose or surround something, see Rackwitz (1969, p. 7).

¹¹⁵See Fischer et al. (1930, pp. 304–205), Mayer (1933, pp. 39–40), Mayercordt (1958, pp. 7–13 and 44–46), Rackwitz (1969:4–7), Seidelmann (1969, pp. 14–17), Baumoller (1970, pp. 8–16), ¨ Beplat (1970, p. 77), Burg (1975:66–69), Richter (1985, pp. 5–6), Michaletz (1994, pp. 64–66), Wood et al. (1995, pp. 164–165), Wood and Johnson (1996:246), Exler (1996, p. 40), Slack (1998, pp. 264–265) or Fleming (2003:74–75) with further references. Beplat (1970, pp. 77–83) e.g. supports this view dividing the history of containerization in four subsequent phases: (1) unsystematical application long before World War I, (2) systematical military application beginning with World War I, (3) systematical intra-continental application in the United States after 1956 and (4) systematical inter-continental application of containers after 1966. Rath (1973, pp. 5–6) even dates the early beginnings of containerization back to the time of the Roman Empire, where cages with wild beasts were transported as unit loads in an intermodal manner from Africa to the Circus Maximus in Rome.

¹¹⁶Cf. Seidelmann (1969, p. 18), Beplat (1970, pp. 77–79), Rath (1973, pp. 26–31), Burg (1975, p. 111–114), Michaletz (1994, pp. 65–66), Talley (2000) or Levinson (2006).

¹¹⁷Cf. Rackwitz (1969, p. 11), Seidelmann (1969, pp. 18–19), Beplat (1970, pp. 79–80), Burg (1975, p. 114–115), Exler (1996, pp. 40–41), Stopford (2000, p. 341) or Talley (2000), or Levinson (2006).

¹¹⁸Of course, other modern handling methods like roll-on/roll-off for rolling cargo and swimon/swim-off for bulky shipments were developed at that time, too. But being only interesting for nice markets of breakbulk cargo or 'neo-bulks', they had a minor impact on ocean shipping industry as a whole, see Kirschnick (1969, pp. 131–133), Kaselau (1971, ¨ pp. 14–18), Williams (1973), Gilman (1977) or G¨ursel (1996, pp. 29–41).

¹¹⁹Cf. Stopford (2000, pp. 340–342).

developed.¹²⁰ In concordance with ISO-Norm 668, standard seagoing containers have common dimensions of 8' height and 8' width with a length of 5', 6' 2/3", 10', 20', 30' or 40'.¹²¹ In 1976, the maximum height of these standard ISO-containers was increased to 8' 6" and later on, high cube containers were introduced with a height of 9'6" in order to provide additional volume for cargo. Today, most of the seagoing containers are either 20' or 40' ones, despite they seem to be a second-best solution with regard to the European pallet standards.¹²²

• Another important feature of the container revolution is the facilitation of inter- or multimodality in cargo transportation in order to build up integrated transport chains allowing a provision of door-to-door services.123 According to Seidelmann (2000b), there are four closely interrelated aspects of integration which have to be taken into account: a technical, an organizational, a pricing and a legal one. First, the usage of standardized containers as a unit load is suitable for all means of transport involved in a multimodal transport chain. This led to an improved transferability of shipments between different means or modes of transport without a need for unloading and reloading it manually at transhipment points.¹²⁴ Second, technical problems being solved, multimodal transport operators (MTOs) could combine the most suitable modes of transport in order to organize a more or less seamless transport of shipments from doorto-door, employing different modes of transport on pre-carriage, main haul and onward-carriage.¹²⁵ Third, being a MTO and providing a door-to-doorservice, an all-inclusive price for the whole multimodal transport chain could

^120'ISO' is an acronym for the "International Standardization Organization." See Rackwitz (1969, pp. 94–105), Seidelmann (1969, pp. 22–34), Kaselau (1971, p. 12), Rath (1973, pp. 49–67), ¨ Burg (1975, pp. 111–125), Michaletz (1994, pp. 54–63), Benson et al. (1994, pp. 204–206), G¨ursel (1996, pp. 21–29), Hebeler (1998), Branch (2000, pp. 71–79), Seidelmann (2000a), Pumpe (2000, pp. 5–11) or Jaeger and Laudel (2003, pp. 200–203) for further details about this ISOstandardization of containers.

¹²¹Notably, the US–American pioneers in container shipping used non-ISO-standard containers like the Sea-Land-Container with a length of 35' and the Matson-Navigation-Container with a length of 24', see Seidelmann (1969, p. 23), Rackwitz (1969, p. 101), Baumoller (1970, p. 5), ¨ Rath (1973, p. 37) or Michaletz (1994, p. 66). In addition to the ISO-Norm 668, further important specifications for containers can be found in ISO-Norms 1496 and 6346.

¹²²See Hebeler (1998) or Pfohl (2003, pp. 158–159) with further comments.

¹²³Cf. Benson et al. (1994, p. 202), Slack (1998, pp. 264–265), Seidelmann (2000b) or Fleming (2002, pp. 73–74). Branch (2000, pp. 94–100) used the term "multi-modalism" instead of intermodality, but his conclusions are likely the same. To him, "multi-modalism is the process of providing a door-to-door or warehouse-to-warehouse service to the shipper which embraces two or more forms of transport, and involves the merchandise being conveyed in a unitized form in the same unit througout transit" (Branch (2000, p. 94)).

¹²⁴Cf. Jaeger and Laudel (2003, pp. 199–200).

¹²⁵Cf. Wolf (2000b,c) or Biebig et al. (2004, pp. 247–261). According to Biebig et al. (2004, pp. 252–256), there are many variants of MTO, including both non-carrier MTOs like freight forwarders or shipping agents and carrier MTOs like shipping companies, road hauliers or railway companies.

be offered to shippers.126 Finally, with the "UN Convention on International Multimodal Transport of Goods," an obligatory legal framework for multimodal transport was attempted in 1980, but it never came into force.¹²⁷ Instead of this, several uniform rules for transport documents and carrier liability in multimodal transport operations were developed, which became a commercial "best practice."¹²⁸ Furthermore, some countries like Germany, the Netherlands or the United States incorporated special provisions for multimodal transports in their national transport laws.¹²⁹

These two factors, namely standardization of seagoing containers and facilitation of intermodality, allowed a tremendous increase in productivity along the whole logistic chain of seagoing cargo. But containerization of shipments and automating cargo handling processes at seaports came along with heavy investments in mechanized systems and equipment for both sea and landside operating companies.¹³⁰ Furthermore, shipping companies investing in purpose-built containerships of ever increasing size perceived huge economies of scale on the sea leg. Whereas the early containerships of the first generation were converted liquid bulk carriers with a loading capacity of less than 1,000 TEUs, the fifth generation of purpose-build cellular container ships can carry more than 7,500 TEUs and even bigger ones with more than 10,000 TEU recently came into service.¹³¹

¹²⁶See Biebig et al. (2004, pp. 263–266) about pricing in multimodal transport chains.

¹²⁷Cf. M¨uller-Feldhammer (1994, p. 273), Puttfarken (1997, p. 180), Asariotis et al. (1999, pp. 21– 22), Spera (2002, pp. 241–248) or Clarke (2002, p. 74).

¹²⁸As a first attempt in this direction, the ICC issued "Uniform Rules for a Combined Transport Document" (ICC-Publication No.273) in 1973, which were revised in 1975 with ICC-Publication No.298 and replaced by the "UNCTAD/ICC Model Rules for Multimodal Transport Documents" (ICC-Publication No.481) in 1992, see M¨uller-Feldhammer (1994, 1996:267–274), Asariotis et al. (1999, pp. 21–25), Wolf (2000a,c), Clarke (2002, pp. 74–80) or Biebig et al. (2004, pp. 259– 261). Further, all these works have been incorportated in the GT&Cs for the "FIATA Multimodal Transport Bill of Lading" of 1992 replacing the "FIATA Combined Transport Bill of Lading" of 1970 and influenced the "FIATA Model Rules for Freight Forwarding Services" of 1996, see M¨uller-Feldhammer (1994, pp. 272–273) or Ramberg (2000).

¹²⁹See M¨uller-Feldhammer (1996, pp. 99–105) or Asariotis et al. (1999, pp. 23–24). In the case of Germany, - 452 HGB gives legal provisions for carriage of goods using various modes of transport on the basis of a single contract of carriage, but their applicability to multimodal transport chains including a leg by sea is more and more questioned by authors like Rabe (2000), Blettgen (2001), Teutsch (2001) or Drews (2003).

¹³⁰See Baumoller (1970, pp. 71–78), K ¨ aselau (1971, pp. 12–14), Rath (1973, pp. 8–9), Benson ¨ et al. (1994, p. 211–214), Slack (1998, pp. 273–274), Bohme (2000c) or Levinson (2006) with ¨ further references.

¹³¹Cf. Slack (1998, pp. 267–268) or Peters (2001, pp. 9–10). Further, AXS-Alphaliner (2010) see a rise in the cellular container ship fleet above 7,500 TEU from 272 (31/12/2009) to 540 until 31/12/2013 of which 202 alone have a capacity of more than 10,000 TEU (projections based on orderbook as of 01/11/2010 assuming that no ships are deleted after this date).

A second, still ongoing technological development is the information processing revolution in ocean transport operations (see Fig. 2.5) with early beginnings in the mainframe computer system era of the 1960s and 1970s.132 In the early 1980s, first seaport-driven Cargo Community Systems (CCSs) were established like the "Daten-Kommunikationssystem" (DAKOSY) at Hamburg or the "Datenbank Bremische Hafen" (DBH) at Bremen and Bremerhaven. ¨ ¹³³ They are on-site value added networks (VANs) for all participants engaged in cargo and transport documentation handling at seaports.¹³⁴ In addition to this, they provide direct links to customs authorities and carriers.

In the mid 1990s, first internet-based maritime information services like SeaNet and carrier-driven CCS like INTTRA, GTN and CargoSmart occurred in the ocean shipping industry.¹³⁵ While internet-based maritime information systems offer "only" all kind of transport-related information, these carrier-driven CCS allow booking of cargo space, transmitting documents as well as tracking and tracing shipments in transit via Internet applications.

Furthermore, some customs authorities like in Canada, Norway and the Netherlands began in the late 1980s to facilitate cross-border movements by employing information technologies, but wide-spread adoption of electronic customs clearance systems started in the late 1990s.¹³⁶ One prominent example is the ongoing EU customs procedures reform, with the main objective to end in a totally paperless customs clearance environment.¹³⁷

Looking at this rapid development and adoption of information technologies in the ocean transport industry, one serious problem is still unsolved: the electronic transmission of B/Ls as the main transport document in sea transport operations with its special fiduciary function.¹³⁸ Although several ambitious approaches like the SEADOCS-Project by INTERTANKO and Chase Manhattan Bank in 1986, the "CMI Rules for Electronic Bills of Lading" of the Comite Maritime International ´ (CMI) in 1990 or the BOLERO Bill of Lading by BOLERO.NET in 1998, paperbased B/Ls are still used if there is a need to transfer title of shipments in transit. The main reason to this seems to be, that such electronic B/Ls are only valid in a special legal environment backed by all participants (e.g. shippers, carriers, freight forwarders and banks) being member of a closed community.

¹³²See Stopford (2002) for a general overview for this and the following.

¹³³See Salzen (2000), Pumpe (2000, pp. 54–59), Jaeger and Laudel (2003, pp. 438–442), Kummer et al. (2010, p. 340–343), http://www.dakosy.deorhttp://www.dbh.de.

¹³⁴See Gudmundsson and Walczuck (1999) or Dautel (2000) with further references.

¹³⁵Cf. Kummer et al. (2010, pp. 339–340).

¹³⁶Cf. Hellberg and Sannes (1991), Heaver (1992), Carr and Crum (1995), Briggs (1997) or Thorby (1997).

¹³⁷Cf. Kummer et al. (2010, pp. 222–226).

¹³⁸See Kummer et al. (2010, pp. 248–251) with further references.

2.2.2.2 Institutional Environment and Market Structure

With the introduction of the first scheduled liner services in the 1870s, a fierce competition on routes served by more than one shipping company broke out leading to a liner conference system.¹³⁹ At this time highly seasonal volumes of cargo came together with an over-capacity of ships and so tariffs were forced down to levels often not covering average costs. In order to relax competition, shipping companies operating on the same trades formed cartels in order to limit sailings and to fix rates. In addition to this, they agreed on ports served, goods carried and even shared freight revenues obtained. The first liner conference was established in 1875 by shipping lines operating between the United Kingdom and Calcutta.¹⁴⁰ This liner conference system builds on the following three blocks.¹⁴¹ First, there are closed liner conferences, which control membership, share cargo and charge lower contract rates for regular shipper. Second, other liner conferences are open ones, which allow any shipping company to join agreeing to the rate agreements. Third, there are shipping companies not member of a liner conference, called outsiders, which usually undercut freight rates offered by the liner conferences. However, a shipping company can be member of liner conference on one trade, whereas be an outsider on the other, because liner conferences are formed for distinct trade routes and ranges as well as ports of call.

Within a few years, this liner conference system got very powerful so that it was constantly under attack and subject to anti-trust measures.¹⁴² Although this, it reached its peak in the beginning 1950s but declined in the following years due to strict legal measures like the "UNCTAD Code of Conduct for Liner Conferences" of 1964 (and set into force in 1983), the US Merchant Shipping Act of 1984, the Regulation (EEC) 4056/86 or the US Ocean Shipping Reform Act of 1998, which all still allow to form liner conferences, but within severe limitation of activities.¹⁴³ Moreover, with Regulation (EC) 1419/2006 the block exemption for liner shipping conferences according to Regulation (EEC) 4056/86 was abandoned

¹³⁹Cf. Svendsen (1958, pp. 334–336), Philipp (1980, pp. 5–7), Scheibe (1980, pp. 58–59), Stopford (2000, pp. 348–349) or Sjostrom (2004).

¹⁴⁰Cf. Svendsen (1958, p. 336), Philipp (1980, p. 5), Scheibe (1980, p. 58), Brooks et al. (1993, p. 222), Wood et al. (1995, pp. 106–108), Branch (1998, p. 453), Bohme (2000d), Stopford (2000, ¨ p. 349), Fleming (2002, p. 380) or Sjostrom (2004).

¹⁴¹Cf. Svendsen (1958, pp. 338–348), Geuther (1990, pp. 29–31), Ordemann (1996, pp. 47–49), Bohme (2000c) or Biebig et al. (2004, pp. 217–219). ¨

¹⁴²Cf. Rosenthal (1950, pp. 39–51), Wood et al. (108–116), Clarke (1997), Bohme (2000d), ¨ Stopford (2000, pp. 349–351) or Sheppard and Seidman (2001:356–358).

¹⁴³Cf. Philipp (1980, pp. 9–16), Franck and Bunel (1991), Brooks et al. (1993, pp. 228–231), Slack et al. (1996, pp. 289–290), Bohme (2000a, pp. 13–14, 2000d), Sheppard and Seidman (2001, ¨ p. 358–360), Reitzes and Sheran (2002), Biebig et al. (2004, pp. 221–223), Dinger (2004) and Benacchio et al. (2007).

so that since 18/10/2008 liner shipping conferences calling ports in the EC are now unsustainable.¹⁴⁴

After the 1960s, huge investments needed for running scheduled container services led to a rapid consolidation in the ocean shipping industry. Some shipping companies focused on specialized fields, other went bankrupt or were objects in mergers or acquisitions.¹⁴⁵ In addition to this, liner consortia and later on in the 1990s strategic alliances of liner shipping companies were formed as an alternative to the traditional liner conference system.¹⁴⁶ Consortia are joint ventures of shipping companies pooling already existing vessels or new built ones to jointly operated fleets whereas strategic alliances are co-operative agreements of otherwise independent shipping companies to establish joint-services on certain trades.147 The most prominent forms of co-operation in strategic alliances are based on (1) vesselsharing agreements to maintain a certain level of service frequency in one trade, (2) slot charter agreements, enabling shipping companies to offer cargo space on a particular route without the need of shipping them by their own vessels or (3) joint service agreements combining both with a considerable higher commitment.¹⁴⁸ After years of ongoing consolidation in the ocean shipping industry, there are two main blocks of container shipping companies remaining:149

• a number of powerful liner alliances or shipping groups like the Grand Alliance (Hapag-Lloyd, P&O-Nedlloyd (until 2006), Nippon Yushen Kaisha (NYK), Oriental Overseas Container Line (OOCL), Malaysian International Shipping Corp. (MISC)) or the New World Alliance (American President – Neptune Orient Lines (APL-NOL), Mitsui Osaka Shosen Kaisha Lines (MOL), Hyundai Merchant Marine (HMM)) and

¹⁴⁴See Czerny and Mitusch (2005), Schmidt (2007), Benacchio et al. (2007) or http://ec.europa.eu/ comm/competition/antitrust/legislation/maritime.

¹⁴⁵Cf. Slack et al. (1996, p. 291), Odrich (1998, pp. 81–83), Stopford (2000, pp. 376–378), Bohme ¨ (2000a, pp. 65–66), DVB (2002, p. 61) or Panayides and Gong (2002). Almost all of the pioneering shipping companies in container shipping of the late 1960s like Sea-Land Service, United States Lines, Moore Mc Cormack Lines or American Export Isbrandsten Lines did not survive.

¹⁴⁶Cf. Burg (1975, pp. 126–130), Richter (1985, pp. 98–117), Geuther (1990, pp. 28–35), Brooks et al. (1993, p. 223–228), Ordemann (1996, pp. 45–73), Slack et al. (1996, pp. 291–292), Clarke (1997, pp. 22–23), Slack (1998, pp. 272–273), Bohme (2000d), Sheppard and Seidman (2001), ¨ Peters (2001, pp. 8–9), Ewert (2006) or Herrebosch (2008).

¹⁴⁷See Rath (1973, pp. 37–40), Scheibe (1980, pp. 74–85), Richter (1985, pp. 100–117), Kindleberger (1993, p. 172–173), Brooks et al. (1993, pp. 223–227), Ordemann (1996, pp. 65–67), Branch (1998, p. 454–455) or Sheppard and Seidman (2001, pp. 352–356). The most prominent consortia is Atlantic Container Lines (ACL) established in 1965 (now part of the Grimaldi Group), see Burg (1975, pp. 127–128). Other consortia founded at this time were Overseas Container Limited (OCL), Associated Container Transportation (ACT) or HAPAG-LLOYD Container Linien (now Hapag-Lloyd), see Seidelmann (1969, pp. 54–56) or Rath (1973, pp. 38–40).

¹⁴⁸Cf. Evangelista and Morvillo (1999, pp. 30–32) or Sheppard and Seidman (2001, pp. 353–355). 149Cf. Stopford (2000, pp. 377–378), Heaver et al. (2001, p. 294), Slack et al. (2002), DVB (2002, p. 62), Midoro et al. (2005, pp. 96–101) or Kummer et al. (2010, pp. 296–297).

• some so-called "outsiders" which are independent carriers just running their business on their own like Maersk, Mediterranean Shipping Company (MSC), Evergreen, and Hamburg S¨ud as the most prominent ones.

2.2.3 The Cargo Segment in the Airline Industry

2.2.3.1 Institutional Environment and Market Structure

From the very beginning, institutional environment was of special importance in the air transport industry as a whole. Even before the first motorized aircraft flight was performed by Wilbur and Orville Wright in 1903, legal provisions were laid down for vehicles based on the principle of the "lighter than air" technology in the eighteenth and nineteenth century.¹⁵⁰ After a first unsuccessful attempt of establishing rules for international aviation at the Paris Air Navigation Conference in 1910, the Paris Peace Conference after the World War I resulted in the Paris Convention of 1919. There, basic rules of cross-border flight operations were fixed and the "Commission Internationale pour la Navigation Aerienne" (CINA) ´ as the first supranational organization for civil aviation issues was established in 1922.¹⁵¹ Also in 1919, six European airlines formed the "International Air Traffic Association" (IATA) in order to standardize their transport operations.¹⁵²

Towards the end of World War II, fundamental issues concerning post-war civil aviation like the freedom of the air rules and basic regulations for scheduled air transport operations were negotiated at a conference in Chicago, resulting in the Chicago Convention of 1944. At the end of this conference, the "International Civil Aviation Organization" (ICAO) was founded as a supranational union of all nations which participate in aviation traffic in order to supersede the CINA and further improve international co-operation on a national and governmental level.¹⁵³ Furthermore, the pre-war IATA of 1919 was re-established under the name "International Air Transport Association" (IATA).¹⁵⁴ Both organizations work closely together to facilitate civil aviation in every aspect. Important outcomes of this co-operation are technical, operational and legal provisions leading to a very secure but heavily regulated transport environment. Prominent examples with

¹⁵⁰Cf. Schaffer (2003, pp. 377–379). ¨

¹⁵¹Cf. Berendt (1961, pp. 194–1995), Buchwald (1973, pp. 54–55) or Schaffer (2003, p. 379). ¨

¹⁵²Cf. Maaß(1927:81–85), Stahlberg (1931, pp. 46–50), Kreuter (1931, p. 11), Berendt (1961, p. 30–32), Buchwald (1973, p. 53), Kehrberger (1996) or Lorenz (2005, p. 505).

¹⁵³Cf. Berendt (1961, pp. 198–199), Buchwald (1973, pp. 55–60), Kehrberger (1996), Opgenhoff (1997, p. 33), Grandjot (2002, pp. 55–56), Schaffer (2003) or Lorenz (2005, pp. 500–501). ¨

¹⁵⁴Cf. Berendt (1961, pp. 75–82), Buchwald (1973, pp. 79–85), Kehrberger (1996), Opgenhoff (1997, pp. 33–34), Grandjot (2002, pp. 57–59) or Lorenz (2005, p. 505).

special importance for air cargo operations are common conditions of carriage and transport documents or provisions for the transport of hazardous materials.¹⁵⁵

Today, the IATA is the most important interest group in civil aviation, resembling to some extent to a huge price and condition cartel of airlines and their approved agents, which is often called the "IATA agency system."156 There are three IATA traffic conference areas, where freight rates as well as other operative and legal provisions are negotiated.157 But today these official airport-to-airport IATAfreight rates published in "The Air Cargo Tariff" (TACT) are usually undercut by up to 85%, so that they can be regarded nowadays merely as a recommended price not reflecting actual competition.158 Furthermore, the IATA serves a unique banking function for their member airlines and approved agents with the "IATA Clearing House" (ICH) and the "IATA Cargo Accounts Settlement System" (IATA-CASS).159

Due to operational needs and an ongoing heavy competition, co-operations between otherwise independent operating airlines are quite common, both in the passenger and air cargo segment.¹⁶⁰ First, a common form of co-operation are interlining agreements backed by the IATA Interlining-System, where a mutual recognition of operational and legal issues allows a "seamless travel" of cargo or passengers on a single waybill or ticket conducted by two or more airlines. In pool agreements, two airlines agree to serve destinations as a joint operation, sharing their costs and revenues. Furthermore, there are numerous code sharing agreements, where a flight performed by one airline called "operating carrier" has several flight designators from other partner airlines marketing this flight, too. In addition to this, some airlines formed cargo alliances within the last years. Examples to this are161

• New Global Cargo or WOW (Lufthansa Cargo, SAS Cargo162, Singapore Airlines Cargo and Japan Airlines Cargo),

¹⁵⁵Cf. Kehrberger (1996), Branch (2000, pp. 160–174), Jaeger and Laudel (2003:467–268) or Kummer et al. (2010, pp. 303–304).

¹⁵⁶Cf. Wood et al. (1995, pp. 148–151), Wood and Johnson (1996, p. 546), Pfohl (2003, p. 351) or Sch¨uller (2003, p. 13). Further, see Berendt (1961, pp. 74–80), Kehrberger (1996), Opgenhoff (1997, p. 34–37) or Lorenz (2005:505–517) for a more elaborated overview of the IATA.

¹⁵⁷Actually, the TACT is subject to a block exemption rule according to Regulation (EC) 487/2009 which demands a free formation of prices within EC and EFTA, see e.g. Spera (2011, p. 254).

¹⁵⁸Cf. Opgenhoff (1997, pp. 34–35), Becker (1999, pp. 45–50), Bertsch (2000a), Grandjot (2002, p. 153–158), Sch¨uller (2003, pp. 17–19), Shaw (2004, pp. 189–192) or Spera (2011, p. 254–257).

¹⁵⁹Cf. Berendt (1961, pp. 133–134), Kehrberger (1996), Opgenhoff (1997, pp. 35–37), Grandjot (2002, p. 58–59) or Kummer et al. (2010, pp. 338–339).

¹⁶⁰Cf. Berendt (1961, pp. 138–142), Gopfert (1994), Kehrberger (1996), Opgenhoff (1997, pp. 42– ¨ 44), Albers (2000, pp. 35–43), DVB (2001, p. 26), Grandjot (2002, pp. 106–114) or Kummer et al. (2010, p. 311–313).

¹⁶¹See DVB (2001, pp. 26–27), DVB (2002, pp. 48–49) and Kummer et al. (2010, pp. 312–313) with further references. In the meantime, the WOW cargo alliance was terminated by Lufthansa Cargo.

^162'SAS' is the abbreviation for "Scandinavian Airlines System," an inter-Scandinavian airline consortia.

• Sky Team Cargo (Air France Cargo, Korean Air Cargo, Delta Air Logistics, Aeromexico Cargo, Alitalia Cargo and Czech Airlines Cargo, KLM Cargo163 joined by Northwest Airline Cargo in 2005).

To some extent, these cargo alliances can be considered as being a natural extension to already existing strategic alliances in the passenger segment.

2.2.3.2 Technological Development and Functional Aspects

In contrast to the ocean transports industry discussed before, the air cargo industry developed very late, only in the twentieth century.164 Shortly after World War I with a rapid development of aircraft technology, a lot of airlines started scheduled air transport operations for passengers, mail as well as cargo backed by heavy governmental subsidies.¹⁶⁵ At that time, aircraft flying range and payload capacity were considerable low, so that intercontinental air traffic could only run by flying boats or airships and other cargo than parcel shipments were quite unusual.¹⁶⁶

World War II was a very important landmark contributing to the continued development of the air freight industry.¹⁶⁷ During the war, aircrafts proved their value in many military logistics operations like landing operations of paratroopers or supply of locked troops. In the late 1940s, even a whole city like Western Berlin was kept supplied with food stuff and fuel by the Berlin Airlift of the Allied Forces during the blockade by the Soviet Union. In the 1960s, civil aviation adopted jetpropulsion engines developed during the World War II which lead to a replacement of propeller-driven by jet aircrafts due to their superior lifting capacity.¹⁶⁸ In the

^163'Koninklijke Luchtvaart Maatschappij' (KLM) – or better say the Royal Dutch Airlines – merged recently with Air France with the latter in a leadership role in this joint company.

¹⁶⁴Wells (1999, p. 364), considered the first air cargo shipment to be five bolts of silk cloth strapped onto the passenger seat of a plane of from Dayton to Columbus, where a department store wanted to sell strips of the cloth as mementos of "the first air shipment." Further, Wells (1999, p. 364) reported about a sack of mail transported from Albany to New York City for the Post Office Department in 1910, whereas Puffer (1949) referred to first experiments in air-mail services between Nassau Boulevard, Long Island, New York and Mineola conducted in 1911, Wood et al. (1995, pp. 127– 128) saw the first occurence of an air freight transport 1911 in India where 6.500 pieces of mail were carried on distance of 5 miles and Grandjot (2000) dated it back to 1911, too, but failed to give more information about.

¹⁶⁵Cf. Maaß(1927:8–71 and 95–107), Schenk (1930, pp. 75–124), Stahlberg (1931, pp. 14–36 and 86–105), Kreuter (1931, pp. 7–14), Ulderup (1935, pp. 14–74 and 101–146), Ziegler (1938, pp. 50–56), Maetz (1937, p. 12–22), Rosenthal (1950, pp. 419–420), Berendt (1961, pp. 12–23 and 34–47), Buchwald (1973, p. 18–21), Graham (1995, pp. 10–12), Wood et al. (1995, pp. 128–131) or Wells (1999, p. 35–47).

¹⁶⁶Cf. Maetz (1937, pp. 39–43), Ziegler (1938, pp. 56–60) or Lorenz (2005, p. 495).

¹⁶⁷Cf. Berendt (1961, pp. 74–63), Buchwald (1973, pp. 22–27), Wood et al. (1995, p. 131), Wells (1999, p. 48–51) or Lorenz (2005, p. 495).

¹⁶⁸Cf. Berendt (1961, pp. 214–218), Heinrich (1966, pp. 92–140), Krebs (1974, pp. 96–100), Graham (1995, p. 12–14), Wells (1999, pp. 51–54) or Fricke and Dussoye (2000).

following decades, bigger and bigger aircrafts were build like the Boeing 747 Jumbo Jet with a payload capacity of over 100 tons in the pure freighter version or former Soviet military aircrafts like the Antonov AN-124 Ruslan with a maximum payload capacity of about 120 tons or the Antonov AN-225 Mrija with a maximum payload capacity of about 250 tons.169

In the early 1960s, palletization of air freight cargo was enforced in order to allow loading and unloading cargo of aircrafts within minimum time.170 In 1963, the IATA started a unitization programme, which lead in the 1970s to a rapid containerization of air cargo.¹⁷¹ This happened also due to a special tariff concept issued in 1969 with bulk unitization charges as a "freight-all-kind"-rate for unit load devices (ULDs) being lower than a normal commodity rate.¹⁷² Today, these standardized air cargo pallets with nets or igloos and containers are indispensable for contemporary air freight operations. There are about 20 different sizes of ULDs existing up to the largest one with the dimensions of a twenty-foot ISO-1 container. But most of these ULDs are much smaller and they have a special design to fit exactly on the main or lower deck of aircrafts without wasting too much space. This extremely adoption to aircraft loading space and the need for roller-bed handling facilities prevent to some extent intermodality in the case of the ULDs.¹⁷³ So up to now, most of these ULDs are build up and/or broken down at the airport or premises nearby.174

Despite "Sales And Business Reservations Electronically" (SABRE), the first airline computer reservation system for passenger transport was established already in 1962, the information processing revolution in air cargo operations started in the early 1970s.175 At about the same time, several airports around the world developed dedicated on-site CCS to control cargo handling operations.176 Furthermore in 1978, an electronic customs procedure system called 'Automatisiertes Luftfrachtabwicklungsverfahren' (ALFA) was established at German airports, which is now

¹⁶⁹See Buchwald (1973, pp. 181–201), Burg (1975, pp. 196–198), Wood et al. (1995, pp. 136– 139), DVB (2001, pp. 15–19), Lorenz (2003, pp. 505–506), Sjogren (2004) or Kummer et al. ¨ (2010, pp. 314–315) with further references.

¹⁷⁰Cf. Seidelmann (1971, pp. 96–99), Ashford et al. (1997, pp. 293–304) or Shaw (1999, p. 141).

¹⁷¹Cf. Seidelmann (1969, pp. 100–104), Rath (1973, pp. 287–289), Smith (1974, pp. 165–173), Burg (1975, p. 89–91) or Ashford et al. (1997, pp. 293–297). Before that time, usage of containers in air cargo transport operations on regular basis was virtually non-existent, see Mayercordt (1958, pp. 47–48).

¹⁷²Cf. Seidelmann (1971, pp. 112–117), Burg (1975, pp. 194–195), Shaw (1999, pp. 141–142), Wells (1999, pp. 380–382), Grandjot (2002, pp. 61–65) or Lorenz (2005, pp. 515–516).

¹⁷³Cf. Seidelmann (1971, pp. 117–119) or Grandjot (2000, 2002, pp. 95–96).

¹⁷⁴Cf. Smith (1974, p. 172), Wood and Johnson (1996, pp. 260–262) or Steiger (2006).

¹⁷⁵Cf. Smith et al. (1992), Bertsch (2000b) or Sch¨uller (2003, pp. 24–25). Specifically, Freight SABRE was established in the 1970s by American Airlines as the first computer reservation system for air cargo shipments, see Arndt and Kinkeldei (1978, pp. 69–70).

¹⁷⁶Cf. Bey (1972, pp. 2–3), Arndt and Kinkeldei (1978, pp. 66–72), Ashford et al. (1997, pp. 305– 307), Graf and Leßmann (2000), Bertsch (2000b) or Sch¨ ¨ uller (2003, pp. 25–27). Examples are the "London Airport Cargo Electronic Data Processing System" (LACES) at London Heathrow and the "Systeme d'ordinateur du Fr ` et International A ˆ erien" (SOFIA) at Paris. ´

a subsystem of the "Automatisiertes Tarif- und Lokales Zollabwicklungsssystem" (ATLAS).¹⁷⁷

In the 1990s, IATA-CASS was established as a clearing system for air freight operations between IATA member airlines and their IATA cargo agents backed by the IATA-Resolution 851.¹⁷⁸ Furthermore, 'Tracking and Tracing Online' (TRAXON) was founded in 1991 as a neutral CCS serving carriers, freight forwarders as well as other participants in the air cargo logistic chain with a fully integrated information technology-driven communication system.¹⁷⁹ Recently, three neutral, air cargo focused computer reservation systems, namely "Global Freight Exchange" (GF-X), "Cargo Portal System" (CPS) and "Ezycargo," started to provide online booking of air cargo shipments.¹⁸⁰

Today, the air cargo industry can be divided in several segments referring to airline's fleet structure and their commitment to air cargo business as shown in Fig. 2.6. 181

First, there are cargo and non-cargo focused passenger airlines, marketing only their lower deck capacity called 'belly' merely as a by-product of their passenger transport operations.¹⁸² The next group are combination airlines, whose fleets are a mix of passenger aircraft and pure freighters. In addition to this, some of them own hybrid forms like "combi carriers" with a movable bulkhead dividing the main deck in a passenger and a cargo section or aircrafts with easily removable seats called "quick change," allowing carriage of passengers by day and cargo at night.183

The other three groups of airlines are mainly employing pure freighters.¹⁸⁴ Integrators offer time definite door-to-door services and are mainly focused on small shipments like documents and parcel size consignments up to a weight of 30 kg, but some of them recently broadened their operations and allow now

¹⁷⁷Cf. Graf and Leßmann (2000) or Kummer et al. (2010, pp. 222–224). ¨

¹⁷⁸See Grandjot (2002, pp. 53–54), IATA (2004, pp. 29–31), Kummer et al. (2010, pp. 338–339) and http://www1.iata.org/cargo/distribution/cass/index.htm.

¹⁷⁹See Damsgaard (1999), Graf and Leßmann (2000), Bertsch (2000b), Grandjot (2002, pp. 162– ¨ 163), Sch¨uller (2003, p. 26), Kummer et al. (2010, pp. 339–340) or http://www.traxon.com.

¹⁸⁰See Grandjot (2002, pp. 170–171) or Kummer et al. (2010, pp. 337–338) with further references.

¹⁸¹See DVB (2001, pp. 9–14) or Grandjot (2002, pp. 97–103). A similar segmentation of the air cargo industry in pure freighter airlines, combination carrier and integrators can be found in Eaton (1994), Wells (1999, pp. 369–371), Albers (2000, pp. 19–23) or HSH Nordbank (2007, pp. 17–25). 182Cf. Eaton (1994), Shaw (1999, pp. 138–140), Grandjot (2002, pp. 9–10), Sch¨uller (2003, p. 11) or HSH Nordbank (2007, pp. 23–32).

¹⁸³Cf. Wood et al. (1995, p. 141), Opgenhoff (1997, p. 10), Shaw (1999, pp. 140–141), DVB (2001, p. 15), Grandjot (2002, p. 10), Vahrenkamp (2002, p. 14) or Sch¨uller (2003, p. 11).

¹⁸⁴See Grandjot (2002, pp. 70–81) for technical descriptions of commonly used pure freighter aircrafts. Most of these pure freighters employed in the air cargo industry are former passenger aircrafts converted into freighters, removing all passenger-related facilities, see DVB (2001, p. 15) or HSH Nordbank (2007, pp. 27–32).

Fig. 2.6 Fleet Structure and Commitment of Airlines (Source: DVB (2001))

shipments of bigger size, too.¹⁸⁵ The most prominent ones are Dalsey-Hillblom-Lynn (DHL), Federal Express (FedEx), Thomas Nationwide Transport (TNT) and United Parcel Service (UPS), having together an overwhelming world market share.186 In addition to this, some freight forwarders had their own fleet of aircrafts, like Emery Worldwide or BAX Global in order to offer an integrated air cargo

¹⁸⁵Cf. Wood et al. (1995, pp. 144–147), Opgenhoff (1997, p. 23), Bess (1996, pp. 48–59), Bjelicic (2000) or DVB (2001, p. 10).

¹⁸⁶Despite many authors like Schneider (1993, pp. 50–54), Opgenhoff (1997, p. 19), Becker (1999, p. 41) or Bachmeier (1999, p. 44) dating the early beginnings of this segment back to the end 1970s, provision of courier-, express-, parcel and/or postal services itself can be traced back to the Ancient Age and some of these four big players were established long before, see Hector (1987) or Bess (1996, pp. 29–32). UPS was established in 1907 in the United States, TNT 1946 in Australia followed by DHL in 1969 and FedEx in 1974 in the United States, see Gutthal (1999, pp. 17–24), Bachmeier (1999, pp. 82–119) or Lucke et al. (2001, p. 294). With the recent development on the European postal market TNT was acquired by the Dutch The Post Groep (TPG) and DHL by the German Deutsche Post World Net (DPWN).

transport service, too.187 Another group of airlines are pure cargo airlines which operate scheduled or charter services on own account, employing their pure freighter fleet like Cargolux, Nippon Cargo Airlines and Kalitta Air.¹⁸⁸ The last group of airlines are mainly focused on the main haul and operate as contract carriers for other airlines as well as integrators and forwarders on a long-term charter basis.¹⁸⁹ Of course, some airlines belong to several groups and therefore the lines between one group and another are rather blurred.

Another aspect in air cargo operations are surface transportation of air cargo shipments between two airports, called "road-feeder-service" or "trucking" backed by IATA-Resolution 507b.¹⁹⁰ Some reasons for this ever increasing partly substitution of air cargo transport by surface operations are (1) insufficient air cargo capacity on flight destinations from/to smaller airports, (2) pooling of air cargo at central hubs by airlines as well as freight forwarders or (3) a general restriction of night flight operations on many airports.¹⁹¹

2.2.4 Mapping of Inter-organizational Interaction and Cargo Movement

In the following sections first, traditional forms of sea transport operations with conventional general cargo as well as contemporary forms with containerized cargo are discussed. Then, the focus is set on contemporary air cargo transport operations.

2.2.4.1 Traditional Sea Transport Operations of General Cargo

As Martin and Thomas (2001) described in the case of break bulk berth operations in the early 1960s, following inter-organizational interaction and cargo movement can be seen in conventional (non-containerized) general cargo trades as shown in

¹⁸⁷Cf. Eaton (1994), Mosler (1999) or DVB (2001, pp. 31–32). Recently, Emery Worldwide was sold to UPS and BAX Global (former Burlington Air Express) was acquired by Deutsche Bahn and merged with their freight forwarding operations under their brand name DB Schenker.

¹⁸⁸See e.g. HSH Nordbank (2007, pp. 20–21).

¹⁸⁹Cf. DVB (2001, pp. 33–34) or HSH Nordbank (2007, p. 24). Examples are Atlas Air Worldwide Holding or Air Atlanta Icelandic.

¹⁹⁰See Arndt and Kinkeldei (1978, pp. 58–66), Mosler (1993), M¨uller-Rostin (1996), Opgenhoff (1997, p. 11), Odrich (1998:95–97), Becker (1999, pp. 55–62), Dautel (2000), Button and Stough (2000, pp. 293–295), Grandjot (2002, pp. 193–194), Sch¨uller (2003, pp. 21–23) or Bottler (2006). According to Dautel (2000), in Germany about 70% of all air cargo is being trucked.

¹⁹¹Further, there is a tendency to a full substitution of air freight operations by road haulage especially in Europe up to a distance of 1,000 km, because of transit time and cost reasons, see Windisch (1996, p. 44), Becker (1999, p. 56) or Bottler (2006).

Fig. 2.7 Interorganizational Interaction and Cargo Movements in General Cargo (Source: adapted from Martin and Thomas (2001))

Fig. 2.7. ¹⁹² The arrows indicate cargo movements, while organizations adjacent to each other have some inter-organizational business relationships.193

Beside providing basic port infrastructure, port authorities usually ensured temporary storage or warehousing of cargo at own premises as well as offering mechanized ship-to-shore cargo handling by cranes.¹⁹⁴
At that time, shipping companies mainly restricted their business operations to sea transport having responsibility for the cargo shipped starting at the port of departure crossing the ship's rail and ending at the port of arrival passing their ship's rail again.195 In the general cargo segment, they were engaged in scheduled

¹⁹²In addition to this, see Schulz-Hanßen (1965:53–70 and 142–68) for a quite similar description of seaport operations or Jansson and Shneerson (1982, pp. 9–26) who focused more on the development of handling capacity at seaports.

¹⁹³Of course, this list of companies working in and around a seaport is by far exhaustible, e.g. there are tug operators, private firms specialized in mooring vessels, ship building yards or other companies staffing, supplying and maintaining vessels, see Schulz-Hanßen (1965:142–154), Benson et al. (1994, pp. 88–96), Diederich (2000) or Jaeger and Laudel (2003, pp. 219–221).

¹⁹⁴Cf. Schlichting (1931, pp. 13–14), Martin and Thomas (2001, p. 280) with further references.

¹⁹⁵Cf. Bohme (2000b). ¨

liner trades which were subject to shipping conferences as already discussed in Sect. 2.2.2.2.

The landside business of shipping companies was usually run by shipping agents and loading brokers.¹⁹⁶ Having an agency contract with shipping companies, shipping agents cared for both ships and cargo on behalf and account of their principals on a regular basis.¹⁹⁷ According to the ships, their duty was to organize all operations needed to supply, maintain and clear seagoing vessels inwards and outwards calling their seaport.¹⁹⁸ Referring to outgoing cargo, they marketed the liner services, concluded contracts of affreightment, collected freight payments, issued and handled transport documents like B/Ls. In the case of incoming cargo, they organized delivery of shipments in concordance to the transport documents including collection of remaining freight payments and secured interests of their principals in the case of loss or damage of cargo. The actual form of relationship between a ship agent and his principal greatly varied both from company to company and from port to port with shipping companies often performing ship agency functions in-house at home ports or at ports of call with a high volume of traffic.¹⁹⁹ However, at the majority of ports of call, shipping companies normally did not run a permanent office and therefore, shipping agents were appointed. Loading brokers were usually engaged in cargo brokerage and chartering business as neutral third parties.²⁰⁰ Doing this, they matched shippers and shipping companies in order to arrange contracts of affreightment or charter parties on a non-regular basis without further involvement.
Wharfingers were small private companies licensed and subcontracted by the port authority to perform on-site movement of cargo.²⁰¹ In addition to this, wharfingers were also responsible of loading, unloading and storage of incoming

¹⁹⁶Cf. Securius and Boning (1941, pp. 64–67), Ohling (1950, pp. 222–224), Trappe (1995), ¨ Puttfarken (1997, pp. 247–249), Knudsen (1998), Herber (1999, pp. 168–171), Bohme (2000b), ¨ Jaeger and Laudel (2003, pp. 219–221) or Dischinger et al. (2005, pp. 269–271). According to Miller (2002, p. 8–11), principally three forms of shipping agents can be distinguished: (1) firms whose principal identity and work was shipping agency, (2) trading agencies running shipping agency department and (3) shipping companies owning shipping agents.

¹⁹⁷Serving this double-role, shipping agents are usually called "husbandry agent" or "ship's husband" if they care only for the vessel and "charter's agent" if they set their focus on cargo, see Puttfarken (1997, pp. 249–250).

¹⁹⁸Cf. Murr (1979, p. 84), Trappe (1995, p. 324) or Knudsen (1998, p. 155).

¹⁹⁹Cf. Martin and Thomas (2001, p. 282).

²⁰⁰Cf. Mittendorf and Oelfke (1974, pp. 228–232), Murr (1979, pp. 83–84), Puttfarken (1997, pp. 248–249), Knudsen (1998, pp. 153–154), Herber (1999, pp. 169–171), Strandenes (2000) or Fiotakis (2005).

²⁰¹Cf. Martin and Thomas (2001, p. 280) with further references. In Hamburg, warehousing of merchandizes like spices, tea, coffee, cacao, wines or carpets is done up to now by "Quartiersleute" in the former freeport area called "Speicherstadt," which is still one of the biggest warehousing complexes in the world, see Ohling (1950, pp. 183–184), Diederich (2000), Lauenroth (2001) and Jaeger and Laudel (2003, p. 220). But with the new "Hafencity" project in Hamburg, this traditional form of warehousing will vanish within a few years, see http://www.hafencity.com/en/home.html.

and outgoing cargo from inland carriers. Doing this, they acted on behalf and account of freight forwarders as well as shippers.

Master stevedores were small private companies engaged into stowing and unstowing cargo onboard of ships as well as lighters.202 They were contracted by shippers, shipping companies or through their respective agents.
Lighter operators were private barge companies licensed by the port authority to provide cargo movements by barges or lighters within the seaport area.²⁰³ Again, they were contracted by shippers, shipping companies or through their respective agents, especially in the case of bulky shipments, transhipment to inland navigation on basis of a through transport contract or simply to save handling charges raised by the port authority for ship-to-shore movements of cargo.²⁰⁴
The dominating inland carrier at that time were national railway companies.²⁰⁵ Especially at German seaports there was a heavily regulated market with special tariffs for hinterland traffic of seagoing cargo performed by the German Reichsbahn. In addition to this, small road hauliers were contracted by freight forwarders for short distance transports as well as inland navigation companies for rather bulky shipments.

In this traditional environment, freight forwarders served a distinct role acting as an agent on behalf and account of a shipper, consignor or consignee.206 Mainly, they arranged affreightment with shipping agents and/or loading brokers for the sea leg on one hand and affreightment of cargo with inland carriers by rail, road or inland navigation for hinterland transport on the other, including the organization of cargo handling operations actually performed by port authorities, stevedores, wharfingers and lighter operators.

In addition to this, they represented the interests of their principals in every aspect, including supervision of cargo handling operations, cargo inspection, arranging customs clearance, crediting duties, fees and charges payable at once. As a "loading agent" for outgoing shipments, a freight forwarder booked cargo space in advance, organized delivery of cargo to the seaport of departure, despatched shipments and handled transport documentation. As a "landing agent" for incoming

²⁰²See Fischer et al. (1930, pp. 239–240), Schlichting (1931, p. 14), Jansson and Shneerson (1982, pp. 24–26), Diederich (2000) or Martin and Thomas (2001, p. 281) with further references.

²⁰³See Martin and Thomas (2001, p. 281) with further references.

²⁰⁴Cf. Schlichting (1931, pp. 14–15).

²⁰⁵Cf. Schlichting (1931, pp. 23–31).

²⁰⁶Cf. Schlichting (1931, pp. 13–58), Seidelmann (1969, pp. 92–93), Mittendorf and Oelfke (1974, pp. 241–265), Murr (1979, p. 237–248), Schumacher (1987:148–149), Herber (1999, pp. 179– 180), Beamer-Downie (1999, pp. 13–21) or Jaeger and Laudel (2003, pp. 263–267). According to Vadnai (1921, p. 19) or Seiler (1981, p. 82–84), some freight forwarders were working as shipping agents at the same time, but this field of activity was not so common and mainly run by some subsidiary company.

shipments, he organized handling operations for unloading shipments, temporary storage, customs clearance and delivery, usually on behalf of his principal.

An additional field of activity at this time was consolidation of small shipments, where a freight forwarder as a "cargo consolidator" collected shipments from several shippers in order to ship them under a single contract of affreightment and split them for delivery afterwards. Doing this, considerable lower freight rates and therefore a high bargain were reached due to usually high minimum rates and charges raised by shipping companies. But these cargo consolidation operations needed at least a correspondent firm at the port of arrival who acted on behalf of the cargo consolidator at the port of departure.

All in all, this break bulk berth system of conventional general cargo was a strictly divided, step-by-step approach to cargo handling at seaports, with:

Freight forwarders and inland carriers being responsible for hinterland operations.
Wharfingers, lighter operators and the port authority as a ship-to-shore-crane operator taking care for on-site and ship-to-shore movement of cargo.
Master stevedores engaged in stowage and unstowage of cargo under supervision of shipping companies' officials onboard their ships.²⁰⁷

Although a clear interdependence existing between all the organizations in several stages of cargo movement at a seaport, they did not work together. Only little communication and a lack of co-ordination occurred, so that cargo handling rates and performance of these break bulk berth systems for general cargo movements were considerably low.²⁰⁸

2.2.4.2 Contemporary Sea Transport Operations of Containerized Cargo

With the advent of containerization, organization of cargo movement as well as inter-organizational relationships at seaports changed dramatically, introducing the container terminal system as shown in Fig. 2.8: 209

²⁰⁷Again, this supervision could be outsourced to agents like shipping agents or tallymen, who were usually engaged on behalf and account of a shipping company to inspect and measure outgoing and incoming shipments, see Ohling (1950, pp. 224–233), Jaeger and Laudel (2003:240), Diederich (2000) or ITZ (2002).

²⁰⁸Cf. Svendsen (1958, pp. 94–102), Baumoller (1970, pp. 62–71), Slack (1998, p. 264) or Martin ¨ and Thomas (2001, p. 282–283).

²⁰⁹Cf. Jansson and Shneerson (1982, pp. 9–26), Slack (1998, pp. 273–276), Pumpe (2000, pp. 29– 44), Juhel (2001), Robinson (2002), Steenken et al. (2004) or Vanelslander (2005, pp. 126–140). According to Martin and Thomas (2001, p. 283) "[t]he name 'terminal' has been introduced, implying an integrated activity providing services not just to the shipping line but also to importers, exporters and transport operators. The name 'terminal' is now applied to facilities that provide a comprehensive range of specialized services to users." Following them, this kind of terminal system is nowadays also valid for other segments of seagoing cargo. But a detailed discussion

Fig. 2.8 Interorganizational Interaction and Cargo Movement in container trades (Source: Adapted from Martin and Thomas (2001))

• Terminal operators today control all ship-to-shore handling operations as well as temporary storage and other container handling operations on their container yards. They integrated virtually all cargo handling operations done before by wharfingers, master stevedores as well as other cargo related handling operations formerly organized by the port authorities.²¹⁰ In order to secure smooth cargo movement at seaports, terminal operators had to completely redesign existing dock structures, invest heavily in suprastructure (e.g. specialized container

about differences between handling of containerized and non-containerized cargo at seaports is beyond of the scope of this work.

²¹⁰See Herber (1999, pp. 173–175), Breitzmann (2000), Pumpe (2000, p. 39), Stopford (2000, pp. 29–32), Peters (2001, pp. 17–25), Martin and Thomas (2001, p. 287), Notteboom and Winkelmans (2001, p. 84), Juhel (2001), Notteboom (2002, p. 263), Miller (2002, p. 18) or Brooks (2004). The role of port authorities nowadays often resembles more than that of a landlord, just providing infrastructure and granting concessions to one or more terminal operators if they are not stakeholders or solemn owners of terminal operations on their premises. According to Baird (2002, p. 275), in 36% of the seaports responding to a survey of the International Association of Ports and Harbours (IAPH), container terminals are run by private companies and in 34% they are operated by the port authority. Further, 62% of the seaports responding offer information services and 63% warehousing as value added services.

handling facilities) and information processing technology.211 Some of them are working now as dedicated container terminals, serving only one shipping company, consortia or strategic alliance on basis of an exclusive agreement.212 Furthermore, they diversified their field of activity to additional services like storing, cleaning, maintaining, stuffing and/or stripping of containers and even engaged in hinterland transports or established inland terminal operations.213 During the last decade, some terminal operators merged or extended their business considerably by taking over and/or setting up new terminals in other ports in order to offer merely a network of terminals than just to market a single site.²¹⁴

• Shipping companies extended their field of operation to cargo handling at sea ports and/or inland operations.215 Some of them are running their own shipping agencies, at least through partly owned subsidiary companies in order to have more control on the marketing of their services.216 Furthermore, they are owning a considerable amount of containers shipped and organize hinterland traffic on own account called carrier haulage and then act as a carrier MTO in door-to-door services.217 Moreover, they have dedicated terminals, some of them even run their own container terminals.218 In addition to this, shipping

²¹¹Cf. Rackwitz (1969, pp. 52–63), Slack et al. (1996, pp. 297–298), Talley (2000, pp. 940–944), Bohme (2000:66–67) or Heaver (2002b, pp. 388–389). ¨

²¹²Cf. Evangelista and Morvillo (1999, pp. 32–34), Heaver et al. (2001, p. 298), Notteboom and Winkelmans (2001, p. 74), Haralambides et al. (2002) or Midoro et al. (2005).

²¹³Cf. Notteboom (2002, pp. 259–260), McCalla et al. (2004) or Debrie and Gouvernal (2006).

²¹⁴Cf. Peters (2001, pp. 17–25), Notteboom and Winkelmans (2001, pp. 78–83), Heaver et al. (2001, pp. 300–302), Notteboom (2002, p. 260–262), Woodbridge (2002), DVB (2002, pp. 74– 75), Midoro et al. (2005), Vanelslander (2005, pp. 20–75), Slack and Fremont (2005), Bichou ´ and Bell (2007), Olivier et al. (2007), Notteboom (2007) or Kummer et al. (2010, pp. 297–298). Prominent examples of this ongoing expansion of terminal operators are Hutchison Port Holding (HPH) from Hongkong, Port of Singapore Authority Corp. (PSA), European Combined Terminals (ECT) of Rotterdam, or the German joint venture of Bremer Lagerhaus-Gesellschaft (BLG) and Eurokai called Eurogate.

²¹⁵Cf. Casson (1986, pp. 23–27), Kindleberger (1993, pp. 173–174), Evangelista and Morvillo (1999, pp. 32–34), Talley (2000, pp. 938–939), Heaver et al. (2000, pp. 365–367), Notteboom and Winkelmans (2001, pp. 74–75), Heaver (2002a, pp. 218–224, 2002b, p. 385), McCalla et al. (2004) Debrie and Gouvernal (2006), Notteboom and Merckx (2006) or Fremont (2009). ´

²¹⁶Cf. Casson (1986, pp. 22–27), Knudsen (1998, pp. 151–152), Heaver et al. (2001, p. 303), Notteboom and Winkelmans (2001, pp. 74–75), Stapleton and Hanna (2002, p. 6) or Miller (2002, pp. 18–19), McCalla et al. (2004), Notteboom and Merckx (2006) or Fremont (2009). ´

²¹⁷According to Branch (2000, p. 68) or DVB (2002, p. 72), about 50% of all containers are owned by shipping companies or other freight carriers and the residue are shipper-owned or leased. Following Notteboom and Winkelmans (2001, pp. 75–76), the average share of carrier haulage in European hinterland traffic is about 30% with a maximum of 70% in the United Kingdom and a minimum of 10% for Switzerland.

²¹⁸Especially the container shipping companies Mearsk Line (APM), COSCO, Evergreen und MSC are engaged in terminal operations, see Heaver et al. (2001, p. 302), Heaver (2002a, 2002b, pp. 383–385), Midoro et al. (2005) or Notteboom (2007).

companies offer their services directly to shippers with high shipment volumes, bypassing all intermediary firms in between.²¹⁹ This is especially valid for comprehensive door-to-door services as well as third-party logistics packages, often performed by fully-owned subsidiary forwarding companies.²²⁰ All in all, shipping companies seem to have a very strong position in the container shipping industry.

Nowadays, shipping agents and loading brokers offering traditional services to ships and cargo as already described in Sect. 2.2.4.1 seem to play a rather diminishing role in the contemporary container shipping industry.221 Most of them vanished if they were not acquired by shipping companies or diversified their traditional operations of shipping agency and brokerage to third-party ship management and/or NVO services on the main haul.222 As a NVO, they are providing scheduled shipping services without operating own ships by buying in cargo space or container slots under flat rates and reselling them to shippers.
Feeder operators are connecting main ports or load centres like Hamburg, Bremen/Bremerhaven, Rotterdam or Antwerp on the north range in Europe with smaller ports in Scandinavia, the Baltic States or Russia not called by deepsea containerships.²²³ Moreover, some feeder operators connect inland container depots like Duisburg with these main ports or load centres by scheduled barge services over the River Rhine.²²⁴ These feeder operations are often done on a through transport basis.
Inland carriers like road hauliers nowadays act either as agents in merchant haulage for a shipper or a freight forwarder or as agents for a shipping company itself in the case of carrier haulage.²²⁵ In addition to this, several railway operators are running scheduled block train services to inland container depots or inland terminals for intermodal transports by railway. These railway operators

²¹⁹See Seidelmann (1969, p. 94), Burg (1975, pp. 135–136), Francesetti Cazzaniga and Foschi (2001:17–18) or Fremont (2009). Notably, Davies (1981) reports about forward integrating ´ exporters directly negotiating with carriers in the case of United Kingdom leading to the same result.

²²⁰Cf. Branch (1998, pp. 188–192), Thorby (2001b), Beddow (2001), Heaver (2002a, pp. 222–224, 2002b, pp. 385–388), Oldenburg (2002), McCalla et al. (2004), Notteboom and Merckx (2006) or Baird (2006) or Fremont (2009). ´

²²¹Cf. Knudsen (1998, pp. 150–154), Eller (2001), Miller (2002, pp. 15–22) or Fiotakis (2005).

²²²See Knudsen (1998, pp. 152–153), Bohme (2000b) or Eller (2001). In the container shipping ¨ industry, the classical chartering business is concentrated in Hamburg with a market share of about 75%, see Knudsen (1998) or Holtappels (2006). Further Branch (1998, pp. 293–302), Panayides (1999), Panayides and Cullinane (2002) or Mitroussi (2003, 2004) for a description of third party ship management.

²²³Cf. Ordemann (1996, pp. 89–93), Neuhof (2000), Stopford (2000, p. 375).

²²⁴Cf. Ordemann (1996, pp. 88–89).

²²⁵According to Notteboom (2002, p. 259), at least two terminal operators (HPH in Felixstowe and ECT in Rotterdam) own road haulier companies, but it seems to be merely unusual for terminal operators to run road transport operations. Moreover, road hauliers are often small private companies who provide simple for-hire trucking services.

are subject to several ownership structures, often in form of a joint venture with railway companies, terminal operators, shipping companies and/or interest groups of freight forwarders as stakeholders.226

Despite this more and more blurring image of ownership structures, co-operation and fields of activities as well as an ongoing process to higher concentration in the container shipping industry, freight forwarders still seem to have a steady source of cargo in the case of consolidated shipments and play an important middleman role between shippers and carriers as described in Sect. 2.2.4.1. ²²⁷ But they have to improve and diversify their range of services offered to vertically integrated transport operations and add third-party logistics services backed by IT-driven operations in order to allow shippers a one-stop shopping.²²⁸ With more and more shippers asking for door-to-door shipments or forwarding at fixed costs, some freight forwarders even entered the ship brokerage business acting as a NVO on the main haul or providing door-to-door service as a MTO.²²⁹ The freight forwarder issues then own transport documents like a FBL as a contracting carrier.

In order to give a full picture of contemporary container transport operations, different forms of pre- and onward-carriage arrangements have be taken into account, too.²³⁰ First, as indicated already before, pre- and onward-carriage of container shipments can be organized either by the shipping company himself (called "carrier haulage") or someone else (so called "merchant haulage"), including consignor, consignee or freight forwarder.231 Second, upon the place of loading (so called "stuffing") and unloading (so called "stripping") of a seagoing container, two different forms, namely "less-than-container-load" (LCL) and "full-container load" (FCL) adding up to four different combinations of pre- and onward-carriage can be distinguished as shown in Fig. 2.9:

• If shipments are stuffed and stripped directly at the container terminal or some container packing station nearby, this container shipment is called "LCL/LCL,"

²²⁶See Heaver et al. (2000, pp. 365–366), Notteboom (2002, pp. 41–43), Lennarz (2001), Heaver (2002a, p. 220–222) or Kummer et al. (2010, pp. 325–328) with further references.

²²⁷Cf. Rackwitz (1969, pp. 77–82), Seidelmann (1969, pp. 93–95), Seidenfus (1973, pp. 4–5), Heaver et al. (2000, p. 369), Cazzaniga Francesetti and Foschi (2001, pp. 18–19) or Lorenz (2005, pp. 492–494). DSLV (2005, p. 17) estimates that about 75% of general cargo leaving Germany are despatched by freight forwarding companies.

²²⁸Cf. Dehn (1987, pp. 15–17), Seidelmann (1969, pp. 94–95) or Heaver (2002b, pp. 380–382).

²²⁹Cf. Seidelmann (1969, pp. 94–95), Seiler (1981, pp. 82–84), Geuther (1990, pp. 36–37) or Wolf (2000b, 2000c). Further, see Herber (1999, pp. 243–244), Nielsen (2000) or Kehagiaras (2004) for a more comprehensive treatment of legal issues.

²³⁰See Ordemann (1996, pp. 103–105), Jaeger and Laudel (2003, pp. 204–206), Biebig et al. (2004, pp. 272–286) or Lorenz (2005, pp. 465–470) for the following and a more elaborated description, including cost and freight issues in container transport operations.

²³¹In addition to this, there can occur some mixed arrangements, where e.g. the carrier forwards an empty container to the consignor with the next steps organized in form of merchant haulage, see Ordemann (1996, p. 103–104).

Fig. 2.9 Forms of container traffic (Source: Kummer et al. (2010, p. 294))

indicating a solely port-to-port container transport. An example are shipments of freight forwarders in groupage service.

Next, "LCL/FCL" are called port-to-door container transports with a container stuffed at the port of departure before the cargo is shipped. At the port of destination, the container is directly delivered to the premises of the consignor. This combination is occurring, if a freight forwarder collects small amounts of cargo in favour of one consignor and then ships them to him in one container.
Door-to-port container transports are called "FCL/LCL," where shipments originating from one consignor are stuffed in a container for different consignees at the consignors' premises and therefore have to be stripped and split off at the port of destination before delivery.
A door-to-door shipment by container is usually abbreviated by "FCL/FCL," because the container content is normally untouched on the whole distance from the consignor's to the consignee's premises.

The actual modal split of pre- and onward-carriage of cargo summing up to the hinterland traffic of a seaport heavily depends on geographic location and proximity to potential shippers.²³² In Hamburg, a lot of short distance road haulage is done due to a high amount of local consignors or consignees, while long distance

²³²Cf. Ordemann (1996, pp. 84–87) or Notteboom (2002, pp. 32–36).

Fig. 2.10 Cargo and documentation flows in sea transport operations. (Source: Kummer et al. (2010, p. 276))

hinterland traffic is mainly served by scheduled block train services. The seaports of Bremen and Bremerhaven have not so many local sources of cargo, so most of the hinterland traffic is done by railway operators to more distant locations. Due to geographic reasons, road haulage is the dominant mode of pre- and onwardcarriage of containers in the United Kingdom, whereas in Rotterdam and Antwerp, the most prominent modes in hinterland traffic are road haulage followed by barge operations.

Before switching to air transport operations, it is worth to have a closer look at the cargo and documentation flow for seagoing cargo as shown in Fig. 2.10 in order to clarify the freight forwarder's function in the documentary flow of seagoing shipments in connection with B/Ls.²³³ First, the unimodal transport case is briefly described, before the through transport concept and multimodal transport chains are discussed.

In an unimodal port-to-port transport, a freight forwarder usually works as the shipper's agent in despatching outgoing shipments, normally based on a forwarding contract. Alternatively, a freight forwarder himself is the consignor of a shipment like in the case of cargo consolidated by him and shipped on own account. In both instances, he usually concludes a contract of affreightment with a shipping agent, a loading broker or a NVO and gets a booking note fixing this agreement as a document of evidence. Principally, the following delivery of cargo at the port of

²³³See Herber (1999, pp. 242–249), Jaeger and Laudel (2003, pp. 230–208) or Lorenz (2005, pp. 471–481) for a more detailed treatment.

departure can be done to a terminal operator or directly to the berth alongside a ship. Upon delivery of cargo, the freight forwarder as a shipper gets either a quay receipt or a mate's receipt, on which basis a B/L is issued by the terminal operator, the shipping company or his shipping agency later on. Sometimes, these dock receipts or mate's receipts are not issued in favour of a B/L at once and offered to the shipper. According to a speciality in German as well as Austrian national maritime law, the term 'shipper' in this context is called "Ablader" and refers to another third party not participating necessarily in the contract of affreightment.²³⁴ Following - 642 HGB or - 642 UGB, he has the right to ask for a B/L according to specifications stated by him after the shipment is stowed onboard a ship (called "Shipped-Onboard Bill of Lading") or received for shipment at the port of departure (called "Received-for-Shipment Bill of Lading"). Being a document of title, these B/Ls issued represent the shipment in transit and can therefore be used as a vehicle to transfer title from one person to another.²³⁵ At the port of destination, any beneficiary person can then release the shipment as described in the B/L either directly from the shipping company or some terminal operator. As often this function as a document of title is not needed any more like in the case of pure intra-firm trade, more and more sea waybills (SWBs) are issued being non-negotiable documents of evidence.²³⁶

If seagoing cargo is shipped on a door-to-door basis, principally three different approaches in terms of contractual scope and liability can be followed. First, a series of unimodal transport contracts can be closed e.g. by a freight forwarder on behalf of shipper for each phase and mode of transport following one after another in a transport chain.²³⁷ At each transhipment point, shipments are then re-expedited under a new contract of carriage. Having a separate unimodal transport contract in each of these subsequent phases, liability of each performing carrier follows its mode-specific legal provisions. Second, a through transport contract can be closed between a shipper and a contracting carrier, where the contracting carrier enters into subsequent contracts of carriage with each performing carrier, if he does not transport the shipment by own means.238 When doing this, usually a "Through Transport Bill of Lading" is issued by the contracting carrier on the main haul, covering the whole transport chain from source to sink.239 By issuing such a through transport document, he undertakes responsibility for the shipment only when he

²³⁴Cf. Trappe (1995, pp. 322–324), Herber (1999, pp. 247–249), Jaeger and Laudel (2003, pp. 225– 226) or Lorenz (2005, p. 450).

²³⁵See e.g. Jaeger and Laudel (2003, pp. 236–238).

²³⁶See Puttfarken (1997, p. 172), Jaeger and Laudel (2003, p. 256), Lorenz (2005, p. 479) or Kummer et al. (2010, p. 278).

²³⁷Cf. M¨uller-Feldhammer (1996, pp. 8–9), Puttfarken (1997, pp. 174–175) or Clarke (2002).

²³⁸Cf. Schlichting (1931, pp. 61–64), Mayer (1933, p. 32), M¨uller-Feldhammer (1996, pp. 7–8), Faber (1996, p. 503–504) or Puttfarken (1997, pp. 173–179).

²³⁹See M¨uller-Feldhammer (1996, pp. 262–266), Wolf (2000b) or Lammich (2000). According to M¨uller-Feldhammer (1994, p. 272), this approach was a common practice up to the 1960s.

has actual control on it or he faults in making arrangements with other performing carriers. In the case of loss, damage or delay of cargo, the owner of a shipment has then always to identify at which stage it occurred in order to address his claim directly to the actual performing carrier.²⁴⁰ The third approach is a multimodal transport contract, where a MTO takes full responsibility for the entire transport chain from source to sink employing different means and modes of transport.²⁴¹ The transport document issued by the contracting carrier is then a "Multimodal Transport Bill of Lading" like the FBL from FIATA in the case of a freight forwarder.²⁴² Depending on the GT&Cs underlying such a multimodal contract, at least two different liability regimes can be distinguished.²⁴³ First, a uniform liability system may be established, where in every stage of a multimodal transport chain the same provisions are imposed, irrespective of the mode of transport actually employed.244 Another approach to liability under a multimodal transport contract is the network liability system, where either existing mode-specific liability provisions are applied if the place where delay, loss or damage of a shipment occurred is known or a uniform liability is imposed if it is virtually unknown.

2.2.4.3 Contemporary Air Cargo Transport Operations

Contemporary air cargo transport operations can be divided into three distinct market segments:²⁴⁵

Traditional air cargo logistic chains, with a close co-operation between freight forwarders, airlines and their respective cargo agents mainly within the IATA agency system.
A charter market for air cargo operations usually performed outside the IATA agency system.
Integrated air cargo logistic chains run by one single operator, usually called "integrator."

In traditional air cargo logistic chains, there is a clear distinction drawn between surface and air transport operations backed by the IATA agency system of IATA

²⁴⁰If not a directly responsible performing carrier can be found, then the liability is determined by the applicable transport law of the contracting carrier issuing the through transport document. As M¨uller-Feldhammer (1994, p. 272) notes, this approach was rather insufficient.

²⁴¹Cf. M¨uller-Feldhammer (1996, pp. 10–22), Faber (1996, pp. 503–504), Slack (1998, pp. 276– 277), Wolf (2000c), Spera (2002, pp. 256–261) or Jaeger and Laudel (2003, p. 235).

²⁴²Cf. M¨uller-Feldhammer (1996, pp. 267–274), Beamer-Downie (1999, pp. 75–82), Wolf (2000a, 2000c), Lammich (2000) or Jaeger and Laudel (2003, p. 235).

²⁴³Cf. Faber (1996, pp. 507–508) or M¨uller-Feldhammer (1996, pp. 275–279).

²⁴⁴Such an uniform system of liability was first proposed in the "UN Convention on International Multimodal Transport of Goods" of 1980.

²⁴⁵Cf. Gompf (1994, pp. 20–43) or Kummer et al. (2010, pp. 307–316).

approved cargo agents.²⁴⁶ Freight forwarders usually hold contacts to shippers, organize pre- and onward-carriage and despatch air cargo shipments to IATA cargo agents whereas airlines restrict their operations on the main haul by air.²⁴⁷

Apart from this institutionalized system of the traditional air cargo logistic chain within the IATA agency system, there exists a considerable huge charter market for air cargo capacity, too. One segment of this charter market are full charter and split charter of pure freighters in non-scheduled flight operations, often employed by freight forwarders.²⁴⁸ This charter market is dominated by non-IATA airlines, because IATA airlines usually restrict themselves not to perform too much split charter operations in order not to cannibalize their scheduled flight operations. With an ever rising demand for air cargo space, today these non-scheduled charters are performed on a regular basis and resemble therefore to scheduled ones. The other segment are highly specialized air transport operations outside the IATA agency system dominated by air freight brokers or general sales agents, like ad hoc charter of aircrafts and air transport of heavy-lift shipments.

Offering fast and reliable time definite services for small shipments, integrators follow the strategy of controlling virtually all stages of the logistic chain.²⁴⁹ They collect shipments at pre-carriage, organize main haulage via own hubs and deliver it at pre-carriage as a fully integrated door-to-door service run by own means. But today, they are also important buyers of air cargo capacity on scheduled flights of passenger, combination and cargo airlines, especially on destinations, where volumes do not justify employment of their own aircrafts.250

Looking at the inter-organizational interaction and cargo movement at airports as shown in Fig. 2.11: 251

• Airport authorities are usually state- or municipal-owned institutions, but recently a lot of them got at least partly privatized.252 At minimum, they provide traffic management for all flight operations, onside infrastructure and basic services for transhipment of passengers, their luggage, mail and freight.253

²⁴⁶Cf. Gompf (1994, pp. 21–31), Opgenhoff (1997, pp. 7–8), Vahrenkamp (2002, p. 16) or Sch¨uller (2003, pp. 16–20).

²⁴⁷Wood et al. (1995, p. 146) comments this co-operation with "If direct airlines are the wholesalers of space, forwarders are the retailers. They are indirect air carriers."

²⁴⁸Cf. Gompf (1994, pp. 31–37) or Lorenz (2005, p. 536).

²⁴⁹Cf. Gompf (1994, pp. 37–43), Opgenhoff (1997, pp. 20–23) or DVB (2001, pp. 15–16).

²⁵⁰Cf. Opgenhoff (1997, p. 24) or DVB (2001, p. 14). On the other hand, they market their excess capacity to other airlines in order to improve their load factors.

²⁵¹Cf. Rossger (1960, pp. 16–18), Windisch (1996, pp. 22–28), Gompf (1994, pp. 44–58), ¨ Opgenhoff (1997, pp. 9–27), Becker (1999, pp. 35–39), Albers (2000, pp. 23–26) or Sch¨uller (2003, pp. 27–35). Further, see Bey (1972, pp. 17–22), Arndt and Kinkeldei (1978, pp. 4–18), Smith (1974, pp. 158–178), Ashford et al. (1997, pp. 286–293) or Grandjot (2002, pp. 130–138) with special reference to documentation flow and physical handling of air cargo shipments at airports.

²⁵²Cf. Graham (2002, pp. 9–52).

²⁵³Cf. Gompf (1994, pp. 114–115), Bender (2000), Grandjot (2002, pp. 128–129) or Maurer (2003, pp. 97–100).

Fig. 2.11 Interorganizational interaction and cargo movements at Airports

Ground handler or ground handling agents are performing all kinds of handling operations at an airport, like despatch and onside transport of passengers, luggage, mail and freight as well as all ramp services needed for supply or maintenance of aircrafts.²⁵⁴ Their ownership structure is quite diverse: Some are airport-owned, airline-owned or independent third-party service providers. For a long time, at most European airports these handling services were offered only by one to two ground handlers at each airport in addition to ground handling operations focused on forefield done by incumbent airlines. But with Regulation (EC) 67/1996, ground handling services had to be liberalized, so that today this market segment is quite competitive with specialized third-party service providers like Menzies World Cargo, Swissport Cargo Services Ltd. or Globeground-Servisair operating on several airports at the same time.²⁵⁵
Airlines are the carriers on the main haul, employing belly capacity of passenger aircrafts as well as pure freighters. Usually, they are not engaged in surface related operations, but some airlines today want to play a more active role in

²⁵⁴See Grandjot (2002, p. 129). Further Ashford et al. (1997, pp. 159–184), Kunz (1999), Pflugbeil (2000), Maurer (2003, pp. 97–100) or Templin (2007, pp. 21–58) for a more elaborated overview. 255Cf. Graham (2002, pp. 116–118), Maurer (2003, pp. 19–21), Wieske-Hartz (2004, p. 43), Conway (2005a) or Templin (2007, pp. 58–94).

the air cargo logistic chain.²⁵⁶ Doing this, they arrange pre- and/or onwardcarriage of shipments (usually called airline trucking), despatch at the airports on own account or through their respective agents and market this directly to shippers. Another aspect of surface operations are extensive road-feeder-services run by several airlines in Europe in order to feed their main hubs with cargo as discussed in Sect. 2.2.3.2.

IATA cargo agents handle air cargo shipments as commission agents on behalf and account of the IATA-airlines based on an agency contract between the agent and the IATA in concordance to the IATA-Resolution 805r (or 805re in the special case of the European Common Aviation Area (ECAA)).²⁵⁷ They are allowed to book cargo space, prepare air cargo shipments and documentation "ready for carriage," organize insurance coverage upon request and issue air waybills (AWBs) on behalf of the airlines. For this agency service, they are paid a fixed percentage on freight and ancillary costs of these air cargo shipments to attract traffic and to reimburse them for services offered. Most of these IATA cargo agents are at the same time freight forwarding companies specialized in air freight operations, so that many authors unfortunately draw no clear distinction between them.²⁵⁸ Moreover, as in the new European Air Cargo Programme (EACP) freight forwarders and other forms of cargo agents working under IATA Resolution 805re are subsumed under the label "IATA cargo intermediaries," the differences seem to blur more and more.²⁵⁹ But for the purpose of this work, freight forwarding and IATA cargo agency operations are clearly separated, because not every freight forwarder engaged in air cargo operations has an IATA cargo intermediary status.
At least two other intermediaries can be mentioned. First, air charter brokers match supply and demand on air charter markets as independent middlemen.²⁶⁰ On the demand side there are shippers or freight forwarders as their agents and

²⁵⁶Cf. Mosler (1993), Gompf (1994, pp. 52–53), Opgenhoff (1997, pp. 17–18) or Grandjot (2002, p. 129–130).

²⁵⁷Cf. Rossger (1960, pp. 18–20), M ¨ ittendorf and Oelfke (1974, pp. 283–287), Weisskopf (1984, pp. 130–35), Gompf (1994, pp. 47–49), Wood et al. (1995, p. 150), Becker (1999, pp. 38–39), Trommer (2000a) or Lorenz (2005, pp. 532–533). Recognizing the changing needs and circumstances in the ECAA, IATA started in 2004 a special European Air Cargo Programme (EACP) in cooperation with IATA member airlines and intermediary representative organizations, see IATA (2004) for further details.

²⁵⁸Cf. Rossger (1960, pp. 17–18), Dehn (1987, pp. 19–20), Windisch (1996, pp. 22–23), Becker ¨ (1999, p. 38–39), Albers (2000, p. 25), Grandjot (2002, p. 128), Bowen and Leinbach (2004) or Neiberger (2007, 2008).

²⁵⁹According to IATA-Resolution 823: "'AGENT' (sometimes referred to as 'IATA Cargo Agent') means a legal person which is a registered IATA Cargo Agent whose name is entered on the Cargo Agency List, having executed an IATA Cargo Agency Agreement having been adjudged to have met the registration and retention criteria as specified in the Cargo Agency Rules. This term also includes European Air Cargo Programme Intermediaries who conduct transactions in accordance with Part 1 of the European Air Cargo Programme form of Cargo Intermediary Agreement."

²⁶⁰Cf. Mittendorf and Oelfke (1974, pp. 280–282), Odrich (1998, pp. 93–95) or Trommer (2000b).

on the supply side airlines. They are usually employed if there is a need for an ad hoc charter agreement. Second, there are numerous general sales agents (GSAs), who run agency services for one or more airlines within a defined territory.261 This allows airlines to achieve cost-effective presence in markets where it is uneconomic for them to maintain its own sales force and premises. Recently, the IATA offered them the chance to apply for certification as a general sales and service agent (GSSA) based on IATA-Resolution 871 as most GSAs are not longer simple sales offices but offer a wide range of on-site services to the airlines.

Apart from the IATA cargo agents, freight forwarders are the most important partners for airlines in the traditional air cargo logistic chain.²⁶² In this environment, a freight forwarder acts as an independent company either as the agent of a shipper or as a cargo consolidator in groupage service on own account.²⁶³ As already indicated above they are quite often at the same time approved IATA cargo agents, working on commission basis for IATA airlines.²⁶⁴ As the agent of a shipper, the freight forwarder books cargo space on aircrafts, organizes delivery of cargo to the airport of departure, despatches shipments and handles transport documentation. In the case of consolidated cargo, he usually acts on own account, collects smaller shipments from different shippers and issues for each of them its own AWB called "House AWB" as a contracting carrier.265 Then he forwards the consolidated shipments as a shipper under his name with one AWB called "Master AWB." At the airport of destination, an own subsidiary or a correspondent breaks up this consolidated cargo for delivery. The difference between the freight rate offered by the airline to the forwarder and the forwarder's own tariff rate to the shipper is then the forwarder's gross yield from such cargo consolidation operation, so that most freight forwarders are engaged in consolidated air cargo shipments.²⁶⁶ To support this cargo consolidation function, there exist horizontal alliances of small and medium sized freight forwarders like Future, Challenge Air Cargo Experts, German Airfreight Partners (GAP), Interessengemeinschaft Luftfracht GmbH (IGLU) or

²⁶¹Cf. Seiler (1981, p. 82), Trommer (2000b) or Mosler (2002).

²⁶²According to estimates by DSLV (2005, pp. 17–19), 98% of all air cargo departuring Germany is despatched and delivered "ready for carriage" by freight forwarding companies with an overwhelming share being consolidated shipments. Further, 75% of air cargo in import is handled by them, too.

²⁶³Cf. Rosenthal (1950, pp. 438–439), Smith (1974, pp. 178–184), Weisskopf (1984, pp. 132– 135), Schumacher (1987, pp. 149–150), Gran (1996), Gompf (1994, pp. 48–49), Bachmeier (1999, pp. 69–78), Trommer (2000c), Grandjot (2002, pp. 99–101), Sch¨uller (2003, pp. 29–31) or Lorenz (2005, p. 533–535).

²⁶⁴Cf. Dehn (1987, pp. 19–20) or Gompf (1994, pp. 46–49). So they serve a dual role, being an agent for both shipper and airline, see Rossger (1960, p. 20) or Gompf (1994, pp. 113–114) with ¨ further comments.

²⁶⁵ Cf. Gran (1996), Becker (1999, pp. 33–35), Trommer (2000a, 2000c), Branch (2000, pp. 133– 135) or Lorenz (2005, pp. 534–535).

²⁶⁶Cf. Smith (1974, pp. 179–180) or Becker (1999, pp. 50–51).

World Cargo Organization (WACO) and Worldwide Independent Network (WIN) to name a few.²⁶⁷ The main objects of these freight forwarder alliances is to pool their air cargo volumes and consolidate it under one common shipment for each destination in order to reach better freight rates, but recently further co-operation seems to be under way.

According to Opgenhoff (1997, pp. 13–19), Bowen and Leinbach (2004) or Neiberger (2007), the freight forwarder business in air cargo operations falls in two main groups, with small, highly specialized forwarding companies focused on air cargo transport on one hand and big forwarding companies offering more integrated air cargo services, including transport, warehousing, packaging and other value added services as a full-service package under their own branding on the other.

Integrators incorporate all these above operations done otherwise by freight forwarders, air freight brokers, IATA cargo agents and ground handlers in order to have full control of their integrated air cargo logistic chain.²⁶⁸ By expanding into heavier shipments and offering third-party logistics services, they started to enlarge their base of operations in the 1980s.²⁶⁹ Therefore, they are considered to be the most threat to both airlines and freight forwarders in the air cargo industry today.²⁷⁰

Another active player in the air cargo industry is a small but increasing number of shippers with considerable high volumes of air cargo, which deal directly with airlines, holding separate agreements with freight forwarders or other surface carriers for pre- and onward-carriage.²⁷¹

According to pre- and onward-carriage of air cargo shipments, a distinction must be drawn between single air cargo shipments and ULDs in a door-to-door-service with the later increasing in importance on main routes.²⁷² Being often consignments of small volume and high urgency, single air cargo shipments are delivered or picked up at the airport directly by the shipper, consignee or consignor (see Fig. 2.12). Alternatively, they are transported in cartage service performed or organized by freight forwarders. Due to specialized equipment needed for the carriage of ULDs

²⁶⁷Cf. DVB (2001, pp. 30–31), Mosler (2001), Sch¨uller (2003, pp. 38–39), Siegmund (2007) or Neiberger (2007). Following Rossger (1960, pp. 26–28), this is not a recent development because ¨ already in 1951, freight forwarders at Hamburg joined together in a "Frachtenkontor" in order to pool their air cargo shipments. This business practice was adopted by other freight forwarders in the air cargo industry within a short time and was finally legalized by the IATA in 1954.

²⁶⁸Cf. Schneider (1993, pp. 91–128), Windisch (1996, p. 27), Opgenhoff (1997, pp. 20–23), Bachmeier (1999, p. 78–81), Grandjot (2002, pp. 102–103), Vahrenkamp (2002, pp. 18–20) or Sch¨uller (2003, pp. 33–35).

²⁶⁹Cf. Schneider (1993, pp. 52–54), Gutthal (1999, pp. 25–27), Bess (1996, pp. 33–63), Bjelicic (2000) or DVB (2001, pp. 32–33).

²⁷⁰Cf. Thaler (1990, pp. 104–109), Gran (1996), Bjelicic (2000) or Grandjot (2002, pp. 97–98).

²⁷¹Cf. Gompf (1994, p. 116), Gran (1996), Opgenhoff (1997, pp. 24–26) or DVB (2001, p. 13).

²⁷²Following Steiger (2006), 33.0%–48.8% of all air cargo shipped between North America, Europe and the Asia-Pacific region are so-called "Build-up Units" – ULDs containing consolidated cargo delivered to the airport "ready for carriage" by freight forwarding companies.

Fig. 2.12 Cargo flow and contractual relationships in air transport operations (Source: Kummer et al. (2010, p. 308))

in a door-to-door service, pre- and onward-carriage of such unitized shipments is generally performed or organized by freight forwarders.273

2.3 Summary and Concluding Remarks on the Functional Perspective

The findings of this part of the book can best be summed up along four environmental effects called (1) cargo structure effect, (2) integration effect, (3) (de)regulation effect, (4) logistics effect as shown in Table 2.3. They were first formulated by Gerd Aberle by describing the ongoing development of transport markets including the freight forwarding business.274

²⁷³Usually, an ULD can be picked up at the airport up to 48 h in advance free of charge in order to be build up at the premises of a shipper or freight forwarder, whereas on the onward-carriage leg, the ULD can be again brought outside the airport for 48 h before a detention charge is raised, see Smith (1974, p. 170).

²⁷⁴Cf. Aberle (1989), but similar statements can be found in Matthai (1981, 1991), Murphy and ¨ Daley (2001), Lorenz (2005, pp. 82–84) or Markides and Holweg (2006), too.

Table 2.3 Environmental effects on freight forwarding business

Level of analysis	Cargo structure effect	Integration effect	(De)regulation effect	Logistics effect
Economy and Trade	Less bulky goods and raw materials, more high valued industrial products	Building up regional trading blocks and political unions	GATT negotiations lowering trade barriers now institutionalized with WTO	From seller to buyer markets
		Free circulation of people, goods, capital and services	Increasing flexibility needed to stay in an ever volatile market environment
Transport Industry	Changes in modal split smaller shipment sizes and more frequent transports	Increased cross-border economic activities	Widespread deregulation of transport markets in the USA and EU	Cost focus set on logistics
	Standardization of handling and transport operations	Geographic dispersion of supply chains	Heavy price competition	New logistics concepts for procurement, manufacturing and distribution
Forwarding Business	Opportunities of specialization to improve quality and service levels	Integrated multimodal transport chains	New entrants	Tailored logistics solutions including value added services in information processing
		International expansion of operations	Alliance building and/or mergers and acquisitions activities

First, a cargo structure effect can be observed since the 1960s, originating from a change in production structure and resulting in a shift from low value bulky goods and raw materials to high value industrial products as an object of transportation.²⁷⁵ This had a decisive impact on the modal split with an ever increasing amount of road transport and air cargo against railway and inland navigation losing ground. Another aspect often referred to are smaller shipment sizes coming along with more frequent transports, which supported a standardisation of handling and transport operations through palletization and containerization of cargo. This in turn had a tremendous

²⁷⁵Cf. Aberle (1989, pp. 6–7, 2003, pp. 91–94).

impact on market structure and operations on all transport markets with the freight forwarding business offering opportunities in specialization on (de)consolidation of cargo or documentary paperwork to improve their quality and service levels.

Referring to the integration effect, several different levels of integration have to be distinguished. On the world level, lowering trade barriers in the framework of the General Agreement on Tariffs and Trade (GATT) boosted international trade on a world wide scale. On a country level, establishment of trading blocks like the EC, EFTA, NAFTA with an almost free circulation of people, goods, capital and services led to increased cross-border economic activities.²⁷⁶ This favoured a geographic dispersion of supply chains, which impacted the level of transport industry considerably: Cross-border cargo flows need reliable transport operations so that shippers of cargo increasingly switched to door-to-door delivery and onestop shopping.²⁷⁷ Therefore, more or less vertically integrated multimodal transport operations were demanded, where international freight forwarding companies seem to be key players.²⁷⁸ Moreover, they expanded considerably their operations on an international scale in order to meet their shipper's requirements.

Apart from the already above mentioned (de)regulation of trade through GATT negotiations now institutionalized with the World Trade Organization (WTO), a (de)regulation of transport markets took place in the United States beginning in the 1970s and in the European Union in the late 1980s.²⁷⁹ Both led to an ongoing excessive price competition among existing players, which is widely considered to be the (de)regulation effect. In addition to this, it allowed market entry of new players like integrators in the air cargo industry, worsening things even more. Moreover, heavy alliance building and/or merger and acquisition activities are a common feature not only in sea or air transportation markets: Since the mid 1990s, the freight forwarding business seems to be in an ongoing consolidation phase, too.²⁸⁰

Last but not least, the logistics effect is a logical step of shippers seeking for rationalization and cost-cutting potentials due to a change of environment from

²⁷⁶Cf. Seidenfus (1973, p. 2), Aberle (1989, p. 8, 2003, pp. 96–98) or Thaler (1990, pp. 110–126).

²⁷⁷See e.g. Suelflow and Hille (1970), Dehn (1987, pp. 15–40), Semeijn and Vellenga (1995) or Aberle (2005, pp. 380–381).

²⁷⁸Cf. Beplat (1970, pp. 128–131, 143–156), Davies (1981), Matthai (1991), Murphy et al. (1992), ¨ Murphy and Daley (1995, 1996a, 1996b, 2001), Gran (1996), Cazzaniga Francesetti and Foschi (2001), Heaver (2002b), Lorenz (2005, pp. 74–79) or Neiberger (2007, 2008).

²⁷⁹Cf. Aberle (1989, pp. 8–8, 2003, pp. 170–229, 2005, p. 380), Schachinger (1989), Simon (1993, pp. 58–94), Opatz (1994, pp. 17–73), Bess (1996, pp. 29–63), Button (2001) or Neiberger (2008).

²⁸⁰See various trade press contributions like Knee (2001), Conway (2005b), Cullen (2006) or MergeGlobal (2008a) commenting actual industry development as well as Dorrenbacher (1990), ¨ Burckhardt et al. (1998), Plehwe (1999), Klaus (1999), Ludvigsen (2000), Lemoine and Dagnæs (2003), Bowen and Leinbach (2004), Wettbewerbsverein (2004), Neiberger (2008) or Schramm and Niedermaier (2008a, 2008b) discussing this phenomenon.

seller-driven to buyer-driven markets.²⁸¹ Furthermore, introduction of new logistical concepts for procurement, manufacturing and distribution raised a demand for more flexible but time-sensitive transport operations and value-added services of information processing.²⁸² A logical consequence of this development was then an outsourcing of logistics functions to third-party logistics services, which offer tailored solutions including a wide range of value-added activities. This is still a booming market with big opportunities for all companies who want to participate.

Coming back to the freight forwarding business in multimodal transport chains with a leg by sea or air as the focus of this work, the following conclusions can be drawn:

In ocean shipping operations a freight forwarder is a freight forwarder in the very legal sense: He organizes shipment of cargo and serves as an agent for shipper, consignor or consignee. This is well documented in contributions like Schlichting (1931, p. 6), Securius and Boning (1941, pp. 59–60) ¨ , Mittendorf and Oelfke (1974, p. 241), Schumacher (1987a, pp. 148–149), Murr (1979, p. 18– 22), Branch (2000) or Dischinger et al. (2005, pp. 268–269) describing activities of a freight forwarder in ocean shipping operations.
In air cargo operations a freight forwarder is the quite natural partner of airlines in the IATA-system: Whereas air cargo freight forwarders deal with virtually all aspects of transport organization, the airlines are the performing carrier. This picture is reinforced by contribution of e.g. Weisskopf (1984, pp. 110–150), Schumacher (1987a, pp. 149–150), Bowen and Leinbach (2004) or Neiberger (2007, 2008).

Moreover, in terms of role theory as outlined in Sect. 1.2 it can be stated that a freight forwarder in multimodal transport chains with a leg by sea or air seems to be "The Architect of Transport" positioned between shipper, consignor or consignee as well as other participants of such multimodal transport chains and its role-set is well defined by legal status (at least in Germany, Austria or Switzerland) and functions served according to FIATA (1975) with two notable additions by the author (namely quasi-banking and transport function). Finally, this view is reinforced if we look at actual fields of activity and specialization in forwarding business in Germany (BSL (2000), DSLV (2005)) as well as Austria (WIFI (1993)).

²⁸¹See Welker (1988, pp. 12–26), Aberle (1989, pp. 7–8, 2003:94–95, 2005:384–385), Ihde (1989), Niegel (1991, pp. 46–65), Simon (1993, pp. 33–39), Schmidt (1996) or Allen (1997).

²⁸²Cf. Matthai (1991), Gompf (1994, pp. 101–112), Aberle (2003, pp. 94–96), Lorenz (2005, p. 79– ¨ 82) or Neiberger (2007).

Chapter 3 New Institutional Economics, Intermediation Theory, and the Intermediary Perspective

After this merely descriptive approach to forwarding business, a closer, more theoretically grounded work on the functioning of multimodal transport chains as well as intermediary roles and functions of freight forwarders therein is demanded. In the last sections, the market structure was taken for granted and freight forwarders were just considered as being one of many players serving some special functions in transport or logistic chains without looking "behind the scenes." Now, the question "why is it so?" is risen, because if freight forwarders are considered to be important intermediaries in multimodal transport and logistic chains, they cannot be only described from a legal and functional point of view. Therefore, a real intermediary perspective has to be developed, which will be done in the following sections.

First, a focus is set on the pure economic as well as to some extent on the law and economics side of New Institutional Economics (NIE) in order to clarify the role of property rights, transaction and agency costs in real-world economy from an institutional point of view, where imperfections are not exceptional but the rule.1 Then outcomes of Market Microstructure Theory (MMT) and Intermediation Theory of the Firm (ITF) are thoroughly presented, which provide a theoretical background for an intermediary perspective. Finally, the findings obtained by this approach are applied on the topic of multimodal transport chains in an international environment in order to provide a well theoretically based intermediary perspective of freight forwarding.

¹These ideas of NIE are wide-spread and were adopted in many disciplines of the social sciences, but to discuss all these schools of thought is far beyond of the scope of this work. Some prominent examples not taken into consideration are e.g. Hayek (1960, 1973) or Buchanan (1975) declaring constitutional law, North (1981) focusing on long-term socio-economic as well as political developments or Picot et al. (2002) describing organizational structures.

3.1 New Institutional Economics

As there is an ongoing battle between advocates of Institutionalism (old and new) and their counterparts favouring (Neo)classical Economics (NCE), a viable starting point of discussion about NIE is to highlight the differences between the NCE and the institutional approach to economics.2

Since the introduction of comparative-static analysis by David Ricardo, the world of NCE comes along with a lot of over-simplifying assumptions, like zero transaction costs, full information structures, full rational, profit maximizing individuals acting on markets, which always reach a general equilibrium.3 Furthermore, as institutions are only considered as some exogenous given constraints, there is a heavy abstraction from contractual relations occurring. Therefore capital structure, ownership structure and the division between ownership and control is regarded to be somehow irrelevant.⁴

3.1.1 Ancedents of the New Institutional Economics

Following authors like Erlei et al. (1999, p. 28), Martiensen (2000, pp. 77–79) or Gobel (2002:48), the roots of Institutional Economics can be even traced back to ¨ the classics of economic theory like Adam Smith and David Hume. But thinking of institutions as a central point of departure is better dated back to the mid of the nineteenth century, when in 1843 the Older German Historical School was founded by Wilhelm Georg Friedrich Roscher, later headed and further developed as the Younger German Historical School by Gustav von Schmoller.⁵ Both German Historical Schools were the first formulating a comprehensive critique on the NCE and proposed an alternative historical approach in order to derive inductively scientific theory in general economic coherence to solve concrete problems in economics. Another stream of Institutionalism originated in Europe is the Austrian School founded in 1871 by Carl Menger with a long line of successors from Eugen von Bohm-Bawerk, Friedrich von Wieser, Ludwig Edler von Mises, Friedrich ¨

²According to Aspromourgos (1991), the term "neoclassical" was first introduced by Veblen (1900) in connection with the work of Alfred Marshall (as the dominant neoclassical economist at that time, see e.g. Yonay (1998, pp. 29–35). Later on, all formal and marginalist theories in economics were subsumed under this label. In the following, "(neo)classical" is used to subsume nonformal classical as well as formal neoclassical economic theory as the great counterparts of Institutionalism without going in too much details.

³See Hutchison (1984) or Kaulmann (1987:1–12) with further references.

⁴Cf. Richter (1991, pp. 398–401) and Richter and Furubotn (2003:7–10).

⁵For this and the following see Hutchison (1984), Backhouse (1985, pp. 85–93, 218–235), Schinzinger (1991), Fusfeld (1991), Erlei et al. (1999, pp. 27–42) or Martiensen (2000, pp. 75–93).

August von Hayek to Israel M. Kirzner and Andrew Schotter.⁶ The main focus of their research was on the origins and functioning of institutions. Doing this, they followed a merely theoretical grounded research conception drawing their conclusions by deduction, which resulted in the so-called "Methodenstreit" between Carl Menger and Gustav von Schmoller. Beginning in the late nineteenth century, a quite heterogeneous movement with leading figures like Thorstein Veblen, Wesley Mitchell and John R. Commons evolved in the United States under the label "Institutional Economics," first introduced by Hamilton (1919).7 Veblen (1898, 1899) was the first social scientist who attempted to develop a theory of economic and institutional evolution essentially along Darwinism lines and provided e.g. with his hard criticism of NCE much of the intellectual inspiration for the US-American Institutionalism (especially in Veblen (1909, 1919)).⁸ Mitchell (1913, 1927) concentrated on empirical studies on cyclical fluctuation and aggregate disequilibrium in a money economy and promoted in Mitchell (1925) quantification and empirical measurement as the main methodological aim or criterion.⁹ Thinking that there is always a struggle for scarce resources, Commons (1924, 1931, 1934) focused his research mainly on the legal framework of collective economic activity with working rules governing individual behaviour and transactions as the basic unit of analysis.10

Although their overwhelming presence in economic research in Germany or Austria in the second half of the nineteenth and the early twentieth century as well as their widespread influence in the United States in the interwar period, the impact of the now called "Old Institutional Economics" (OIE) declined heavily afterwards because of various reasons.¹¹ Certainly, Institutionalism as a field of economic research was not extincted at all, but their later protagonists like the Freiburg School

⁶Cf. Kirzner (1991) or Vanberg (1998a).

⁷See Backhouse (1985, pp. 221–235), Samuels (1991), Hodgson (1998), Yonay (1998, pp. 50–76) or Rutherford (2000, 2001) focusing on the US-American stream of Institutionalism. Remarkably, some of these early US-American institutionalists have been heavily influenced by the Younger German Historical School, because they spend some or almost all of their time during graduate studies in Germany, see Backhouse (1985, p. 127), Samuels (1991, p. 864) or Rutherford (2001, pp. 176–177).

⁸Cf. Hodgson (1998, pp. 166–169), Rutherford (2001, pp. 174–176) or Martiensen (2000, p. 87).

⁹Cf. Backhouse (1985, pp. 181–183), Rutherford (2001, p. 177) or Martiensen (2000, p. 88).

¹⁰Cf. Rutherford (1983), Hutchison (1984, p. 21), Backhouse (1985, pp. 230–235), Rutherford (2000:176, 2001:293) or Martiensen (2000, pp. 88–89).

¹¹First, Institutionalism in Germany as well as in Austria was mainly settled in single schools of thought rather than being a broad movement, and the American institutionalists failed to agree apon, let alone develop a systematic theoretical core. Second, their logic underpinnings were partially disabled by a combined result of profound shifts in social science from instinct psychology and pragmatist philosophy to behaviorist psychology and positivist philosophy and a rise of mathematical modelling methods in NCE as well as econometrics for empirical studies, outdating the institutionalism movement as a mainstream of economic thought. Further, more and more extensions to NCE followed later on, like the wide field of Industrial Organization Theory (IOT) reflecting not fully competitive markets as well as Game Theory modelling individual information

around Walter Eucken, the Austrian economists Friedrich August von Hayek, Israel M. Kirzner and Andrew Schotter as well as Clarence E. Ayres and John K. Galbraith in the United States played merely a niche role.¹²

In order to briefly sum up this historical review it can be asserted, that the OIE moved from general ideas concerning human agency, institutions and the evolutionary nature of economic processes to specific ideas and theories, related to specific economic institutions or types of economy.¹³ The core ideas of Institutionalism concerning institutions, habits, rules and their evolution was facilitated by a specific and historically located approach to economic analysis.¹⁴ Moreover, as Williamson (1981:550–552, 1985:2–7) showed, startling insights into the nature of economic institutions like Frank Knight, John R. Commons, Ronald Coase, Karl Llewellyn and Chester Barnard were made in the interwar period, which will be discussed briefly in the following section.¹⁵

3.1.2 Common Assumptions of New Institutional Economics

At least four assumptions can be identified, which crucially matter the contemporary institutional approach to economics, namely (1) individuals are forming institutions as the main focus of interest (2) property rights and other contractual relationships have economic impacts, (3) positive transaction costs are always present and (4) imperfect information structures and bounded rationality have to be taken into

structures and actions en detail. See Hutchison (1984), Hodgson (1998), Yonay (1998, pp. 184– 195), Rutherford (2000:298–301, 2001:182–185) or Hodgson (2003) for further discussion.

¹²See especially Vanberg (1998b), Erlei et al. (1999, pp. 35–38) or Martiensen (2000, pp. 90–93) for the Freiburg School and Backhouse (1985, pp. 373–380) for the Austrian School as well as US-American Institutionalism after 1945.

¹³Cf. Hodgson (1998, p. 168).

¹⁴Hodgson (1998, pp. 173–174) further summed up the institutional approach as follows:

^"First, there is a degree of emphasis on institutional and cultural factors that is not found in mainstream economic theory. Second, the analysis is openly interdisciplinary, in recognizing insights from politics, sociology, psychology, and other sciences. Third, there is no recourse to the model of the rational, utility-maximizing agent [. . . ]. Fourth, mathematical and statistical techniques are recognized as the servants of, rather than the essence of, economic theory. Fifth, the analysis does not start by building stylized facts and causal mechanisms. Sixth, extensive use is made of historical and comparative empirical material concerning socio-economic institutions. In several of these aspects, institutional economics is at variance with much of modern mainstream economic theory."

¹⁵Notably, only John R. Commons and Ronald Coase can be regarded as "real" institutionalists, while the others have a different background and can be at best regarded as peripheral figures despite their proximity to OIE: Frank Knight was merely a (neo)classical economist, Karl Llewellyn a lawyer and Chester Barnard an organization theorist.

consideration.16 In the following, this common core of NIE will be discussed along its roots and main initial contributions contrasting NCE.

3.1.2.1 Methodological Individualism and Institutions

A basic assumption of NIE is methodological individualism¹⁷ and therefore, institutions (like societies, states, communities, firms or other forms of organizations) are formed by individuals with different preferences, goals, intents or ideas.¹⁸ As there is a long scientific history of thought dedicated to institutions and their various outcomes, a myriad of definitions and interpretations of institutions were raised, which provide at the same time a demarcation line between NCE and the different streams of Institutionalism.¹⁹ Following e.g. Rutherford (1983, p. 722), (neo)classical economists usually treat institutions as "[. . . ] having the role of given constraints. On the other hand, institutionalists have tended to be more concerned with questions of institutional change and many have emphasized the instrumental role of institution." Furthermore Williamson (1998a, p. 88) asserted, that "[t]he institutions of principal interest to the NIE are the institutional environment (or Rules of the Game – the polity, judiciary, laws of contract and property [North (1991)]) and the institutions of governance (or play of the game – the use of markets, hybrids, firms, bureaus)."

In a more recent contribution defining institutions as sanctionable expectations in reference to behaviour of one or more individuals, Picot et al. (2002:11–25) provided a conceptual framework of superior and subordinate institutions forming a hierarchical structure. For them, the highest level of this institutional hierarchy

¹⁶Notably, Hamilton (1919, pp. 312–317) proposed quite similar five characteristics an economic theory must be able to meet, namely (1) it should unify economic science, (2) it should be relevant to the modern problem of control, (3) the proper subject-matter is institution, (4) it is concerned with matters of process and (5) it must be based upon an acceptable theory of human behavior, and to him, all these propositions are met by Institutional Economics.

¹⁷The term "methological individualism" was originally defined by Schumpeter (1908a:88–98, 1908b), see Erlei et al. (1999, p. 6) or Furubotn and Richter (2003:3) with further explanations.

¹⁸Following e.g. Furubotn and Pejovic (1972:1137), "[t]he organization [sic!] per se is no longer the central focus; rather, individuals are assumed to seek their own interests and to maximize utility subject to the limits established by the existing organizational structure."

¹⁹See Erlei et al. (1999, pp. 23–27), Gobel (2002, pp. 1–21), Martiensen (2000, pp. 10–20) ¨ or Richter and Furubotn (2003:7–8). Some of the very prominent definitions are as follows: According to Veblen (1899), institutions are "[. . . ] settled habits of thought common to the generally of men." For Commons (1931, p. 648), institutions are problem solving instruments or mechanisms through which collective control is exercised and therefore he defined them as "[...] collective action in control, liberation and expansion of individual action." Schotter (1986:117, as cited in Erlei et al. (1999, p. 23)) considered that "institutions are seen as a set of rules that contain individual behavior and define the social outcomes that result from individual action." And last but not least, Furubotn and Richter (2003:7) defined institutions as a system of mutually bounded explicit (formal) or implicit (informal) rules (norms) including their enforcement.

consists of human rights, basic social rules and norms, language and money which all result from an evolutionary process and therefore cannot be influenced or altered. The lower level institutions are derived ones like law, court judgements, contracts and organizational rules which can be established to restrict individual behaviour. Finally, there are constitutional institutions in between, which arise out of social contracts closed between individuals codifying common rules and norms. These mutually recognized rules and norms allow then individuals to live together in one society, because they oppose rights and duties on individual behaviour.

3.1.2.2 Property Rights and Other Contractual Relationships

As already indicated in the last section, important topics of contemporary Institutionalism are the existence of specific property right structures and their economic impacts, which led to alternative institutional arrangements with different rights to use (or abuse) given resources.²⁰ In contrast to this notion of NIE, NCE regards them as a "datum" or something given and therefore they are not of interest.²¹

As various more or less incomplete as well as context-specific approaches can be found in economic literature to define the term "property rights," it seems to be quite useful to start with a merely legal approach.²² Following Hume ([1739/40] 1978:484–534) and his individualistic view of "human society," there are three "fundamental laws of nature" on whose strict observance the peace and security of such a human society entirely depends: (1) the stability of possession, (2) its transfer by consent, and (3) the performance of promises. Transferring this in more contemporary terms, the legal system of a civil state has to rule at least (1) ownership rights, (2) freedom of contracts and (3) obligations or liabilities from promises as well as illegal actions.²³ According to Richter and Furubotn (2003, pp. 95–105), property rights can be divided into three distinct groups: (1) absolute property rights, (2) relative property rights and (3) other forms of contractual relationships which will be discussed in the following.

²⁰Cf. De Alessi (1980, p. 2).

²¹Cf. Demsetz (1967) or Richter and Furubotn (2003, p. 88). De Alessi (1980, p. 2) further asserted, that

^"in the process of developing a rigorous statement of the formal theory [. . . ] economists gradually eliminated institutional constraints from considering by assuming either explicitly or implicitly, that all rights to the use of resources were fully allocated, privately held, and voluntarily exchanged at zero information and transaction (negotiating, contracting, and policing) costs."

²²See Kaulmann (1987:14–15), Tirole (1999, pp. 742–742), Martiensen (2000, pp. 221–223), Richter and Furubotn (2003:16–18) citing some definitions from economic literature.

²³Cf. Richter and Furubotn (2003, pp. 16–18). Moreover, Posner (1998, p. 35) showed a similar view of things, as he divided Anglo-Saxonian Common Law into three main parts, namely (1) the Law of Property, (2) the Law of Contracts, and (3) the Law of Torts.

Absolute Property Rights

First, there are absolute property rights, or ownership rights, which are a permission to use, enjoy and dispose of things fully and freely subject to the limits and conditions for doing so determined by law, like clearly stated in the case of - 903 DBGB.²⁴ This imposes, as Posner (1998, p. 38) asserted from an economic point of view, "that if every valuable (meaning scarce as well as desired) resource were owned by someone (the criterion of universality), ownership connoted the unqualified power to exclude everybody else from using the resource (exclusivity) as well as to use it oneself, and ownership rights were freely transferable, or as lawyers say alienable (transferability), value would be maximized." Furthermore, not only material but also immaterial goods can be subject to such absolute property rights, and this is well reflected in contemporary intellectual property law (including patents, copyrights, trademarks, trade secrets and privacy) or human rights.25

However, as already indicated above, the term "absolute" does not implicate, that such property rights are totally unrestricted.²⁶ There are a lot of legal regulations set to the execution of property rights in order to prevent harm to others or provide resolution in the case of conflicting property rights, so that a holder of a property right has always to obey not to offend his "neighbourhood."27 As such conflicts are often not easily solvable just by referencing legal rules, extensive court trials are not unusual, if the costs of specification and enforcement of these property rights in question are not too high. From an economic point of view, such restrictions imposed on property rights in connection with incomplete specification and costly contracting lead to a decrease of their value, commonly called "attenuation" of property rights.²⁸

Moreover, property rights can be shared among several individuals leading to a common ownership.29 One example of such shared property rights are common or public goods which lead to the well known "tragedy of the commons" because

²⁴Cf. Gobel (2002, p. 68) or Richter and Furubot ¨ n (2003, pp. 95–96). In Anglo-Saxon literature like Alchian (1965, p. 818–819), Furubotn and Pejovich (1972, p. 1139), De Alessi (1980, p. 4) or Alchian (1991, p. 1031), the term "private property rights" is often used, indicating a private ownership of goods by individuals.

²⁵Cf. Posner (1998, pp. 43–50), Richter and Furubotn (2003, pp. 97–99) or Gobel (2002, pp. 66– ¨ 67). Following Gobel (2002, pp. 66–67), even reputation of a firm or a person as well as human ¨ capital (e.g. special knowledge) and social capital (e.g. a personal network of contacts) can be regarded as property rights in a wider sense.

²⁶See Gobel (2002, pp. 68–71), Richter and Furubotn (2003, pp. 96–97) going into much more ¨ detail concerning German Law.

²⁷See Coase (1960) with some real legal cases showing clearly the problems occurring in execution of absolute property rights.

²⁸Furubotn and Pejovich (1972, pp. 1146–147) seem to be the first authors who used this term. See further Tietzel (1981:211–218), Schoppe (1995, pp. 140–142) or Picot et al. (2002, pp. 55–58) for more discussion about this topic.

²⁹Cf. Alchian (1965, pp. 819–823), Demsetz (1967), De Alessi (1980, pp. 5–9), Erlei et al. (1999:280–287), Gobel (2002, pp. 71–77) or Martiensen (2000, pp. 230–253). ¨

there is always a tendency to underinvest in preservation (e.g. of wild animal stock) or maintenance (e.g. of commonly owned assets), paired with their overusage until extinction, breakdown or total destruction.30 Closely related to this problem is the pooling of several privately held resources in an organization with team production where a final product resulting from such a joint production has a non-decomposable and non-attributable value so that it is not possible to identify how much of the final product comes from each resource input.³¹

Another aspect overlooked so far is the possibility of partitioning of property rights.³² More specifically, an absolute property right can be split into:33

The right to use the property (Lat. "ius usus").
The right to dispose the property (Lat. "ius abusus").
The right to take and keep fruits or revenues produced by the property (Lat. "ius usus fructus").
The right to transfer some or all rights on the property to someone else (Lat. "ius successionis").
The right to exclude others from usage of the property.

By augmenting this, a transaction of goods can then be considered as a transfer of bundles of property rights connected with these goods and some of them can be given up and transferred from one to another with the other property rights still held by the first individual. Straight forward examples are lending, renting, leasing or chartering goods from an original owner to another individual, coming along with a temporary divergence of ownership and possession.³⁴

Relative Property Rights

The second group of property rights are relative ones, which are either contractual obligations or all sorts of liabilities, especially in connection with a contractual relationship.³⁵ The very difference to the absolute property rights discussed above is

³⁰Cf. Demsetz (1967), De Alessi (1980, pp. 6–7) or Alchian (1991, p. 1033).

³¹This idea was first developed in Alchian and Demsetz (1972). According to them, a viable solution is to assign a team leader, (1) being the central party common to all contracts of the jointed resources, (2) having rights to renegotiate any input's contract independently of contracts with other team members, (3) holding the residual claim and (4) being allowed to sell his central residual status. See Alchian (1991:1032), Richter (1991, pp. 401–406), Schoppe (1995, pp. 142–146), Erlei et al. (1999, pp. 70–74) or Martiensen (2000, pp. 255–262) for further discussion.

³²This aspect of property rights was first explored by Coase (1960) and Alchian (1965), see also Gobel (2002, pp. 66–67,78). ¨

³³Cf. Alchian (1965, pp. 819–820), Furubotn and Pejovich (1972, pp. 1139–1140), Tietzel (1981:210–211), Schoppe (1995, p. 137), Wolff and Neuburger (1995, pp. 79–80) or Gobel ¨ (2002:66).

³⁴Cf. Gobel (2002, p. 67). ¨

³⁵Cf. Richter and Furubotn (2003, pp. 100–102, 145–172).

that relative property rights are subject to a single or a certain group of individuals. Most of them come along with the partition and transfer of absolute property rights, where an original owner of a good obtains a relative property right towards the borrower, lessee or charterer in form of a claim like a promise to pay interest, rent, lease or charter.³⁶ Another sort of relative property rights are liabilities growing out of damage or loss of goods as well as hurt or even death of individuals arising from their own fault.37 Two of many examples are (1) the contractual liability of freight forwarders in Germany for any damage resulting from loss of or damage to the goods occurring during the time between taking them over and their delivery as stated in - 425 HGB or (2) the liability rule valid for Austrian freight forwarders according to the provision in - 407–415 UGB in connection with - 388–390 UGB, which considers them to be only liable in conducting their activities to the extent that they are at fault.³⁸

In the context of relative property rights, special interest are synallagmatic contracts relying on the principle "do ut des" (Lat.), where a sequence of mutual exchange of goods and/or property rights takes place, based on law–specified contractual duties and obligations.³⁹ Examples are sales contracts of goods as well as property rights, labour contracts and so on.⁴⁰

Other Forms of Contractual Relationships

In addition to these absolute and relative property rights clearly defined by law, there are "other" forms of contractual relationships not fully backed by legal measures.41 Furthermore, as every transaction is essentially driven by some contractual relationship, various forms of either legal or non-legal contractual relationships have to be considered as means with economic impact.⁴² An important early contribution

³⁶Demsetz (1967, p. 347) referred to this situation as follows: "When a transaction is concluded in the marketplace, two bundles of property rights are exchanged. A bundle of rights often attaches to a physical commodity or service, but it is the value of the rights that determines the value of what is exchanged." But he left open, what sort of property rights are exchanged.

³⁷Cf. Richter and Furubotn (2003, p. 102).

³⁸For a formal economic interpretation of these quite different rules of liability in general see Erlei et al. (1999, pp. 287–295) with further references. Further, it has to be noted, that in both cases there are more provisions for exclusion or limiting maximum liability as well as full liability in the case of intent or gross negligence, but describing this en detail is beyond of the scope of this work. 39Cf. Richter and Furubotn (2003, pp. 151–153). The term "synallagmatic" itself is derived from "synallagma" (Gr.), which refers to developing actions on written promises, see Llewellyn (1931, p. 740).

⁴⁰See Gobel (2002, pp. 79–92) or Richter and Furubotn (2002:165) discussing some examples ¨ from German Law.

⁴¹Cf. Richter and Furubotn (2003, pp. 102–105 and 165–171).

⁴²Llewellyn (1931, pp. 707–708) even regarded the term "contract" itself as "[. . . ] an ambiguous concept, ambiguous particularly when more is concerned than unmixed legal doctrine. (1) The word is used especially to indicate business agreements-in-fact, as such, irrespective of their legal

in this direction is Llewellyn (1931), who distinguished between "iron rules" and "yielding rules" in contracting, and proposed a concept of "self-government" supporting relationships between parties involved therein, which is not necessarily guided by conventional law measures.⁴³ At the same time, he shared a mutual commitment with Common's (1924) legal approach to economics.44

In an empirical study of non-contractual relationships in the manufacturing sector, Macaulay (1963) observed actual contracting practice in every-day business transactions. Doing this, he drew the conclusion, that "businessmen often fail to plan exchange relationships completely, and seldom use legal sanctions to adjust these relationships or to settle disputes. [. . . ] Transactions are planned and legal sanctions are used when the gains are thought to outweigh the costs" (Macaulay (1963, p. 55)). But often, contracts are not needed, because products are standardized with a well accepted specification or can be tested to check them against what was ordered originally.⁴⁵ Furthermore, there are many effective nonlegal ways to sanctioning with commitments to be honoured in almost all situations and product quality to be ensured out of self-interest. Another important reason not to stick on well defined contracts is, that they come along with a loss of "flexibility." Therefore, "relative contractual practices" are often employed in order to e.g. avoid problems occurring from complexity of agreed performance over a long time. These rather vague propositions inspired law scholars like Summers (1969) and Macneil (1974, 1978, 1980, 1983) to explore the legal side of such contractual practice apart from the ordinary system of (neo)classical contract law with the later developing a legal theory of relational contracting.

Macneil (1974, 1978) thoroughly discussed the differences between transactional (classical or neoclassical) and relational contracting, whereas in Macneil (1980, 1983) he further developed his notion in a much broader context along a system of common contract norms as well as other, supracontractual norms.⁴⁶ His analysis

consequences – irrespective indeed of whether they have legal consequences. . . . – Or the word is used to indicate agreements-in-fact with legal consequences. . . . Again, the word indicates the legal effects, if such there be, of promises [. . . ]. I shall endeavor to reserve "promise" for the promise-in-fact, "contract" for the legal effects of such a promise. (4) A fourth current meaning of the word, the writing embodying an agreement (commonly assumed to be one with legal consequences) may here be disregarded."

⁴³Cf. Llewellyn (1931, pp. 727–734). More specifically, he imposed that "[. . . ] the major importance of legal contract is to provide a framework for well-nigh every type of group organization and for well-nigh every type of passing or permanent relation between individual and groups [...]–a framework highly adjustable, a framework which almost never accurately indicates real working relations, but which affords a rough indication around which such relations vary, an occasional guide in cases of doubt, and a norm of ultimate appeal when the relations cease in fact to work." (Llewellyn (1931, pp. 736–737). See also Richter and Furubotn (2003, pp. 165–166) for further discussion.

⁴⁴Cf. Llewellyn (1931, p. 706).

⁴⁵See for this and the following Macaulay (1963, pp. 62–65).

⁴⁶Following Macneil (1980:40–59, 1983:347), there are ten common contract norms: (1) role integrity, (2) mutuality (as in Macneil (1980)) or reciprocity (as in Macneil (1983)), (3)

of contracts is based on the recognition of the fact that contractual relationships are conducted within a "social matrix."⁴⁷ Exchange is only possible within a society that provides at the minimum (1) a means of communication so the parties can understand each other, (2) a system of order so that the parties use exchange rather than force to get what they want, (3) a payment mechanism, and (4) a mechanism to enforce promises if some performance has to take place in the future.⁴⁸ Since a contract can only be made against such a social background, all contracts have to be considered relational to some degree, in the sense of involving social relationships and being embedded in a much broader social context. Some contracts are, however, less relational than others in the sense that they emphasize "discreteness" (a oneoff transaction between the contract parties involved) and "presentation" (detailed propositions stipulating all the contract parties' expectations and remedies for breach) so that there exists a spectrum of contractual behaviour with highly relational contracts at one end and merely discrete transactions at the other:49

At the discrete end of this spectrum are discrete transactions based on classical contracting, which comes along with (1) treating the identity of the contract parties as irrelevant to the transaction, (2) commodifying as much as possible the subject matter of contracts, (3) limiting strictly the means employed in establishing the substantive content of the transaction, (4) limiting the contract remedies available, so that e.g. the consequences from non-performance are relatively predictable from the beginning and are not open-ended, (5) drawing clear lines between being in and not being in a transaction, (6) discouraging the introduction of third parties into a contractual relation, (7) equating of the legal effect of a transaction with the promises creating it, (8) supplying a precise, predictable body of law to deal with all aspects of the transaction not encompassed by the promises and (9) stressing on expectation remedies, whether specific performance or damages measured by the value of performance, tends to bring the future into the present, since all risks, including market risks, are thereby transferred at the time the "deal is made."50
Neoclassical contracting provides some remedies to alleviate the strict limitations of pure classical contracting in long-term relationships, as it allows planning

implementation of planning, (4) effectuation of consent, (5) flexibility, (6), contractual solidarity, (7) restitution, reliance, and expectation interests, (8) creation and restraint of power, (9) propriety of means (added to the list in Macneil (1983)), and (10) harmonization with the social matrix. Further, social and political norms being not particularly contractual but have to be taken into account in contractual relationships are subsumed under the label "supracontract norms," see Macneil (1978:898, 1980:70, 1983:350).

⁴⁷Cf. Macneil (1974:710–711, 1980:11, 1983:344).

⁴⁸To some extent, this notion resembles very much to the conceptual framework of superior and subordinate institutions forming a hierarchical structure proposed by Picot et al. (2002, pp. 11–25) as already shown above.

⁴⁹Cf. Macneil (1974:735–805, 1978:862–865).

⁵⁰See Macneil (1978, pp. 863–863), where he considered (1)–(6) as ways to implement discreteness and (7)–(9) merely focusing on presentation.

flexibility by (1) proposing contract adjustment tied to external standards, direct third-party determination of performance, one-party control of specific contract terms, compensation schemes based on costs or simply agreements to agree on (re)negotiation about contract details later on, (2) deviation of original planning in existing contractual relations by actual realized (non)performance as well as additions, subtractions or terminations based on mutual agreements or unilateral concessions by one of the contract parties, (3) implementing non-disruptive dispute settlements like mediation or arbitration and (4) proposing clear rules for termination.⁵¹ But, as Macneil (1978, p. 885) asserted, "neoclassical contract law can free itself only partially from the limitations posed by obeisance to the twin classical goals of discreteness and presentation. This obeisance is imposed by adherence to an overall structure founded on full consent at the time of initial contracting."

• In relational contracting, social relationships and their preservation, harmonization of relationship conflicts as well as the influence of supracontractual norms play a significant role, whereas discreteness and presentation aspects are of minor importance.52 In order to illustrate this radical departure from the (neo)classical contracting approaches, Macneil (1974, pp. 744–805) compared in depth differences between discrete transactions and relational contracts along 12 categories and took the process of dispute resolution as an example in Macneil (1978, pp. 891–892). Regarding (neo)classical contracting, "this process is rather like the discrete transaction itself: sharp in (by commencing suit) and sharp out (by judgment for defendant or collection of a money judgment by plaintiff)" (Macneil (1978, p. 891)). In the world of relational contracting, however, (1) the scope of such a dispute is not exogenously given by contract terms but is shaped by both contract parties, some third-party arbitrator and by the entire relation in the past, the present and the future, (2) the contractual relationship is not strictly bilateral but sprawling and amorphous, (3) the fact inquiry is not only historical and adjudicative but also predictive and legislative, (4) relief is not conceived primarily (or sometimes at all) as compensation for past wrong doing in a form logically derived from the substantive liability and confined in its impact to the immediate parties, (5) the remedy is not imposed but negotiated and mediated, (6) the award will require continuing administration by the arbitrator, (7) he is further active, with responsibility not only for credible fact evaluation but also for organizing and shaping the dispute processes to ensure a justified and viable outcome, and (8) the subject matter of the dispute is not between private individuals about private rights but is a grievance about the operation of policies of the overall contractual relation.

⁵¹See Macneil (1978, pp. 865–883) with a more detailed description of neoclassical contracting measures and their impact on long-term contractual relationships.

⁵²Cf. Macneil (1978:883–886, 1980:64–70).

It were then Goldberg (1976a,b) and Williamson (1976), who first introduced this concept of relational contracting as developed by Macneil (1974) into the economic field.53 Furthermore, Richter and Furubotn (2003:187–191) recently provided an extensive list of enforcement mechanisms for such contractual relationships: (1) self-enforcing mechanisms (Telser (1980), Kronman (1985)), (2) embeddedness of actors in a social network context (Granovetter (1985, 1992), Uzzi (1996, 1997)), (3) "tit-for-tat" strategies from game theory (Axelrod (1984)), (4) enforcement of contracts with the help of third-parties (Williamson (1985)), (5) governmental regulation (Goldberg (1976a)) as well as (6) forming a union or a unified, vertical integrated organization, (Williamson (1975, 1985), Kronman (1985)). Especially the first category is worth to explicate shortly, because the other enforcement mechanisms are either self-explicable or will be thoroughly discussed later in subsequent sections: Telser (1980) laid ground for a formal theory of selfenforcing contracts on the principle of "honesty pays"54, whereas Kronman (1985) summarized more practical concepts of contractual self-enforcement in cases of non-legally enforceable agreements, like exchange of hostages, collaterals or handstying agreements.55

3.1.2.3 Positive Transaction Costs

One of the most crucial departures of NIE from (neo)classical thoughts seems to be the presence of positive transaction costs. Commons (1931, p. 662) first advanced the proposition that "[. . . ] a unit of activity – a transaction with its participants" must be the basis unit of economic analysis. Furthermore he asserted, that "transactions are the means, under operation of law and customs, of acquiring and alienating legal control of commodities, or legal control of the labour and management that will produce and deliver or exchange the commodities and services, forward to

⁵³Cf. Schoppe (1995, p. 157) or Furubotn and Richter (2003:207).

⁵⁴More concrete, Telser (1980) drew three important conclusions: (1) Self-enforcing agreements are not feasible if the sequence of occasions for transactions has a definite known last element, (2) for a given sequence of gains the expected horizon must be long enough or there can be no self-enforcing agreement and (3) parties would not enter a self-enforcing agreement expecting the other parties to violate it. Further see Martiensen (2000, pp. 432–438) or Richter and Furubotn (2003:276–284) for alternative formal explanations of self-enforcing contracts.

⁵⁵See Richter and Furubotn (2003, pp. 182–185). Following Kronman (1985), a hostage is anything (or anyone) of value to a promissor handed over temporarily to a promissee in order to get it back after performing his promise, whereas a collateral is an asset given as a hostage with a direct use or exchange value for the promissee so that is a direct substitute for performance. In the case of hands-tying, a promissor takes actions like public observable commitments to performance making his promise more credible, because if he does not perform successfully, he automatically harms himself e.g. by loosing his reputation. Remarkably, he is not the first author describing these concepts initially (as indicated by his references to Schelling (1956), Klein and Leffler (1981) and Williamson (1983)), but he brought them into a wider context.

the ultimate consumers."⁵⁶ Relying on a classification based on both legal and functional criteria, he distinguished between three major types of transaction, namely bargaining, managerial and rationing transactions, and was aware of the substitutability between the first two.⁵⁷

Another early contribution is the break-through seminal article by Coase (1937), where he argued transaction costs to provide a way of understanding "the Nature of the Firm."⁵⁸ Specifically, he asserted that the characteristic feature of the firm is, that within it, decisions about the allocation of resources could be made administratively, rather than through the market. For him, the main reason for this are simply transaction costs and therefore he formulated a common transaction cost economizing framework to derive optimal size and boundaries of a firm.⁵⁹ Almost half a century later, Coase (1960) highlighted again the need to take positive transaction costs explicitly into consideration as he observed "the Problem of Social Costs" with a focus set on "[. . . ] those actions of business firms which have harmful effects on others" (Coase (1960:1)). Discussing several legal cases, he came to the conclusion that in a world of zero transaction costs, the allocation of resources would be independent from a given legal environment since people can always costless negotiate, subdivide and combine property rights whenever this would be favourable to them.⁶⁰ Therefore, all actions "[. . . ] that could be taken which cost less than the reduction of damage they would bring, and they were the least costly means available to accomplish such a reduction, they would be undertaken" (Coase (1990, p. 175)). But taking positive transaction costs into account, property rights matter, because then the allocation of property rights among individuals has decisive economic impacts.⁶¹

Within a few years after Coase (1960), the concept of transaction costs evolved to be a well respected general approach to assess theoretically frictions in realworld economy.⁶² Accordingly, Arrow (1969, p. 48) defined transaction cost as "the

⁵⁶Commons (1931, pp. 656–657).

⁵⁷See Commons (1931, pp. 662–663). According to Rutherford (1983, pp. 724–726), he provided another classificatory scheme, where he draws a distinction between "strategic" and "routine" transactions, with the first giving control over an otherwise "limiting factor" and the second dealing with "complementary factors," after the first being under control.

⁵⁸Several years later he asserted that in Coase (1937) he showed that "[. . . ] in the absence of transaction costs, there is no economic basis for the existence of the firm" (Coase (1990, p. 14)). Notably, Coase (1937) initially did not use the term "transaction costs" but called them "the cost of using the price mechanism," "the cost of carrying out a transaction by means of an exchange on the open market" or simply "marketing costs," see Coase (1990, p. 6) or Spulber (1999, pp. 232–233).

⁵⁹See Cheung (1983, pp. 3–10), Coase (1990, pp. 5–7), Spulber (1999, pp. 233–234) or Richter and Furubotn (2003:80–81).

⁶⁰Later on, Stigler (1966, p. 113) restated this as the "Coase Theorem," see Casson (1990:10– 16 and 157–185), De Meza (1998) or Richter and Furubotn (2003, pp. 111–116) with further comments.

⁶¹See Coase (1990, pp. 174–179) for further discussion.

⁶²Cf. Williamson (1985, pp. 15–16).

costs of running the economic system"63, whereas Williamson (1985, pp. 18–19) considered them to be "the economic equivalent of friction in physical systems" and have to be distinguished from production costs. Furthermore, Williamson (1985, pp. 20–22) asserted, that there should be a distinction drawn between ex ante (before contracting) and ex post (after contracting) types of transaction costs.⁶⁴ The first are costs of drafting, negotiating, and safeguarding an agreement, whereas the second include (1) maladoption costs incurred when transactions drift out of alignment, (2) haggling costs incurred if bilateral efforts are made to correct ex post misalignments, (3) setup and running costs associated with the governance structures to which disputes are referred and (4) bonding costs of effecting secure commitments.⁶⁵ Especially Oliver E. Williamson as one of the leading figures in contemporary transaction cost economics literature refined and extended these early treatments of Commons (1931) and Coase (1937, 1960) on transaction costs in many ways, which will be shown later on in Sect. 3.1.3.1.

3.1.2.4 Imperfect Information Structure and Bounded Rationality

After defining the role of institutions, property rights and transaction costs in Institutional Economics, there is still one important point of criticism left against NCE: an overwhelming discrepancy between (neo)classical assumptions of full rational, profit maximizing individuals with perfect information structures and the empirical evidence from real world economy. Acknowledging the importance of studying "human nature as we know it" (Knight (1921:270)), he was one of the first authors challenging these simplifications of full knowledge, expectations and rationality in NCE and exploring the impacts of risk and uncertainty on profits as well as on individual behaviour:

"The facts of life [. . . ] are in a superficial sense obtrusively obvious and are a matter of common observation. It is a world of change in which we live, and a world of uncertainty. We live only by knowing something about the future; while the problems of life, or conduct at least, arise from the fact that we know so little. [. . . ] If we are to understand the workings of the economic system we must examine the meaning and significance of uncertainty; and to this end some inquiry into the nature and function of knowledge itself is necessary" (Knight (1921:199)).

There are two aspects, which are worth to be further elaborated: (1) the presence of imperfect information structures which rise from the unavailability of full

⁶³See Williamson (1985:18, 1999:5), Schoppe (1995, p. 149), Erlei et al. (1999, p. 42) or Martiensen (2000:272).

⁶⁴See further Picot (1982, p. 270) for a quite similar definition well reflected in German transaction cost literature or Richter and Furubotn (2003, pp. 57–71) with a recent, more detailed definition distinguishing between market, firm and political transaction costs.

⁶⁵Moreover, these costs of contracting are interdependent and therefore, they must be addressed simultaneously rather than sequentially.

information about all states of nature (or common knowledge) as well as each individual's private knowledge and (2) the fact of bounded rationality as individuals cannot be regarded as a fully rational calculator taking all this information available to him into consideration.

Imperfect Information Structures

Talking about imperfect information structures, a distinction has to be drawn between (1) adverse selection, (2) moral hazard, (3) hold-up and some combinations of these first three basic information problems as shown later on in Sect. 3.1.3.2. Despite economists having long recognized all these information problems and their impacts on economics and markets, modern theoretical treatment providing solutions to alleviate them began in the 1960s.⁶⁶

Adverse selection as a pre-contractual information problem was first explored explicitly by Akerlof (1970), who investigated a stylized market for used cars and showed, that there is a "Lemon's Problem" regarding their quality which led to a market breakdown. It was further generalized by Wilson (1980), taking different price-setting mechanisms into account.
Conversely, moral hazard is the technical term for a variety of post-contractual information problems arising from opportunistic behaviour with individuals maximizing their own utility to the detriment of others in situations where they do not bear the full consequences or enjoy the full benefits of their actions.67 One of the first explicit treatments of moral hazard problems was made by Arrow (1963a), who showed the impacts of medical insurance policies on health care incentives.68
In a hold-up problem situation, an information problem arises from the timing of unobservable investments done by individuals in connection with closing a contract. These investments are sunk at least to some extent, because they are not fully rewarded by their counterparts. This hold-up problem was first introduced in the field of economics by Goldberg (1976b).

⁶⁶One prominent earlier example is Knight (1921), who discussed in his treatment about risk and uncertainty problems of adverse selection and moral hazard on insurance markets and principal– agent relationships, but did not recognize that these problems can be alleviated or eliminated with properly structured contracts, see Hutchison (1984, pp. 24–26), LeRoy and Singell (1989), Langlois and Cosgel (1993) or Laffont and Martimore (2002:19).

⁶⁷Cf. Kotowitz (1990).

⁶⁸Following Lafont and Martimort (2002:13), it was again Arrow (1963b) who first adopted the idea of moral hazard to principal–agent situations.

Bounded Rationality

Apart from Knight (1921) already cited above, other early contributions coping with limited information processing capabilities of individuals are Commons (1924, 1931, 1934) considering them constrained by working rules and merely directed by habits, instincts and passions than by reason or Barnard (1938) and Simon (1945) concerning with the intended rationality of individuals inside given formal organizational structures.69

It was then Herbert A. Simon, who wrapped up and explicitly formalized these initial concepts of individuals suffering from somehow "constrained" rationality and formulated a principle of bounded rationality, widely accepted today in the field of NIE (and even beyond):70

"The capacity of the human mind for formulating and solving complex problems is very small compared with the size of the problems whose solution is required for objectively rational behaviour in the real world – or even for a reasonable approximation to such objective rationality." (Simon (1957:198–199))

More explicitly, his concept of bounded rationality shows the limitation of a decision maker along three dimensions: (1) He has an imperfect or limited knowledge of his environment, (2) he is unable to anticipate and consider all options to solve his problem due to limited abilities of calculation and (3) he is not able to process all available data, because of his limited skills of attention.⁷¹ Furthermore, this imperfection of individuals leads to a limited search under possible alternatives, accepting the first satisfactory one instead of an exhaustive search to find the very best, so that the goal of "maximizing" has to be replaced by the goal of "satisficing" or finding an alternative that is "good enough."⁷²

3.1.3 Building Blocks of New Institutional Economics

As the NIE is an ever evolving field of research, it is subject to a steady further development and nearly every new contribution attempting to develop a framework of NIE brings in another view of things. But most authors agree, that NIE has (at least) three building blocks, namely property rights theory (PRT), transaction cost theory (TCT) and agency theory (or principal–agent theory (PAT)), which form a

⁶⁹Cf. Forest and Mehier (2001, pp. 591–597) and Williamson (1985:5–7, 1999:29–44).

⁷⁰Of course, as contributions by Conlisk (1996), Kreps (1998) or Rubinstein (1998) clearly showed, this conception of bounded rationality is not the only one but it is the most prominent.

⁷¹Cf. Simon (1991a), Forest and Mehier (2001, pp. 593–595) or Martiensen (2000, pp. 179–191).

⁷²Cf. Simon (1957, pp. 204–205), Simon (1991b), Erlei et al. (1999, pp. 9–12) or Conlisk (1996, p. 675–767).

"theoretical core" of NIE.73 But this is not the one and only possible classification of the economic side of NIE:

Williamson (1985, p. 29) proposed a "cognitive map of contract," where an efficiency branch of contract is split into two parts with the incentive alignment literature (PRT and PAT) on one side and TCT with the measurement branch as well as the governance approach on the other.
Nilakant and Rao (1994) divided NIE in an incentive alignment stream with PRT and PAT emphasizing ex ante side of contracts which safeguard the parties involved against all possible contingencies and another stream with TCT focusing merely on ex-post side of contracts.
Erlei et al. (1999) initially saw the economic part of NIE split in PAT and TCT, but in the following they subsumed team production from PRT and PAT under the label of transaction costs named "Transaktionskosten I," while "Transaktionskosten II" was dedicated to the measurement branch of TCT and a more formal treatment of the governance approach centering around incomplete contracting. Furthermore, PRT was placed in another chapter, discussing the institutions of law.
Martiensen (2000) started with PRT followed by TCT and a chapter about economic contract theories, where he discussed explicitly not only PAT but also theories of incomplete, self-enforcing as well as relational contracts.
Richter and Furubotn (2003) regarded NIE as a conglomerate of TCT, PRT and contract theory with the later consisting of PAT, incomplete, self-enforcing and relational contracts added by the theory of implicit contracts and the management theory of the firm.⁷⁴

For the purpose of the study, (1) TCT with the organization failure framework as well as the governance approach to contractual relations and (2) PAT including adverse selection, moral hazard and hold-up problems are briefly described, because especially these two streams of thought in NIE add useful insight to intermediary relationships discussed later on in this piece of work.

3.1.3.1 Transaction Cost Theory: Organization Failures Framework and Governance of Contractual Relations

As already indicated above, contemporary TCT is closely connected with the ongoing research agenda of Oliver E. Williamson well documented in three monographs

⁷³See Wolff and Neuburger (1995, pp. 75–83), Gobel (2002, pp. 60–65), Picot et al. (2002:54–156) ¨ or Richter (2005:166–174).

⁷⁴Following Rosen (1985), the theory of implicit contracts goes back to Knight (1921) and copes mainly with labour market exchanges in terms of contracts. The managerial theory of the firm originates in a management study by Berle and Means (1932) and focuses on manager preferences and opportunistic behavior, see Richter and Furubotn (2003, pp. 220–224).

(Williamson 1975, 1985, 1996), where he developed an interdisciplinary theory of transaction cost economics (TCE) from a law, economics and organizational perspective.75 For him, "[t]ransaction cost economics (1) eschews intuitive notions of complexity and asks what the dimensions are on which transactions differ that present differential hazards. It further (2) asks what the attributes are on which governance structures differ that have hazard mitigation consequences. And it (3) asks what main purposes are served by economic organization. Because, moreover, contracting takes place over time, transaction cost economics (4) inquiries [sic!] into the intertemporal transformations that contracts and organizations undergo. Also, in order to establish better why governance structures differ in discrete structural ways, it (5) asks why one form of organization (e.g., hierarchy) is unable to replicate the mechanisms found to be efficacious in another (e.g., the market)" (Williamson (1999, p. 13)).

Organization Failures Framework

Williamson (1975, pp. 20–40) proposed an organization failures framework as a general approach to economic organization ordering. To him, markets as well as more hierarchical forms of organizations are alternative instruments for completing a related set of transactions, and their employment depends on the relative efficiency of each mode. Human behaviour (namely bounded rationality and opportunism) as well as environmental factors (more specifically uncertainty, task complexity and small numbers bargaining) can explain when individuals may decide to by-pass the market environment and resort to more hierarchical modes of organization (and vice versa). These key elements of the organization failures framework can be described as follows:

As already shown above, bounded rationality means that individuals may intend to make a rational decision but their capacity to evaluate accurately all possible decision alternatives is physically limited. To him, this fact poses a problem in situations when transactions have to be managed under complexity and/or uncertainty conditions. Then the given ability of individuals to make fully rational decisions is impeded, because the sheer complexity of a task and/or unforeseeable contingencies avoids them to develop a complete decision tree needed to design fully specified contracts.
Second, opportunism, or "self-interest seeking with guile," has profound implications for choosing between alternative instruments.⁷⁶ He recognized that

⁷⁵Cf. Williamson (1985:387–388, 1999:6–11). Further, Williamson (1999, pp. 18–20) considered these three monographs to form a trilogy. But looking more closely on their theoretical contents, Williamson (1996) seems to be merely a restatement of the theoretical framework in Williamson (1985) so that in the following, a focus is set on the first two.

⁷⁶See Williamson (1975:26–30, 1979:131, 1985:47–50). Further Williamson (1998b) with a focus set on hazards borne out of opportunism.

individuals will sometimes seek to exploit a situation to their own advantage by making self-disbelieved promises. This does not imply to them that all those involved in transactions act all the time opportunistically, rather, it considers that the risk of opportunistic behaviour by someone is just present. Following Williamson (1975:26–29), it is greater when there exists a small numbers bargaining problem, where the counter parties are able to act in a merely strategic manner.77

• Furthermore, if uncertainty and opportunism is present, informational asymmetry as already discussed above influences heavily small numbers bargaining situations, too, in two ways.78 The first is called ex ante opportunism and refers to situations where information is hidden prior to a transaction, which leads to an adverse selection problem. The second informational asymmetry is ex post opportunism which occurs after a transaction and leads to problems of moral hazard, because individuals may have the incentive to act opportunistically to increase their economic welfare when their actions are not directly observable by their counterparts.

All these elements of human behaviour and environmental factors occasionally lead to a market failure, because market transactions are unable to be safe-guarded by fully specified contracts and therefore contracting parties tend to prefer more hierarchical modes of organization. But more hierarchical forms of organizations like a vertical integration can suffer to some extent from the same key elements, too, which was further discussed in detail in Williamson (1975). Revising and broadening this organization failure framework, he focused in later contributions like Williamson (1975, 1975) on the topic of transaction costs and the efficient governance of contractual relations as the outcome of a transaction cost economizing behaviour, which will be discussed in the following.⁷⁹

Characterization of Transactions

According to Williamson (1979:246–254, 1981:555–556, 1985:52–61, 1999:58– 61), transactions principally differ in the dimensions of (1) asset specificity, (2)

⁷⁷In Williamson (1985, pp. 61–63), he referred to this fact as the "fundamental transformation," where a "large numbers" bidding condition by competing rivals on a market is effectively transformed into one of a "small numbers" bilateral monopoly by establishing some contractual relationship.

⁷⁸Williamson (1975, p. 31) initially called this a derived condition of "information impactness."

⁷⁹Strictly speaking, "[. . . ] the object is to economize on the sum of production and transactions costs" (Williamson 1975:245). Further, for this and the following see Schoppe (1995, pp. 48–70), Erlei et al. (1999, pp. 175–193), Gobel (2002:135–149), Martiensen (2000, pp. 271–302) or Richter ¨ and Furubotn (2003, pp. 193–199).

uncertainty and (3) frequency, with the first being the most important one.80 Regarding asset specificity, there are at least four different types to distinguish:81

site specificity, highlighting the immobility of transaction-specific investments.
physical asset specificity, stressing on the non-redeployability of transactionspecific investments.
human asset specificity in connection with training efforts undergone to reach specialized skills and working experience through learning-by-doing.
dedicated assets which are investments in assets undertaken for dedicated usage in the prospect of doing business with a certain customer.

The main reason for asset specificity being critical seems to be, that the value of investments done in advance for other usage is much smaller than for which they have been intended, so that there occurs a lock-in effect. Next, there are two basic sources of uncertainty to consider, which were already discussed in the preceding section: (1) overall state-contingent environmental uncertainty and (2) relationspecific behavioural uncertainty. In both instances, increasing uncertainty usually leads to closer ties between the contracting parties. In addition to this, there are some "interaction effects" occurring in connection with asset specificity. Referring to frequency of transactions, it is quite clear that economies of scale or scope arising from learning-by-doing, division of labour force and/or pooling firm resources are usually considerable high to allow decreasing transaction costs with an increasing amount of transactions.

Governance of Contractual Relations

Assuming, that a sufficient degree of uncertainty is present to pose an adaptive, sequential decision requirement, Williamson (1979:245–254, 1985:72–79) focused on the transaction characteristics of frequency and asset specificity (expressed as the degree of investment idiosyncrasy) to develop a typology of efficient governance structures for contractual relations.⁸² Doing this, he discovered a striking

⁸⁰Discussing merely influences on transaction costs, Picot (1982, pp. 271–273) provided an alternative view by distinguishing between situational ambiguity, environmental uncertainty and frequency of transactions as well as transaction infrastructures offered by law and technological progress.

⁸¹Cf. Williamson (1983:526, 1985:55–56). Masten et al. (1991, p. 9) added this classification "temporal specificity" in connection with tight time schedules, where "[. . . ] an opportunistic supplier may be tempted to seek a larger share of the gains from trade by threatening to suspend performance at the last minute." Further, "[e]ven though the skills and assets necessary to perform the task may be fairly common, the difficulty of identifying and arranging to have an alternative supplier in place on short introduces the prospect of strategic holdup." Later on, Williamson (1991:281, 1998b:707) added a sixth type called "brand name capital," centering around the problem of maintaining reputation as described by Gobel (2002, p. 140). ¨

⁸²Other simplifying assumptions are (1) individuals dealing with each other intent to be in business on a continuing basis, (2) there is no ex ante monopoly of ownership in specialized resources,

			Investment Characteristics
		Nonspecific	Mixed	Idiosyncratic
	Occasional		Trilateral Governance (Neoclassical Contracting)
Frequency	Recurrent	(Classical Contracting) Market Governance	Bilateral Governance	Unified Governance (Relational Contracting)

Fig. 3.1 Illustrative commercial transactions (Source: Williamson (1979:247, 1985:73))

correspondence between these six types of illustrative commercial transactions (see Fig. 3.1) and Macneil's (1974, 1978) three-fold classification of contracting as already discussed in Sect. 3.1.2.2 and matched them with four governance structures of contractual relations as shown in Fig. 3.2:

First, there is market governance as the most efficient one for non-specific transactions of both occasional and recurrent contracting backed by classical contract law. Transactions are discrete, highly standardized, intensify presentiation, and there is no need for safe-guarding relationships against opportunism as long as there are enough alternative contracting partners present.
Second, trilateral governance based on neoclassical contracting is recommended for occasional transactions in connection with mixed or highly specific investment characteristics, because once the principals of such transactions have entered into a contract, they have strong incentives to see the contract through completion. Specialized investments done in advance, unsatisfactory market alternatives and/or switching cost considerations are the main reasons that force contract partners to stick on neoclassical contract law, which allows to relax the strict limitations of discreteness and presentation in order to provide planning flexibility. Quite often, third-parties are engaged in evaluating performance and/or resolving disputes through arbitration so that this governance structure is called a trilateral one.
In bilateral governance, the non-standardized nature of transactions in connection with a medium degree of asset specificity makes primary reliance on

⁽³⁾ the frequency dimension refers strictly to buyer activity in the market and (4) the asset specificity dimension refers to the characteristics of investments made by suppliers, see Williamson (1979:246–247, 1985:72).

			Investment Characteristics
		Nonspecific	Mixed	Idiosyncratic
	Occasional	Purchasing Standard Equipment	Purchasing Customized Equipment	Constructing a Plant
Frequency	Recurrent	Purchasing Standard Material	Purchasing Customized Material	Site-Specific Transfer of Intermediate Product across successive Stages

Fig. 3.2 Efficient governance structures (Source: Williamson (1985, p. 79))

market governance hazardous, while their recurrent nature permits the costs of specialized governance structures to be recovered so that relational contracting between otherwise autonomous contracting parties seems to be an efficient mode of governance structure. Furthermore, both parties have always an incentive to sustain their usually tight social relationships through discrete adjustments rather than to terminate it immediately.

• In the case of a unified governance, highly idiosyncratic transactions are organized within an internal organization structure under relational contracting, because human and physical assets needed for production are extensively specialized and mostly dedicated to a single use. Hence, these assets are less transferable to other uses and the choice of organizational mode turns to favour a governance structure with superior adaptive properties like vertical integration under a single (unified) ownership.

While in this section a focus was set on commercial transactions, Williamson (1979) indicated that these general considerations can be applied to other types of transactions, too – at least after some modification. Moreover, Picot (1982, pp. 273–277) developed a quite similar transaction economizing framework, where he classified transaction characteristics again in three dimensions (frequency, uncertainty and degree of ambiguity), but ended in twelve distinguishable types of transactions matched with a continuum of contractual forms as well as governance structure between market and hierarchy.

Some Applications of Transaction Cost Theory

Many authors were attracted by the TCE framework of Williamson (1975, 1985), so that in the following these Williamsonian ideas were adapted to clarify various real economic situations in the field of transport and logistics.⁸³ Their applications of TCE can be divided into three broad categories: (1) guideline of contractual governance or industry governance structure descriptions, (2) support of outsourcing or make-or-buy decisions, and (3) deliver theoretical background for diverse treatments.

One of the very first applications of TCE in the field of transport and logistics were detailed descriptions of contractual governance like Palay (1984) in the rail freight sector, Jones and Pustay (1988) in the US airline industry between World War I and II, Rahman (1989) in interorganizational vertical relationships between shippers and freight forwarders in international shipping or Pirrong (1993) in bulk shipping markets. These were closely followed by Stahl (1995a, 1995a) discussing corporate governance structures in international transport networks established by German freight forwarding companies or Lunnan and Reve (1995) about the corporate governance structure of Laboremus, a small Norwegian industrial shipping company. Furthermore, surveys of industry governance structure like Casson (1986) focusing on the UK shipping industry or Ojala (1994, 1995) about shipping companies from Finland, Sweden and Norway were conducted, too. Later on, Hubbard (1999, 2001) as well as Lafontaine and Masten (2002) explored contractual governance considerations between carriers and their truck drivers in the US trucking industry. Moreover, Pfohl and Large (1992), Hobbs (1996) as well as Skjøtt-Larsen (1999a, 1999a) discussed contractual governance in third-party logistics relationships and/or freight forwarding and Panayides (2002) offered a transaction cost approach for assessing the effectiveness of appropriate governance structures in organizing intermodal transport systems including a leg by sea. Another contribution is Lopez (2003) conducting a field study among ocean carriers and intermodal transport providers in the United States to map their diverse governance structures in organizing empty container relocation operations. Recently, Halldorsson et al. (2004, 2005) as well as Wathne and Heide (2004) employed ´ TCE as a theoretical base to assess relationship governance issues in supply chain management. All authors found the Williamsonian governance approach to be a viable tool for describing contractual relationships and/or governance structure in the field of transport, logistics and/or supply chain management.

The second category are quite popular applications of TCE to outsourcing and make-or-buy decisions in transport and logistics, like Bretzke (1989, 1998), Kleer (1991, pp. 69–96), Aertsen (1993), Maltz (1993, 1994), Andersson (1997, pp. 87–127), Hanna and Maltz (1998), Bienstock and Mentzer (1999) or Panayides and Cullinane (2002) – only to name a few topical contributions.⁸⁴ Furthermore,

⁸³See Masten (1995) for a collection of early empirical studies and Menard (2000) or Shelanski ´ and Klein (1995) for general overviews.

⁸⁴See Rindfleisch and Heide (1997) for a broad overview mainly focusing on marketing literature, but some of these above mentioned authors were included, too. Further, Hobbs (1996), Stock (1997) as well as Skjøtt-Larsen (1999a) considered make-or-buy decsions as being the foremost application of TCT in the field of logistics or supply chain management.

Stapleton and Hanna (2002) applied TCT to sales force operations in ocean liner shipping, where they explored, whether a technological innovation such as the availability of double stack train hinterland connections impacts the business decision of maritime shipping companies to provide their sales force function internally or rely on third-party intermediaries.

The last category consists of a group of very heterogeneous treatments employing TCE as a theoretical background for further investigation. In a survey about future prospects of small and medium seaports in the Hamburg-Antwerp Range, Hinricher (1993) employed TCT to determine occurring asset specificity and derive an efficient firm size for them. Silverman et al. (1997) analysed the mortality of large motor carriers in the US-American for-hire trucking industry after deregulation on basis of the Williamsonian organization failures framework, whereas Nickerson and Silverman (2003b) focused on inappropriate governance, performance and adoption of motor carriers in the same context. Furthermore, other authors used the theoretical framework of TCE to explain cooperation in institutional arrangements like Becker (1999) observing strategic alliances of air cargo freight forwarders engaged in consolidating shipments.

3.1.3.2 Agency Theory: Agency Costs, Adverse Selection, Moral Hazard and Hold-Up

Agency theory copes with agency costs of firms or other organizations consisting of principals on one side and agents on the other, which arise through the presence of asymmetric information structures. Mainly, there are two distinct approaches to principal–agent problems: a positive and a normative one.85

The positive, less formal approach is closely connected with the works of Jensen and Meckling (1976), Fama (1980) and Fama and Jensen (1983a,b) and can be regarded as a direct answer to the questions raised by Alchian and Demsetz (1972).⁸⁶ Considering firms to be a "nexus of contracts"87, they focused on the broad problem of separating ownership from control as well as its impact on organizational design and emphasized how managers can be disciplined by incentive schemes, external labour markets and capital markets.⁸⁸ According to Jensen and Meckling (1976, p. 308), agency costs are "[. . . ] the sum of the monitoring expenditures by the principal, bonding expenditures by the agent costs, [and] the residual loss."

⁸⁵Following Richter and Furubotn (2003:290), the term "principal–agent problem" was first introduced by Ross (1973).

⁸⁶According to Schoppe (1995, p. 181) or Richter and Furubotn (2003:205) the term "positive agency theory" was first introduced by Jensen (1983) in order to distinguish his work from the "normative agency theory".

⁸⁷Cf. Jensen and Meckling (1976, pp. 310–312) or Richter and Furubotn (2003, p. 201).

⁸⁸See Nilakant and Rao (1994). Further Erlei et al. (1999, pp. 76–106) or Richter and Furubotn (2003, pp. 176–180) for a brief overview.

Table 3.1 Contributions to normative principal-agent theory

Hidden information Moral Hazard	Hidden action Moral Hazard
Contributions
Ross (1973), Stiglitz (1974, 1975a), Mirrlees (197 Grossman and Hart (1983); Harris and Raviv (197 Townsend (1979), Jewitt (1988), Laffont and Mart	9); Holmström (1979); Shavell (1979a);
Hidden characteristics adverse selection	Hidden intensions hold up
Contributions	Contributions
Guesnerie and Laffont (1984); Mirrlees (1976); Rees (1985b); Salanié (1990); Theilen (1996), Laffont and Martimort (2002)	Alchian and Woodward (1987); Grossman and Hart (1986); Klein (1980); Klein et al. (1978); Klein and Leffler (1981), Hart and Moore (1988, 1990), Aghion et al. (1994); Chung (1991); Edlin and Hermalin (2000); Laffont and Martimort (2002); Nöldeke and Schmidt (1995); Rogerson (1992)*
Generalized principal agent problems
Double Moral Hazard	Moral Hazard and adverse selection
Contributions	Contributions
Agrawal (2002); Bhattacharya and Lafontaine (1995); Cooper and Ross (1985); Demski and Sappington (1991); Dybvig and Lutz (1993); Emons (1988); Gupta and Romano (1998); Kambhu (1982); Lal (1990); Lutz (1995); Mann and Wissink (1988); Romano (1994)	Baron (1982); Baron and Besanko (1987); Baron and Holmström (1980); Guesnerie et al. (1988); Myerson (1982); Picard (1987); Sappington (1991); Theilen (1996), Faynzilberg and Kumar (2000); Laffont and Martimort (2002); McAfee and McMillan (1987)

^*Indicates that a contribution falls in more than one category.

Therefore Jensen and Meckling (1976) and Fama and Jensen (1983a) focused on the relationship between agency costs and capital structure whereas Fama (1980) and Fama and Jensen (1983b) focused on the relationship between agency costs and organizational design.⁸⁹

The normative, more formal stream of PAT takes ownership and allocation of firms as given and concentrates on the design of ex ante fully specified contracts and/or dedicated ex post monitoring or reporting systems. This is a flourishing field of research activity as shown in Table 3.1, where some of the main works on several aspects of normative PAT are depicted. ⁹⁰ It seems to be the best way to classify this myriad of contributions by dividing them into works on: ⁹¹

Hidden characteristics leading to adverse selection.

^89

Cf. Eisenhardt (1989, pp. 59-60) or Erlei et al. (1999, pp. 168-169).

^90

An asterix "*" indicates that a contribution falls in more than one category. Further there exists a bulk of reviews on normative PAT like MacDonald (1984), Rees (1985a,b), Arrow (1986), Spremann (1987, 1988, 1990), Levinthal (1988), Eisenhardt (1989), Breid (1995), Kah (1993), Theilen (1996), Saam (2002) or Laffont and Martimort (2002).

^91

Cf. Spremann (1990), Breid (1995), Wolff and Neuburger (1995, pp. 81–83), Saam (2002, pp. 28–35), Picot et al. (2002:85–148) or Göbel (2002, pp. 100–123).

Hidden action leading to moral hazard.
Hidden information leading again to moral hazard.⁹²
Hidden intentions leading to hold-up.

Furthermore, more generalized principal–agent problems like double moral hazard, the occurrence of moral hazard and adverse selection at the same time and multiple principal and/or agent situations were modelled, but to discuss them here is beyond the scope of this work.93 But some of these more generalized principal– agent problems will be discussed in Sect. 3.2.4.

Adverse Selection

Focusing on ex ante information asymmetry, there is an adverse selection problem, because the principal cannot observe contract-relevant characteristics of his agent and so he is unsure upon contracting with him at all. Remedies like signalling by the agent (Spence (1973, 1974)), screening by the principal (Stiglitz (1975b)) and the establishment of self-selection mechanisms (Ross (1973), Arrow (1986)) are usually proposed to alleviate the given ex ante information asymmetry. Furthermore, an alternative approach to adverse selection is the alignment of interests of the agent to the principal's ones like in the case of retailers providing guarantees or restitutions to their customers.⁹⁴

Moral Hazard

Moral hazard of the hidden action type occurs if there is an ex post information asymmetry, where the principal can observe some outcome of his agent, but he cannot see the actual effort level of him. The information asymmetry can be costly alleviated either by extensive monitoring efforts by the principal or reporting efforts by the agent. Other approaches are closing contracts including incentive schemes and/or self-selection mechanisms or the agent makes a hands-tying agreement in advance (Schelling (1956), Kronman (1985)).

In the case of moral hazard of the hidden information type, there is an ex post information asymmetry about the state of nature which comes together with resource plasticity and opportunistic behaviour of "shirking" agents (Alchian and

⁹²Notably, Arrow (1985, pp. 38–40) provided in a quite influential contribution a somehow puzzling taxonomy mixing up the problem of (ex post) hidden information with (ex ante) adverse selection.

⁹³In addition to this, Gobel (2002, pp. 105–109) or Richter and Furubotn (2003, pp. 290–294) ¨ named other variants and extensions of simple principal–agent problems including alternative incentives, more agents and relative outcome measurement, multi-task problems, common agency, hierarchical structures or dynamic modelling.

⁹⁴Giving such a promise, the agent's self-interest is then delivering performance as expected by the principal with his reputation being at risk, see Picot et al. (2002, pp. 92–93).

Woodward (1987)). The agent is considered to have more information about the state of nature after contracting and therefore he can adjust his effort level accordingly. Again, costly monitoring by the principal or contracts with incentive schemes and hands-tying agreements by the agent are viable solutions.

Hold-up

In a hold-up situation, (1) ex ante commitments like initial investments are made by the agent non-observable by the principal result in an ex post dependency situation for the agent and/or (2) not fully specified contracts closed between an agent and a principal raise opportunities for the agent to extract specialized quasi rents.⁹⁵ Until recently (e.g. Laffont and Martimort (2002) or Gibbons (2005)), this sort of principal–agent problem has been widely negated in Anglo-Saxon PAT and therefore mostly unrecognized in mainstream literature there.96 Some authors like Breid (1995, pp. 824–825) argued that these hold-up problems arise from a mix between ex ante hidden characteristics and ex post hidden information, whereas others like Picot et al. (2002, pp. 89–90) saw no direct information asymmetry between principal and agent at all but against third-parties which lead to nonsanctionable opportunism. Remedies for the first situation of hold-up are signalling by the agent, building up reputation (Klein and Leffler (1981)), exchanging hostages (Williamson (1983), Kronman (1985)), developing renegotiating schemes (Chung (1991), Aghion et al. (1994), Noldeke and Schmidt (1995), Edlin and Hermalin ¨ (2000)) or a vertical integration solution (Klein et al. (1978)).⁹⁷ Principally, there are three viable solutions in the second hold-up situation as described above: (1) reconstruction of a corresponding complete contract by court, (2) transformation in a complete contract by assignment of residual claims to one party, which undergoes transaction-specific investments or (3) designing of self-enforcing contract schemes.⁹⁸

⁹⁵Cf. Rogerson (1992). Laffont and Martimort (2002, pp. 370–372) label the first "non-verifiability and contracting" and the second "non-verifiability and bargaining." Further, Klein (1998) considered three factors necessary for the occurrence of a hold-up, namely (1) specific investments done in advance, (2) an imperfect contract not covering every possible contingency and (3) engaging in a hold-up is wealth-maximizing, because short-run gains outweigh long-run costs when doing so.

⁹⁶See e.g. reviews by Arrow (1986), Spreman (1986), Levinthal (1988), Eisenhardt (1989), Kah (1993, p. 21–23), Petersen (1993), Barney and Hesterly (1996, pp. 124–129) or Theilen (1996, pp. 1–4). But contributions like Klein et al. (1978), Alchian and Woodward (1987, 1988), Chung (1991) or works under the label "incomplete contracts" (Grossman and Hart (1986), Hart and Moore (1988, 1990)), essentially describe such hold-up situations in principal–agent relationships, despite they are commonly not considered to be part of PAT.

⁹⁷Cf. Picot (2002, p. 95).

⁹⁸Cf. Martiensen (2000, pp. 361–367) or Schmitz (2001). This stream of literature is often subsumed under label of "incomplete contracting" initially assessed by Grossman and Hart (1986), Hart and Moore (1988, 1990), who later proposed to provide a 'property rights theory of the firm'. See also Schwartz (1998), Tirole (1999), Schmitz (2001) or Richter and Furubotn

Some Applications of Agency Theory

Although so many and diverse treatments of principal–agent problems exist in other fields of economic activity (like insurance markets99, real estate brokerage100, professional services101 or marketing102), only quite a few recent contributions dedicated to the field of transport and logistics are known to the author.103

First, some authors assessed to integrate TCT and PAT in order to cope with principal–agent relationships. Kowtha (1997) suggested performance ambiguity and employee skill level to be significant predictors of control in employing crews from different nationalities in merchant shipping, Fernandez et al. (2000) discussed ´ quasi-integration (subcontracting of owner-operators) versus vertical integration (employing own vehicles and drivers as employees) in European LTL-trucking, whereas Nickerson and Silverman (2003a) explored the trade-off a US-American for-hire trucking carrier makes in its choice between contracting owner-operators and employing company drivers on company-owned vehicles. All three come to the conclusion, that agency as well as transaction costs matter in the trucking industry. Furthermore, Logan (2000) as well as Halldorsson and Skjøtt-Larsen (2003) focused ´ on long-term outsourcing relationships in transport and/or logistics and proposed that there is a need to look from both sides of the contract which helps to provide and support an environment of trust, whereas Sch¨uller (2003) employed PAT (as well as TCT and PRT) to derive managerial implications for integrated transport chains in the air cargo industry.

Second, Hubbard (1998, 2000) as well as Baker and Hubbard (2000, 2002, 2003) observed the impact of on-board information technologies like trip recorders and electronic vehicle management systems on monitoring driver's performance, job design and asset ownership patterns in the US-American trucking industry.

Moreover, Mitroussi (2003) discussed third-party ship management as a separation of ownership and management in the shipping context from a PAT point of view. Within his theoretical framework, he examined the relevant factors contributing to the recent emergence and evolution of a third-party ship management sector.

^(2003, pp. 301–305) with recent reviews about contributions dedicated to their formal models of incomplete contracting. Further, they provided some conceptual criticism.

⁹⁹See treatments on moral hazard and/or adverse selection in insurance markets e.g. by Spence and Zeckhauser (1971), Wilson (1977), Shavell (1979b) or Arnott and Stiglitz (1988, 1991).

¹⁰⁰See contributions e.g. by Anglin and Arnott (1991), Arnold (1992) or Yavas (1995).

¹⁰¹See e.g. Sharma (1997) or Saam (2002).

¹⁰²See Eisenhardt (1989), Bergen et al. (1992) or Lassar and Kerr (1996) with extensive reviews about this topic.

¹⁰³See Zenios (2004, p. 118), considering PAT to be generally underrepresented in the operations management literature. Stock (1997) provided some evidence to this, reviewing four major peerrefereed journals with a logistics focus from 1980 to 1996. He found no contribution applying directly to PAT, but he considered it to provide further insights in principal–agent relations on the basis of contractual agreements.

Last but not least, a growing stream of contributions considered PAT to be a good theoretical base to approach interorganizational relationships in supply chain management. Discussing principal–agent problems in supply chain management, Kaluza et al. (2003) developed a taxonomy of principal–agent relationship problems, agency costs connected with them and workable solutions as shown in Table 3.2. For them, it seems to be straight forward, that different types of agents cause different principal–agent problems and therefore different approaches have to be followed. Furthermore, Halldorsson et al. (2004, 2005) employed PAT (together ´ with TCT) to develop an institutional perspective of supply chain management in order to "[. . . ] explain and understand this particular form of inter-organizational arrangement" (Halldorsson et al. (2004:248)) and authors like Zsidisin and Ellram ´ (2003) or Agrell et al. (2004) used PAT to model explicitly information asymmetry in multi echelon inventory problems.

3.1.4 Conclusions and Some Comparisons

As it was suggested in the last sections, TCT and PAT are distinguishable theory approaches of NIE, but closely interrelated in their theoretical core.¹⁰⁴ Basically, although telling different stories they aim to solve quite the same sort of problems: TCT focuses on transaction costs and the ways to economize on them by choosing efficient governance structures in a discriminating way whereas in PAT, the object of interest are agency costs borne out of information asymmetry between contracting parties that have to be resolved by special incentive-compatible arrangements. In TCT, the problems causing costly distortions are called opportunism (ex ante and ex post), whereas in PAT they are considered as problems of adverse selection, moral hazard and/or hold-up. First, it is evident, that ex ante opportunism can be easily restated in terms of an adverse selection problem because both deal with likely the same pre-contractual situation. Second, Alchian and Woodward (1988, p. 67) claimed that the various PAT situations of moral hazard or hold-up are subsumed under the label of ex post opportunism in Williamsonian TCT without making further distinction between them.¹⁰⁵ But the very difference between these two concepts is, that PAT usually employs a neoclassical rationality concept that allows to write ex ante full specified contracts whereas bounded rationality and subliminal opportunism of individuals prevent them to do this in TCT. Furthermore, TCT can be regarded as a more comprehensive approach than PAT, because it is not only centred around the impact of imperfect information structures, but takes other transaction characteristics into consideration like asset specificity having the

¹⁰⁴For this and the following see Barney and Hesterly (1996, p. 124), Kowtha (1997, pp. 54–55), Martiensen (2000, p. 353–355), Gobel (2002, pp. 64–65 and 133–138) or Picot et al. (2002, p. 64). ¨ 105See Alchian and Woodward (1988, pp. 67–70) criticizing exactly this in their book review of Williamson (1985).

	Cooperative relationships		Non-cooperative relationships
Agent type	Parts producer	Network partner	Employees	mous supplier Anony	Technology specialist
Description	subcontracting Outsourcing,	m or supply chain partnership Syste	In-house Production	Arms-length-buyer- supplier	Independent specialist, not interested in
				relationship	cooperation
ms minant proble Do	Hidden action	Hidden characteristics,	Hidden action	Hidden characteristics	Hidden action, hidden
		action, intensions			intension
minant agency Do	Monitoring costs	Bonding and monitoring	Monitoring costs	Residual loss	Residual loss
costs		costs
minant solutions Do	Monitoring	mes Incentive sche	Monitoring and	Monitoring, abandoning	Monitoring, holding
			mes incentive sche	of relationship is	contact, always
				sub-performing	searching for
					alternatives

most decisive impact, even in absence of asymmetric information.¹⁰⁶ Moreover, both incentive (measurement and specification) as well as coordination (searching and matching) problems occurring in an imperfect world with extensive division of labour are covered by TCT at the same time. Another rather technical aspect is that PAT (especially the normative part) relies heavily on mathematical exposition, but Williamson (1991:281–286) later restated his initial solely verbalized reasonings of TCT in a more formal way, too. In addition to this, some authors considered the theory of incomplete contracts (in the sense of Grossman and Hart (1986) or Hart and Moore (1988, 1990)) as another formal expression of transaction costs economizing behaviour, but Holmstrom and Roberts (1998) as well as Williamson ¨ (2000, p. 604–607) provided striking arguments against such a comparison.

3.2 Market Microstructure and Intermediation Theory

In the previous sections, several facets of actual business like the governance of contractual relationships, the role of transaction costs and the consequences borne out of information asymmetries (namely adverse selection, moral hazard and hold-up) were theoretically reflected from an institutional point of view and some applications in the context of transport or logistic chains were given. But despite all these theories giving useful insights how economic environment imperfections affect business operations, establishment and running of markets to allocate goods and/or services or the important economic role played by intermediaries on markets seem to be rather unsolved. Therefore in the following, a closer look on market structure and intermediaries' role in markets is provided, based on MMT and ITF.

MMT itself can be shortly defined as the study of intermediation and the institution of exchange on financial markets.¹⁰⁷ Daniel F. Spulber recently expanded this approach of MMT on financial markets to other market institutions, leading to an "intermediation theory of the firm" backed by theoretical findings from NIE, IOT as well as NCE.¹⁰⁸ Following Spulber (1999:xiii),

"[t]he intermediation theory of the firm provides an explanation for why firms exist. Firms are formed when intermediated exchange provides greater gains from trade than direct exchange between consumers and suppliers. The theory also helps to explain how markets work. Markets reach equilibrium through strategic pricing and contracting by intermediaries."

¹⁰⁶See Gobel (2002, pp. 134–136) for a more detailed discussion. ¨

¹⁰⁷Cf. Spulber (1996b). A more elaborated definition was provided by O'Hara (1995, p. 1): "Market microstructure is the study of the process and outcomes of exchanging assets under explicit trading rules. [. . . ] Market microstructure research exploits the structure provided by specific trading mechanisms to model how price-setting rules evolve in markets."

¹⁰⁸See Spulber (1999, pp. 13–21) for a comparison of his intermediation theory approach with NIE, IOT and NCE.

More precisely, he considered intermediaries to offer specific advantages over direct exchange like (1) reducing transactions costs, (2) pooling and diversifying risks, (3) lowering costs of searching and matching, (4) alleviating adverse selection, (5) mitigating moral hazard and opportunism and (6) supporting commitment through delegation. 109 Some of these advantages are derived from some superior coordination abilities, others are borne out of economies of scale and/or scope or stem from their interest in longevity of relationships and economic incentives to build up reputation

Reviewing the literature in the field of MMT and ITF, at least four basic functional roles of an intermediary can be distinguished (see Tables 3.3 and 3.4

Table 3.3 Contributions about intermediaries: Part 1
-------------------------------------------------------------

Table 3.3 Contributions about intermediaries: Part	1
Market makers, specialists, dealers, marketers or merchants (Dyadic relationships)	Matchmakers or Brokers (Triadic relationships)
Contributions focused on Financial Intermediation:	Contributions focused on Financial Intermediation:
Gurley and Shaw (1955, 1956), Baumol (1965), Demsetz (1968), Garman (1976); Stoll (1978); Tinic (1972), Bryant (1980), Ho and Stoll (1980, 1981), Amihud and Mendelson (1980); Bryant and Wallace (1980); Copeland and Galai (1983); Diamond and Dybvig (1983); Glosten and Milgrom (1985); Grossman and Miller (1988); O'Hara and Oldfield (1986), Gorton and Pennacchi (1990), Easterwood and Morgan (1991); Laffont and Maskin (1990), Beneviste et al. (1992); Bhattacharya and Yavas (1993), Freeman (1996a, 1996b), Holmström and Tirole (1998); Yavas (2001)	Baumol (1965), Demsetz (1968), Easterwood and Morgan (1991), Yavas (2001)
Other Contributions:	Other Contributions:
Garella (1989); Hackett (1992); Lim (1981); Spulber (1985); Stahl (1988); Wernerfelt (1988)* Voyag (1992)* Gebrig (1993)	Anglin and Arnott (1991); Myerson and Satterthwaite (1983); Rubinstein and Wolineky (1987); Vinger (1981) Yeve

(1988)*, Yavas (1992)*, Gehrig (1993), Havila (1993a, 1993b, 1996)*, Spulber (1996a), Fingleton (1997a, 1997b), Agrawal and Seshadri (2000), Johri and Leach (2002), Rust and Hall (2003)*

Wolinsky (1987); Yinger (1981), Yavas (1992, 1994, 1996a)*, Hackett (1992)*, Havila (1993a, 1993b, 1996)*, Bloch and Ryder (2000); Cosimano (1996); Gehrig (1996); Margolin (1993); Rauch and Watson (2002); Rust and Hall (2003); Yavas et al. (2001)* Cummins and Doherty (2006)

^*Indicates that a contribution falls in more than one category.

¹⁰⁹Cf. Spulber (1999:xiii).

Table 3.4 Contributions about intermediaries: Part 2

Hidden Characteristics Issues
Information Producers
Contributions
Campbell and Kracaw (1980); Chan (1983); Etgar and Zusman (1982); Leland and Pyle (1977); Ramakrishnan and Thakor (1984), Millon and Thakor (1985), Boyd
and Prescott (1986), Admati and Pfleiderer (1986, 1990)
Guarantors or Certificators
Contributions
Biglaiser (1993); Chu and Chu (1994); Marvel and McCafferty (1984), Biglaiser and Friedman (1994, 1999); Li (1998), Lizzeri (1999), Garella and Peitz (2000)

Indicates that a contribution falls in more than one category.

for a list of selected contributions in chronological order, an asterisk indicates contributions fitting to more than one category):¹¹⁰

A market maker (specialist, dealer, marketer or merchant) in dyadic relationships, purchasing goods and/or services from market takers (sellers, suppliers or producers) for resale to their customers (buyers).
A matchmaker (broker) in triadic relationships, helping buyers and sellers to get together for executing market transactions.

¹¹⁰Of course, this enumeration is by far not complete, but these contributions can be regarded as the most topical ones providing original thoughts and ideas. Most of the contributions not cited here fall in the main categories of unpublished working papers, replications and/or extensions to existing ideas initially presented by other authors.

• An information producer, guarantor and/or certificator, generating a business out of information asymmetry of the adverse selection type.
An agent employed for specific tasks like delegated expertise, bargaining, contracting and/or monitoring.

In the following, the focus is set first on each of these four categories one after another presenting a selection of topical contributions, before again some applications in the field of transport and logistics are given.111 For brevity in explanation, only basic assumptions, stylized facts and major conclusions for all of these more or less theoretical treatments as shown in Tables 3.3 and 3.4 are provided.

3.2.1 Market Makers, Specialists, Dealers, Marketers and Merchants

Following Spulber (1996a), there are at least three types of individuals acting on a market: customers (buyers), market takers (sellers, suppliers or producers) and market makers (specialists, dealers, merchants or marketers).¹¹² Market takers take price signals and market institution as given, whereas market makers create and/or run markets either as firms setting prices in a given market institution or being itself the market institution by coordinating transactions between all participants of a particular market.¹¹³ Furthermore, they always take title on goods and/or services bought in from their suppliers and resell them to their customers. Doing this, they make a bargain out of the bid-ask spread between their purchase and their selling prices. These intermediaries are usually named specialists or dealers on financial markets and operate as marketers or merchants like retailers, wholesalers or used car dealers on product markets. But apart from this simple price setting and intermediate trading function in dyadic relationships between both sides of the market, market makers offer even more service to markets, as will be shown in the following.

¹¹¹A more detailed overview of these contributions taken into consideration for this work and others not mentioned explicitly here can be found in reviews of Stoll (1985), Gertler (1988), Yanelle (1989), Yavas (1994b), O'Hara (1995), Easley and O'Hara (1995), Allen and Santomero (1998), Spulber (1999), Benjamin et al. (2000a,b), Madhavan (2000), Gorton and Winton (2002) or Stoll (2003).

¹¹²See O'Hara (1995, pp. 8–12) for a quite similar view, but she includes brokers as a fourth main type and asserts that some individuals acting on a market can belong to two of these categories at the same time.

¹¹³Other authors like O'Hara (1995) use the term "trading mechanism" instead of "market institution," which "[. . . ] can be viewed as a type of trading game in which players meet (perhaps not physically) at some venue and act according to some rules." (O'Hara (1995, p. 8)).

3.2.1.1 Early Contributions on Financial Intermediation

Looking at existing financial markets, the assumption from NCE of competitive markets where a fictitious Walrasian auctioneer allows suppliers to sell and customers to buy any amount they desire at a given market price resulting from a tatonnement process is in fact quite unrealistic. ˆ ¹¹⁴

Probably one of the first contributions highlighting the role of financial intermediaries is Gurley and Shaw (1955, 1956). Looking the saving-investment process supported by financial markets, they defined the business of a financial intermediary as lending at one stratum of interest rates and borrowing at a lower stratum which allows individuals to obtain external financing.¹¹⁵ Furthermore, they draw a distinction between these sort of non-monetary intermediaries engaged in borrowing and lending money and banks as monetary intermediaries offering bank deposits and money transaction systems as a mean of payment.

Baumol (1965) and Demsetz (1968) explored the business of specialists as market institutions at stock exchanges. To Baumol (1965, p. 14), "[t]he specialist's basic function is to bring buyers and sellers together." Furthermore, he distinguished between (1) the specialists's brokerage function facilitating market transactions between buyers and sellers and (2) the specialist as a dealer trading for his own account if needed.¹¹⁶ Serving this second function, a specialist sets market prices and acts as a residual buyer or seller on demand in order to reach a market clearing equilibrium. Moreover, the specialist stabilizes financial markets by providing a continuous source of supply and a continuous demand in times of low market activity or trading against temporary upward or downward swings of market prices. Demsetz (1968) investigated empirically effects of trading volumes on transaction costs at the New York Stock Exchange. Doing this, he observed that "the askbid spread is the mark-up that is paid for predictable immediacy of exchange in organized markets; in other markets it is the inventory mark-up of retailer and wholesaler."¹¹⁷ In another contribution, Tinic (1972) focused on liquidity service issues, adding structural determinants of demand for liquidity services and the economics of the specialist function to Demsetz's (1968) transaction cost approach.

All in all, these shown above and many other early contributions to financial intermediation were rather descriptive in nature and therefore failed to model

¹¹⁴See Baumol (1965, pp. 10–12), Yavas (1994b), O'Hara (1995, pp. 3–5) or Spulber (1996a). According to O'Hara (1995, p. 7), the price fixing process at the London Gold Market resembles in many ways to such a Walrasian price mechanism with a series of preliminary auctions until demand equals supply at some equilibrium price, but actually no trading occurs there.

¹¹⁵Cf. Gurley and Shaw (1956, pp. 259–260).

¹¹⁶Cf. Baumol (1965, pp. 14–16). Further Demsetz (1968, pp. 37–38) for a similar description of the specialist's intermediary role.

¹¹⁷Demsetz (1968, pp. 35–36).

explicitly the underlying market mechanism.118 But they laid ground for further, more formal investigation of financial markets as a micro economic phenomena.

3.2.1.2 Transaction Cost Reduction, Risk Bearing and Immediacy Service

Closely after a pioneering study conducted by Garman (1976), researchers focused on matching problems in financial market processes where a market maker sets prices to balance supply and demand of goods or services traded across time which in turn leads to a reduction of transaction costs to all market participants.¹¹⁹ Furthermore, as such a market maker usually takes title on goods or services traded, he bears risks and offers at least some immediacy service, so that these sort of MMT models are often also called inventory models.¹²⁰

Garman (1976) focused on the nature of order flow in determining equilibrium market prices on financial markets by introducing market uncertainty due to a Poisson process of buy or sell orders. Based on this stochastic market process approach, he modelled a risk-neutral monopolist intermediary setting prices in a dealership market for securities and contrasted it with a double auction market model of a stock exchange where market maker trades on temporarily fluctuations of supply and demand. Although his approach was considered in the following to be too mechanistic and full of simplifying assumptions to be applicable to actual market settings, it had a great influence on preceding work in the field of MMT.¹²¹ Amihud and Mendelson (1980) developed Garman's (1976) dealership market model further and considered a specialist adjusting his price-setting based on inventory levels to maximize his profit per unit of time in a continuous-time, multiperiod setting.122 Deriving an optimal policy, they showed, that prices set by the specialist are monotone decreasing functions on the stock at hand and the resulting bid-ask spread is always positive. Furthermore, the optimal policy of the specialist implies the existence of a preferred inventory position.

Another, more detailed dealer-based approach to market maker modelling was provided first by Stoll (1978). He analyzed directly the dealer's price-setting problem where he or she is one out of several risk averse traders in a financial market, who is willing to alter his own portfolio away from desired holdings to accommodate the trading desires of other market participants. The resulting bidask spread reflects then the elasticity of demand and supply and the dealer's degree

¹¹⁸For a review of these early contributions, see Benston and Smith (1976), Cohen et al. (1979) or Stoll (1985).

¹¹⁹See Benston and Smith (1976), Cohen et al. (1979), O'Hara (1995, pp. 13–55) for a more comprehensive overview of the literature about market making on financial markets.

¹²⁰Cf. O'Hara (1995, pp. 13–14) or Yavas (2001).

¹²¹See O'Hara (1995, pp. 14–23) for further critical comments.

¹²²Further, see Spulber (1999, pp. 31–58) for some other extensions to the original dealership market model of Garman (1976).

of risk aversion. Ho and Stoll (1981) extended this original two-stage model of Stoll (1978) to a multi-period setting in which both order flow and portfolio returns are stochastic. There, the dealer faces demand following a continuous time Poisson process and is subject to return risk on his stock as well as the rest of his portfolio. The bid-ask spread then tends to increase with longer planning horizon of the dealer because adding more periods provides him more opportunities for price adjustment. But at the same time, his risk position increases, which leads to a need for greater compensation and therefore the bid-ask spread is widened. Furthermore, Ho and Stoll (1980) outlined the problem of determining the reservation prices of dealers operating in a competitive multiperiod market environment who are subject to return and transaction uncertainty. One of their results is, that each dealer's pricing strategy in this market setting depends not only on his own current and expected inventory position as well as his other characteristics, but also on the current and expected inventory and the other characteristics of his competitors.

Furthermore, O'Hara and Oldfield (1986) considered the dynamic pricing policy of risk averse dealers receiving both market as well as limit orders and facing inventory value uncertainty in a discrete-time, multiperiod trading environment.123 Doing this, they showed, that the resulting bid-ask spread can be decomposed into three parts: (1) a portion for the known limit orders, (2) risk-neutral adjustment for expected market orders and (3) a risk adjustment for market and inventory uncertainty. Therefore, a risk-averse dealer may set a smaller bid-ask spread than a risk-neutral specialist with his inventory position affecting both the placement and size of the spread.

Bhattacharya and Yavas (1993) considered a dealer to be a "trader of the last resort" if traders can bypass him and trade directly with each other. There, potential traders approach him only when they cannot find a better match somewhere else. An increase in the quoted bid-ask spread intensifies search among traders and may result in more order imbalances which in turn attracts other dealers even with lower inventory holding costs to rise their bid-ask spread, too. Yavas (2001) developed this idea further studying explicitly the coordination role of a specialist in a simple search market model in which traders can either conduct direct search for each other to trade or go to the specialist. The coordination role of the specialist is due to the fact that he provides an immediacy service, with traders not dealing with him only if they expect to get a higher payoff from their own search efforts. However, he noted, that in the market equilibrium, the specialist never has a nonzero inventory position, thus he never has to execute his ability to sell (buy) a unit without having to buy (sell) another one and therefore the sheer presence of his ability alone is enough to induce coordination. And this result is robust to many relaxations of assumption as long as the specialist can provide such an immediacy.

¹²³Market orders are orders to buy and sell for immediate execution and limit orders are orders to buy or sell at some prespecified price. As O'Hara (19995:37) notes, models with a stochastic order flow like Amihud and Mendelson (1980) or Ho and Stoll (1981) allow only market orders.

3.2.1.3 The Impact of Asymmetric Information on Market Making

Another stream of contributions on financial market intermediation regarded the trading process as a game involving traders with asymmetric information leading to problems of adverse selection.¹²⁴ Bagehot (1971) first proposed to discuss the economics of market making by introducing three kinds of traders: (1) informed traders possessing insider information, (2) liquidity-motivated traders who have no special information but merely want to trade cash against securities (or vice versa) and (3) traders acting on information which they believe has not yet been fully discounted in the market but which in fact has. To him, a market maker always bears losses in trading with informed traders which have to be recovered from trading with the other two forms of uninformed traders.

Following Bagehot (1971), Copeland and Galai (1983) and Glosten and Milgrom (1985) examined formally how market makers establish rational bid and ask prices, taking into account that some traders have superior private information.125 The basic principle is that the bid price and ask price should reflect all available market information, including past, current and pending transactions. Thus the bid price (ask price) is a rational price conditional on a sale to (on the purchase from) the intermediary. Since a sale to the intermediary has a higher probability of being placed by an informed trader (insider) with superior private information in order to make profit than by a liquidity trader (outsider) who has no further private information and trades only for liquidity reasons, it implies a signal to an increased probability of a stock price decline. Conversely, a purchase from the intermediary might be a signal to an increased probability of a stock price increase. In this environment, the market maker establishes a bid-ask spread based on the average information contained in a sale to him or a purchase from him.¹²⁶ Doing this, he suffers losses from trading with informed traders, which have to be recovered by trading with liquidity traders. Therefore, the profitability of a market maker depends heavily on the returns from his immediacy service to liquidity traders, which must be sufficient to offset losses from dealing with informed traders.

Laffont and Maskin (1990) examined a model where a market maker possessing considerable market power is an informed trader while all other traders are uninformed so that he can set a bid-ask spread based on his private information.¹²⁷ If the possible variation of returns to a risky asset privately known to him is not

¹²⁴See O'Hara (1995, pp. 53–178) for a comprehensive overview on intermediation issues on financial markets with asymmetric information, including strategic trader modelling first introduced by Kyle (1984, 1985).

¹²⁵See Stoll (1985:78), O'Hara (1995, pp. 53–66) or Spulber (1999, pp. 203–211) for a more comprehensive review of these models.

¹²⁶In Copeland and Galai (1983), this bid-ask spread depends on the relative proportion of insiders and outsiders in a one-period model whereas Glosten and Milgrom (1985) model a dynamic setting of bid-ask spreads, where the market maker updates his beliefs and adjusts his bid-ask spread accordingly.

¹²⁷See Spulber (1999, pp. 215–219) for a brief overview.

too large, the market maker has always an incentive to conceal that information rather than revealing it through his price setting, leading to a pooling equilibrium. Although this uninformativeness of price setting by the market maker, all other traders are better off than in any separating equilibrium, where equilibrium prices aggregate information effectively.

Beneviste et al. (1992) showed that long-standing relationships between floor brokers as traders and risk-neutral specialists can mitigate the effects of asymmetric information.¹²⁸ Moreover, a specialist who actively attempts to differentiate between informed and uninformed traders and has the power to sanction traders exploiting private information can achieve separating equilibria that pareto-dominate pooling equilibria. So they come to the conclusion, that a specialist system can be regarded as a market mechanism that improves both the welfare of market participants at the stock exchange and the terms of trade by reducing incentives to exploit informational asymmetry.

3.2.1.4 Liquidity Service and Consumption Smoothing

According to Grossman and Miller (1988), market liquidity is determined by the demand and supply of immediacy, which is provided by market makers through their continuous market presence and their willingness to bear risk by acquiring and holding stock during the time between the arrival of selling and corresponding buying orders.129 Doing this, the market maker charges a risk premium to a trader who wants to sell immediately to him by offering him a price that is not uncertain, but on average lower than the seller could expect from a postponed sale. This margin can be regarded as the cost of liquidity service provision.

Focusing on banks as financial intermediaries, Gorton and Winton (2002) motivated three other forms of liquidity service provision: (1) establishing a payment system for spatial separated buyers and sellers which cannot meet each other face to face and transact at a central place, (2) offering bank deposits reducing trading losses for liquidity traders faced by other traders with private information or (3) providing credit lines as a form of contingent liquidity if moral hazard limits the effectiveness of transactions on financial markets between traders with excess liquidity and traders that need liquidity.

The first view of banks as liquidity providers concerns the non-monetary role of banks in the payment system. Freeman (1996a,b) set up a model where transaction partners are spatially separated and the timing of transactions is such that they

¹²⁸In this context, floor brokers are agents acting on behalf of their customers and represent therefore market orders driven either by liquidity needs or private information.

¹²⁹See also O'Hara (1995, pp. 216–223). Another explanation of market liquidity is asserted by Pagano (1989) considering liquidity not to be a measure of the cost of tasking risk and waiting, but rather a function of the scale of trading. For him, liquidity arises endogenously as a result of individual trader's decisions to trade more likely in markets that have more traders than in markets that have fewer traders, see O'Hara (1995, pp. 223–233).

cannot simultaneously trade at a central location. The real problem is, that buyers wish to obtain goods from sellers, but have neither goods they want to offer them in barter-like exchange nor any money for instant payment, though later they might be able to sell their goods in exchange for money. So, buyers issue IOU's – promises to pay at a central location later on with fiat money – to the sellers.¹³⁰ Fiat money is used to settle this debts, but money and private debt coexist. At the central clearing location it may happen that all creditors and debtors (or their respective agents) arrive simultaneously, in which case clearing occurs directly. If arrival is not simultaneous, however, payment can take place through a clearing house.¹³¹ The clearing house accepts money in payment of IOU's and pays off immediately IOU's presented. However, if creditors arrive first, then the clearing house must have some liquidity stock to pay them before the debtors arrive. Furthermore, a basic point of the author is to develop the notion of the clearing house issuing its own IOU's like private bank notes, that can circulate and be redeemed for fiat money later.

Gorton and Pennachi (1990) provided another rationale for liquidity provision by banks showing that bank deposits can mitigate losses of uninformed traders associated with information asymmetry on financial markets. They argue that splitting the cash flows of an underlying portfolio in order to create debt and equity can create a liquid security, namely the debt, which can be used by uninformed traders to sell securities to satisfy their liquidity needs. If the debt provided is riskless, then there can be no information advantage that other traders could possess. These characteristics allow uninformed traders to trade without systematically loosing money to informed ones. Finally, financial intermediaries are the natural entities to create such securities, as they hold diversified portfolios of assets.

In Holmstrom and Tirole (1998), a bank creates a security called credit line, ¨ which is valuable because it is state contingent. Instead of holding cash or securities from other firms at stock as a liquidity reserve a firm might open a credit line at a bank in advance, which is only drawn on when there is a sudden need for liquid financial resources at once. Capital market securities cannot replicate such statecontingent payoff, so that they can make a business out of this third form of liquidity provision.

Building on previous work by Bryant and Wallace (1980) and Bryant (1980), Diamond and Dybvig (1983) developed the idea, that liquidity provision and consumption smoothing are important intermediation aspects of banking services.132 For them, bank liabilities do not only function as a medium facilitating financial transactions. Moreover, banking firms offer some special insurance coverage to

¹³⁰The abbreviation IOU stands for "I owe you" and IOU's can be either promissory notes with the buyer as the debtor being drawer and drawee at the same time backed by an aval credit or banker's cheques issued by the buyer's bank on behalf of them, see Shubik (1991) or Kummer et al. (2010, pp. 227–235) with further references.

¹³¹See Freeman (1996a, pp. 106–111) for a detailed treatment of the clearinghouse function.

¹³²See also Gorton and Winton (2002, pp. 16–21) an overview of other consumption-smoothing related models of banking and financial intermediation.

their customers against temporarily running out of liquidity. Individuals are facing uncertainty consumption needs which are often inconsistent with payoffs from irreversible investments done before, but banks offering demand deposits give them flexibility over the timing of withdrawal from deposits in endowment opened beforehand. Offering such transformation services from illiquid assets into more liquid liability payoffs, banks are always in danger of going bankrupt in case of bank runs.¹³³ This exemplifies the importance of strict governmental regulations in the banking sector, like minimum reserve holding obligations, coverage of deposit insurances and the role of state-owned central banks as a "lender of the last resort."

3.2.1.5 Market Making on Product Markets

On product markets, intermediaries actually reduce risk of market transactions borne from a high volatility of demand by holding inventories of goods on stock, which in turn allow them to offer an immediacy service to their customers.¹³⁴

According to Lim (1981), retail and wholesale intermediaries serve another important role – beside offering immediacy service and lowering transaction costs as already shown in the case of financial intermediaries – which is diversifying supplier risks by purchasing and reselling not only one but a variety of products. The gains from trade with intermediaries as risk bearers do not necessarily require that the intermediaries have better information regarding consumer demands or less costly retailing technologies. Rather, risk pooling via product diversification done by them is sufficient to provide a profit incentive for intermediation. Furthermore, price discounts offered to intermediaries' bulk orders can then be regarded as a form of risk premium for his insurance service provision.

Spulber (1985) showed, that manufacturers and wholesalers pool common inventories in central warehouses to counter market volatility, smoothen demand across their customers and reduce the risks of excess or insufficient inventories. Furthermore, they often enter into additional financial risk sharing arrangements with retailers like consignment or repurchase contracts. Moreover, large retail chains achieve important advantages through their ability to diversify demand risk across individual retail stores.

¹³³Reasons to bank runs are manifold, see Gorton and Winton (2002, pp. 63–77). According to Diamond and Dybvig (1983), a bank run occurs in an undesirable equilibrium with preference shocks, where all depositors of a bank panic and withdraw immediately, including even those who would prefer to leave their deposits in if they were not concerned about the bank to fail. Chari and Jagannathan (1988) give another reason to bank runs by modelling informed depositors, who withdraw because they get information that future returns are likely to be low followed by uninformed depositors just observing this, inferring their bank is about to fail and precipitate a bank run. Jacklin and Bhattacharya (1988) even characterized information-based bank runs with two-sided asymmetric information, where banks cannot observe the true liquidity needs of their depositors while depositors in turn are uncertain about bank asset quality.

¹³⁴Cf. Spulber (1999:xviii–xix).

Stahl (1988) examined the effects of winner-takes-it-all Bertrand price competition for inputs and market making of homogeneous goods, where intermediaries set prices sequentially, competing first for suppliers and then for customers (or vice versa).¹³⁵ This leads at least to a partial equilibrium model, in which the Walrasian price arises not from a fictitious auctioneer, but from price-setting behaviour of market making intermediaries. Sticking to the fact that supply and demand in competing markets are usually interdependent, Fingleton (1997a) extended Stahl (1988) to the case where direct trade between suppliers and consumers is possible.136 He found Stahl's (1988) results are robust, because direct trade lowers but does not eliminate the monopoly rent, leaving a possible incentive for intermediaries to corner the market.¹³⁷

Garella (1989) showed that a market making intermediary is able to improve market functioning in Akerlof's (1970) Lemon's Problem even if neither signals nor quality screening is available. By proposing a random pricing scheme to potential suppliers for sale of the units they own, the intermediary obtains an improvement of average quality if he is a monopolist or the suppliers do not have the possibility to meet more than one intermediary.

Spulber (1996a) examined market making by competing heterogeneous intermediaries with consumers searching for the lowest ask price and suppliers searching for the highest bid price. In order to highlight the role of time as one form of transaction costs, they consider search for a transaction partner as time consuming and suppliers, customers as well as intermediaries discount future returns with a common discount rate. This assumption makes it possible that buyers may settle for a higher price and sellers may accept for a lower price than expected instead of continuing search in the case of non-zero transaction costs.¹³⁸ The market making intermediaries post constant bid and ask prices to maximize their present discounted value of profits before a search process begins. The results are contrasted with the standard Walrasian supply and demand framework leading to the following conclusions: (1) For any rate of discount, the market is cleared so that supply equals demand, (2) the range of ask prices is above the Walrasian price and the range of bid prices is below it, and total output is less than the Walrasian output, (3) as the discount rate goes to zero, thus lowering the time-costs of search, the market

¹³⁵Notably, this winner-takes-it-all competition in the first stage with the inputs divided equally among the competing intermediaries or all allocated to one intermediary cause the second stage either to be a duopoly with equal capacities or a monopoly.

¹³⁶Direct trade is modelled as a bilateral matching market with a centralized price coordinator, who is not at the same time an intermediary. See also Spulber (1999, pp. 61–80) for some other extensions like differentiated products and purchases or non-zero consumer switching costs.

¹³⁷In a companion contribution, Fingleton (1997b) focused on a diminishing welfare effect of introducing the possibility of direct trade with a common price scheme and the threat for the intermediaries to be disintermediated by direct trade.

¹³⁸In addition to this, different willingness to pay among the customers and different opportunity costs among the suppliers are assumed which lead to different cost of search in both groups of market participants.

equilibrium with price-setting intermediaries approaches the Walrasian equilibrium and (4) as the discount rates become large, thus raising the time-costs of search, the market equilibrium output falls reaching the monopoly solution in the limit.

Agrawal and Seshadri (2000) showed, how an intermediary such as a purchase group or a sourcing agent can reduce the financial risk faced by risk averse retailers with uncertain customer demand in a single-period inventory (newsvendor) problem. Due to their risk aversion, retailer's order quantity is usually smaller than the profit maximizing order quantity for a single product from a common vendor under a special newsvendor contract.139 According to the authors, a risk neutral intermediary can eliminate this loss of efficiency by purchasing goods from the vendor under the original newsvendor contract and sell them to the retailers through mutual beneficial risk sharing contracts by offering a menu of risk reducing pricing schemes with a fixed and variable component.

3.2.2 Matchmakers and Brokers

Beside the Walrasian auctioneer, neoclassical assumptions of perfect competitive markets include full information on both sides.¹⁴⁰ But considering real existing markets, it is more realistic to assume, that buyers and sellers usually have imperfect information about location, reservation prices and the preferences of their potential transaction partners. The unknown location of potential transaction partners implies that sellers as well as buyers have to invest search efforts in order to find each other. This search process then comes along with two sources of inefficiency. The first source is an uncertainty about finding a satisfactory match, which can be split into two components: (1) the uncertainty about finding a transaction partner and (2) given that such a transaction partner is found, the uncertainty about the buyer's reservation price (e.g. his willingness to pay) is above the seller's reservation price (e.g. his opportunity costs) and bargaining between them results in an agreement. The second source of inefficiency associated with such a search process are externalities involved therein. It is quite clear that an increase in seller's search efforts might increase buyer's gains from search (and vice versa), leading to positive externalities. If potential transaction partners do not take these positive externalities into account when choosing their personal search effort, there will be too little or too much search, depending on the magnitude of these opposing externality effects. Furthermore, such search economies might exhibit multiple

¹³⁹Under such a newsvendor contract, each retailer purchases a certain quantity at a regular purchase price, which is the same for all retailers. If the realized demand is greater than the quantity ordered, the retailer has the option to purchase the units that are short at an emergency purchase price higher than the regular one. Conversely, if the demand is less than the order quantity, the retailer has the option to get the remaining inventory at a salvage price lower than the regular one.

¹⁴⁰See Yavas (1994b) or Spulber (1996b) for this and the following.

equilibria, where pareto-dominated could occur leading to a so-called coordination failure.¹⁴¹

Introducing now matchmakers, usually called "brokers," in such a search market to coordinate market transactions by matching sellers and buyers might alleviate some if not all of these above mentioned market inefficiencies. Furthermore, matchmakers reduce at the same time transaction costs for buyers and sellers by centralizing exchange, because they are not only engaged in matching one but many transactions at the same time. Many forms of such matchmaking intermediaries exist in real-economy markets focused on stocks, real estate, insurance, labour force, travel and tourism, marriage and dating, transportation and so on. The very difference between them and the market makers already discussed before is, that they all just match buyers and sellers in triadic relationships without taking title on goods or services being traded. Furthermore, they get a commission for their matching service, whose source and amount depends on the market institution or results from bilateral agreements closed in advance. Remarkably, all these matchmakers have always to compete with decentralized search and bargaining of goods and/or services, where suppliers and customers try to find each other and negotiate prices directly.

3.2.2.1 Matchmaking on Financial Markets

As it can be clearly seen in Table 3.3, matchmaking on financial markets does not seem to be very common and therefore it is only sparsely reflected in literature. Although Baumol (1965) as well as Demsetz (1968) saw a brokerage function of financial intermediaries, they thought that their ordinary business is market making. But following Grossman and Miller (1988, pp. 619–620) or Yavas (2001), specialists as well as floor traders on stock exchanges not offering immediacy and/or liquidity services can be regarded as matchmakers on financial markets. This situation occurs especially in thin markets with low trading activity where continuous presence and willingness to buy and sell at any time most likely is not profitable. In addition to this, possible asymmetric information about true values of assets lead financial intermediaries to provide brokerage, but no immediacy and/or liquidity services.

3.2.2.2 Matchmaking on Other Markets

Starting with a classic study of Yinger (1981) explaining the business of real estate brokers and the value of multiple listing services (MLSs), research on real estate brokerage seemed to mushroom.¹⁴² In his model framework, real estate brokers

¹⁴¹Cf. Diamond (1984).

¹⁴²See Margolin (1993, pp. 8–14), Yavas (1994b) or Benjamin et al. (2000a,b) for reviews especially dedicated to this stream of brokerage literature.

must consider uncertainty. First, they search for buyers of houses on one side and collect listings of houses for sale on the other, but they neither can be certain about the number of buyers asking for houses nor the number of sellers listing with him during any given time period. In addition to this, matching of buyers with listings of houses is also uncertain because houses have many attributes and tastes of potential buyers vary widely. Simultaneously taking these three uncertainties into account, Yinger (1981) constructed a formal model of the search-and-match behaviour of real estate brokers. Furthermore, he introduced the possibility that a real estate broker might share his pool of buyers and sellers with other brokers in form of a MLS and share the commission obtained for a successful match.143 He compared his results with and without MLS, coming to the conclusion that too much resources are devoted to brokerage service in both cases although the creation of an MLS increases overall market efficiency.

In a more general market model by Rubinstein and Wolinsky (1987), buyers, sellers and intermediaries are randomly matched. The matching process is assumed to be time-consuming and all market participants discount future benefits so an intermediary can reduce the time-preference losses that occur when buyer and sellers must search for their counterparts for matching. The intermediaries are considered to increase the number of market participants that can buy or sell a good traded, as long as the likelihood of a buyer's or a seller's encountering an intermediary whom they can trade with is at least as great as the likelihood of exchanging directly with each other. Analysing gains from trade for all market participants, they explored two cases how the intermediary takes part in the matching process. First, he might facilitate the matching process by setting a bid-ask spread, which leads to a market equilibrium biased in favour of the buyers because the price paid by the intermediary to the seller is sunk when he meets the buyer and the following bargaining is over the full surplus from intermediation. If the intermediary works for the seller based of a consignment agreement, the asymmetry disappears and the surplus is divided symmetrically among the three participants, because the consignment also strengthens the position of the seller against the intermediary, since price concessions made by him will partly be rolled over to the buyer.

While analysing brokerage contracts, Margolin (1993) developed a theory of brokerage formation based on TCT. For him, brokers provide uniquely efficient solutions in markets where particular patterns of strategic behaviour interact with stochastic demand and costly information. There, brokers do not solve informational problems but solve problems associated with the exchange of information.

¹⁴³Following Yavas (1994b, pp. 176–177), such a MLS can be considered as a market mechanism that eliminates trade friction resulting from imperfect information, bringing the housing market closer to a perfect competitive market outcome. Because there are economies of scale in a matching process, the quality and quantity of matches increases with the size of such a pool of buyers of sellers. For them, a MLS means in turn a higher probability of a match and lower search costs, because it eliminates the need for a buyer (seller) to visit several real estate brokers in order to ensure a sufficient coverage of (exposure to) the market.

Furthermore, commission compensation systems applied by them provide unique governance of free riding and resale problems in line with overcoming informational pricing problems. More specifically, brokers price their services on a value-ofservice basis, the pricing method most commonly associated with simple, spotmarket exchanges of ordinary goods and services.

In Gehrig (1993), suppliers and customers can choose between using intermediaries to trade at a known price and the risky option being randomly matched on decentralized matching market, because the customers do not know the suppliers' opportunity costs and the suppliers do not know the customers' willingness to pay.¹⁴⁴ The market making intermediary chooses a profit-maximizing bid-ask spread in competition with the decentralized market option and posts common observable prices to suppliers and customers. At the market equilibrium, customers with a willingness to pay above a critical level (greater than the ask price) purchase from the intermediary, whereas suppliers with opportunity costs below a critical level (less than the bid level) sell to him. Consumers and suppliers with values between these two critical levels enter the matching market. If there is only one monopoly intermediary, both markets are active and if there are competing intermediaries, the bid-ask spread collapses to the Walrasian price and all customers and suppliers transact through intermediaries.

Cosimano (1996) presented a matching model where trading through an intermediary being more efficient in matching buyers and sellers lowers the probability of an unsuccessful trade. By posting a bid-ask spread at which sellers and buyers are guaranteed a successful trade, the intermediary may increase the expected gain from trade. Market participants with extreme valuations of the good choose to buy from (sell to) the intermediary and experience thereby an increase in their welfare since the expected net gain to trade is higher relative to the benefits of direct trade. However, there is also a harm imposed on direct traders and those market participants who are indifferent whether to trade or not at all. These traders are worse off because on average they will be dealing with counterparts who expect a lower gain from trade. The direct sellers (buyers) have a lower expected gain since they trade with direct buyers (sellers) who place a lower (higher) value on the product relative to the market participants trading through the intermediary. Furthermore, a market for homogeneous goods will have a lower bid-ask spread since the demand faced by the intermediary is more elastic and more direct trades will occur while a market with heterogeneous goods tends to have a high bid-ask spread and will be dominated by intermediated trade.

Gehrig (1996) analyzed island economies with intermediating matchmakers, who invest in information networks allowing them to inform other market participants

¹⁴⁴Yavas (1996b) further generalized Gehrig (1993) by endogenizing intermediaries's search intensities, by allowing the buyers and sellers to become an intermediary even if they have engaged themselves on the matching market, and by enabling the intermediary to provide the service of immediacy. See also Spulber (1999, pp. 118–130) for some further extensions to the original model framework by altering the bargaining process and introducing additional types of transaction costs.

on spatial dispersed islands about their price offers in a homogenous market with fixed costs of entry.¹⁴⁵ Larger networks allow them to reach more markets and a larger set of potential clients so that the probability of successful matching enhances which leads to a more efficient brokerage service. He showed that the number of intermediaries active in a symmetric equilibrium is bounded independently of the size of the market as long as investments in information networks are costly. The market structure itself resembles then to a natural oligopoly market structure with a small group of large intermediaries with similar network size and a competitive fringe of much smaller ones occupying niche markets.

Taking marriage markets as an example, Bloch and Ryder (2000) analyzed the provision of matching services in a bilateral search environment with infinite horizon, but without transferable utility. The matchmaker sets up a centralized matching procedure and proposes to match his clients according to his scheme based on an exclusive brokerage contract. The most interesting results of their study are that a matchmaker charging a uniform participation fee (like marriage brokers or travel agents) only services the upper part of the market, leaving potential clients of low quality apart. On the other hand, a matchmaker charging a commission on the matching surplus (like real estate agents or head hunters) only attracts clients of lower quality.

Rauch and Watson (2002) constructed a simple general equilibrium model of entrepreneurship in international trade, where individuals first accumulate deep knowledge of buyers and sellers by working as employees and then leave their jobs in order to provide then brokerage services on own account, based on their existing network of business contacts. They showed that in equilibrium, there can be too much or too little intermediation in a pure matching market, depending on their bargaining power and technology of intermediation.

Finally, Cummins and Doherty (2006) analyzed the economic functions of independent insurance intermediaries (brokers and independent agents) with a special focus on the commercial property and casuality insurance market in the US. As the market process and product offers are complex and multidimensional, such intermediaries help buyers to identify their coverage and risk management needs and match them with appropriate insurance companies. Furthermore, they help to alleviate adverse selection problems of the insurer because they are usually better informed about the risk of their clients than the insurer. Moreover, they increase market competition by providing access to a wider range of possible products and help their clients to select insurers on the basis of price, coverage, service and financial strength.

¹⁴⁵According to Gehrig (1996), this model applies to corporate activities of bank in Continental Europe, real estate brokers, marriage brokers, head hunters or even computer reservation systems in the airline industry, where the perception of product quality is maintained by service networks of different sizes.

3.2.2.3 The Impact of Asymmetric Information on Matchmaking

Again, asymmetric information seems to have a decisive impact on intermediation and market microstructure outcome. A first example are Myerson and Satterthwaite (1983), who showed that an intermediary can replace direct bargaining with an incentive-compatible and individual rational trading mechanism for a buyer and a seller who are exchanging on a single indivisible object.146 Such a trading mechanism determines the payment and the likelihood of trade given truthful announcements from the buyer and the seller of their respective private information to the intermediary. Furthermore, it represents the outcome of such bargaining procedures as first-and-final offers by either the buyer or the seller as well as alternating offers and encompasses double auctions, in which the good is exchanged at some average of buyer and seller bids, as long as the buyer's bid exceeds that of the seller. Doing this, the intermediary can overcome the present asymmetric information structure which allows him to tax the intermediated buyer and seller exchange by setting a profit-maximizing bid-ask spread.

Yavas (1994a) examined the role of matchmakers in a one period bilateral search economy with private information about goods valuation of buyers and sellers and showed that the introduction of such a monopolistic intermediary will narrow the set of buyer and seller types who search for a match on their own.¹⁴⁷ More specifically, sellers with a high reservation price and buyers with a low willingness to pay will immediately drop out of the search market and trade through him. Moreover, the presence of a matchmaker leads to a decrease of equilibrium search intensities for those market participants who are still engaged in searching. Therefore, the matchmaker improves welfare by affecting search behaviour of market participants, if search is very costly and inefficient (and vice versa).

Furthermore, Yavas (1996a) examined the optimality of matching strategies employed by real estate brokers under different commission structures, namely percentage commission, flat-fee and net listing systems.148 In order to shift the focus away from search and resource allocation issues of brokerage towards a matching problem, he assumed that the intermediary has already a pool of buyers and sellers to match under asymmetric information. Specifically, the buyers and sellers know their own reservation price but do not know each other's reservation prices and the intermediary knows (or manages to find out) all reservation prices. He showed

regardless of the actual transaction price, which is determined through bargaining between him and

the buyer, see Yavas (1996a, pp. 103–105).

¹⁴⁶See Spulber (1999, pp. 173–193) for this basic model of Myerson and Satterthwaite (1983) and some extensions.

¹⁴⁷Originally the study was dedicated to exploring the role of employment agencies in labour markets. But as Yavas (1994a) noted, that his model can be applied to most bilateral search markets. 148Under the percentage commission structure, the intermediary receives a percentage of the transaction price from the sale of each property whereas under the flat-fee system he raises a fixed fee per property sold, regardless of transaction price. If the intermediary and the seller sign a net listing contract, the seller specifies a price the intermediary has to pay upon the sale of the property

that the profit-maximizing matching strategy for the intermediary under percentage commission and flat-free systems also maximizes the number of houses sold, but minimizes the buyer's as well as seller's surplus. On the other hand, net listing results in fewer house sales, but yields larger surplus for buyers and sellers.

In another contribution, Yavas et al. (2001) reported experimental evidence on the impact of an intermediary in bargaining over the sale of a single good. Through an experimental analysis, they examined how he affects the price, the likelihood of a successful negotiation and the time it takes to complete a negotiation. Their results show that an intermediary, whether informed about reservation prices of buyers and seller or not, (1) increases the sale price, (2) reduces the likelihood of an agreement and (3) increases time to reach an agreement (though the number of bargaining rounds decline). So they suggested the benefits of brokerage to be predominantly in the matching of buyers and sellers rather than in the facilitation of bargaining.

3.2.2.4 Comparative Studies

After looking on market making on one side and matchmaking on the other in isolation, in this section the findings are summed up and the results of some comparative studies are shown.

Focusing on costs and benefits of intermediation on financial markets for external financing with costly search as well as costly contracting, Easterwood and Morgan (1991) showed the differences between market making and matchmaking types of intermediaries.¹⁴⁹ In this environment, the market maker enters into two costly contracts, one to provide funding for a borrower and another to obtain and repay funds from lenders, whereas the matchmaker enters into a single costly contract with a borrower to search for acceptable lenders for contracting over external financing for a certain investment.150 Intermediaries of either type operate relative to other intermediaries in a perfect competitive market for their intermediary services offered to the borrowers.151 The model allows for three modes of external financing: (1) direct financing between borrowers and lenders without an intermediary, (2) indirect financing through a matchmaker and (3) indirect financing through a market maker. Considering the level of search cost on one hand and contracting cost on the other, Easterwood and Morgan (1991) showed that a matchmaker will be employed by a borrower, if search costs are high and cost of contracting is low. Furthermore, direct financing will always occur for very low search costs but with increasing search

¹⁴⁹Throughout their paper, they denoted market makers as "depository financial intermediaries" and matchmakers as "brokerage financial intermediaries".

¹⁵⁰Due to the project of the borrower seeking for external funding is assumed to have a certain outcome, risk considerations of leding and borrowing are not considered in this study.

¹⁵¹Easterwood and Morgan (1991) noted, that their results are robust to alternative specifications of the degree of competition among the intermediaries, because they have the same necessary and sufficient conditions for market participation in their cost-benefit comparisons.

costs, a shift towards the market maker happens, who is the dominant intermediary type in case of high search as well as high contracting costs.

Using a bilateral search model with a monopolistic intermediary, Yavas (1992) compared both forms of intermediation. Referring to an earlier paper (Yavas (1991)), he asserted that market making yields to more profits and higher welfare than matchmaking when the valuations of the market participants are common knowledge. But introducing a bilateral search framework, where two market participants (a buyer and a seller) with private information about their value are searching for each other or opt for trading through an intermediary, profit and welfare results depend heavily on market characteristics.¹⁵² Comparing expected profits for both forms of intermediation, they came to the conclusion that when the valuations of the market participants are private information, market making might yield higher or lower profits and welfare effects than matchmaking, depending on the efficiency, the cost of search and the valuation distribution of the market participants.153 The intermediary in question prefers e.g. to be a market maker, if search is efficient and costless. Conversely, if search is relative inefficient and sufficiently costly, matchmaking leads to higher profits and welfare. Furthermore, they noted that there are many other reasons for an intermediary to be a matchmaker like the inability to perform market making on labour or marriage markets or there are intermediaries who provide both market making and matchmaking services like specialists on stock markets.

Hackett (1992) provided a comparative analysis of merchant and broker intermediation contracts on product markets with uncertain demand, highlighting the differences in ownership patterns and compensation structures as a key source of different performance. He set up a model framework where the intermediation process starts with commitment to one of the two contractual forms, then the intermediaries take actions (like investment in efforts in order to increase expected demand), demand is realized, and trade occurs. Merchants have to commit themselves to a quantity purchase contract with the supplier, take title on intermediated goods in order to be compensated with residual surplus after the demand is realized and goods are sold. On the other hand, brokers are intermediaries who facilitate exchange without buying and selling goods and their compensation consists of a revenue sharing commission rate specified in advance. To Hackett (1992), a principal advantage of merchant intermediation relative to intermediation through brokers is that sellers do not have to pay a commission. Furthermore, merchants will provide the expected joint surplus maximizing level of effort, because they are residual claimants. The merchant intermediary form is therefore best suited to conditions in which demand has low variance, but is highly responsive to intermediary's effort

¹⁵²Specifically, the intermediary can be either a market maker setting a bid-ask spread in order to buy or sell on own account or a match maker simply matching two parties on basis of an exclusive brokerage contract, obtaining a percentage commission on the transaction price.

¹⁵³Additional results are that both lead to smaller price dispersions (at least of the matches facilitated in the case of the matchmaker) and decreasing equilibrium search intensities.

level. Brokers never take title on goods, and providing investments to increase demand for a seller as the owner of the good leads to a less than expected joint surplus maximizing level of effort. Therefore, brokerage through intermediaries most likely occurs when demand variance is high and independent of intermediary's effort level.

Extending Spulber (1996a), Rust and Hall (2003) presented a model in which the microstructure of trade in a commodity or asset is endogenously determined. Suppliers and consumers of a commodity (or buyers and sellers of an asset) who wish to trade can choose between two competing types of intermediaries: brokers and market makers. Market makers post publicly observable bid and ask prices, whereas the prices quoted by brokers with heterogeneous transaction costs are private information that can only be obtained through a costly search process. Starting from an initial equilibrium search model where are no market makers but free entry of brokers and all trade is intermediated by them, they analyse the effect of a market entry by a single market maker. The post-entry relative share of trade intermediated by brokers rather than the market maker depends on three parameters: (1) the intertemporal discount rate, (2) the per unit transaction costs of the market maker and (3) the per unit transaction costs of the most efficient broker. If the market maker's per unit transaction costs exceed that of the least efficient broker already operating on the market, then an entry by the market maker is not profitable and therefore only brokers will operate in the post-entry equilibrium. For a broad range of parameter values, the market maker coexists with a competitive fringe of brokers undercutting the market maker's publicly posted bid-ask spread and limiting his market power. If the per unit transaction costs of the most efficient broker are sufficiently high relative to the transaction costs of the market maker, then the entry of the market maker drives all brokers out of business. However, brokers can survive the entry of a market maker even if the transaction costs of the most efficient broker exceeds the transaction costs of the market maker – provided that the gap between them is not too big. Furthermore, they showed that all suppliers and consumers choosing to participate on the market enjoy a strict increase in their expected gains from trade after entry of a market maker and Pareto improvements occur even in cases where the market maker's entry drives all broker out of business, monopolizing the intermediation of trade.

3.2.3 Information Production, Guarantors and Certificators

As buyers (customers, lenders or investors) on one side and sellers (suppliers, producers or borrowers) on the other are usually subject to asymmetric information, there are opportunities for intermediaries to help and fill this gap by collecting and supplying information for both sides of the market.154 Reviewing

¹⁵⁴Cf. Spulber (1999, pp. 147–148).

the intermediary-related literature, at least three different situations of such intermediary action can be identified: (1) Intermediaries undergo investments or sample information to get informed and give credible signals to their customers, (2) they sample information in order to sell it to everyone who is interested in it or (3) they provide certificates and guarantees on product quality.

3.2.3.1 Information Production and Signalling

Leland and Pyle (1977) suggested that an intermediary can signal his informed status and credibly produce information by investing in assets about which he has private knowledge. The starting point of their chain of argumentation is that a single intermediary has private information about an investment opportunity, but insufficient resources to undertake the necessary investments on his own. Since other potential investors do not observe his private information, there is an adverse selection problem. To alleviate this, the intermediary signals his private information by the fraction of equity put into a project whose quality is known only to him and sells the remaining fractions to other investors. This idea was further reinforced and generalized by Campbell and Kracaw (1980). According to them, "intermediaries emerge as information producers because the production of information, the protection of confidentiality, the provision of transactions services, as well as other intermediary services are naturally complementary activities" (Campbell and Kracaw (1980, p. 864)).

In another contribution, Boyd and Prescott (1986) modelled agents facing asymmetric information prior to contracting and investing in loans. The adverse selection problem is here, that agents are of different types and this information is private to each agent. Each agent, however, owns a technology to evaluate projects – or better say – determine agent type, allowing him to produce ex ante information about the projects. This ex ante information production can alleviate the adverse selection problem, so that here an agent evaluates his own project, and then issues securities to investors that promise specified returns. Alternatively, a coalition of agents can offer investors a claim on group returns. Financial intermediaries are then coalitions of agents that evaluate projects, invest in those determined to be high-value projects, and share the returns from the portfolio of projects.

Chan (1983) considered a venture capital market, where entrepreneurs select the qualities of projects and their prequisite consumption, about which investors are imperfectly informed, but can conduct costly search in order to get an informed investor.155 If all investors have positive search costs, there is a "Lemon's Problem,"

¹⁵⁵The entrepreneur selects the quality of project (which depends on time and effort expended on that project) and his perquisite consumption diverted from the return on the project. Although investors have rational expectations regarding the entrepreneurial actions in equilibrium, the actions are unobserved leaving the investors uninformed. However, they can learn about the entrepreneur through search, which can be done at a cost that vary across investors. In turn

because in a market without informed investors the entrepreneurs are always induced to offer only unacceptable inferior projects. In such an undesirable situation, competitive financial intermediaries may evolve, who conduct search for good projects and invest for their clients against a profit maximizing fee per investor. These intermediaries increase the welfare of the investors by inducing the entrepreneurs to offer high return projects.

3.2.3.2 Information Production and Selling

Building on a communication network model in marketing channels first formulated by Balderston (1958) and further extended by Baligh and Richartz (1967), Etgar and Zusman (1982) developed the idea of a marketer as an information seller on product markets. For them, information like conditions of prices and quantities in the supply side or willingness to pay on the demand side carries economic value, so that it becomes a commodity by itself. This allows a marketer to collect information about suppliers and sell it to buyers (or vice versa). Since information collection and transmission is costly, the marketer has to bear communication cost proportional to the channel contacts established by him. Furthermore, information-processing costs occur reflecting the costs associated by encoding, storing, retrieving and organizing information, which raise considerably from a certain level due to diseconomies of organization. Assuming profit maximization by the marketers and free entry, competition will force the price for information to equal average costs. Referring to hierarchical marketing channels, one important implication of this model is, that marketing channels tend to adopt a pyramid-like structure whereby the number of intermediaries at each level is smaller than at the preceding one.¹⁵⁶

Ramakrishnan and Thakor (1984) observed financial intermediaries producing information in order to sell it. Individual firms issuing new shares to the public can hire such an intermediary to produce information about their quality. As information production is a costly but unobservable effort leading to an ex post noisy indicator, an information producer might think of avoiding these costs without being detected. But being risk averse, information producers would prefer to avoid the risk that the noise in the indicator prevents them from obtaining compensation for their efforts. Furthermore, there are returns to mergers between such intermediaries if they can monitor each other's effort at gathering information within a joint organization. Risk sharing is not sufficient to create gains from such mergers, but if internal monitoring is costless and the joint organization becomes considerable large, the expected costs

the entrepreneurs take this into account, so that their actions will depend on the distribution of the investor's search costs.

¹⁵⁶The logic is as follows (see Etgar and Zusman (1982, pp. 513–514)): The profit maximization of information sellers imposes that the price of information must decline as information moves from one channel level to the next one. This gives higher level intermediaries to obtain information for a lower price which in turn allows him to operate at minimum average costs with more information buyers and generates a need for fewer intermediaries at his level.

of monitoring converge to lower limit, indicating that there are benefits from fully centralized intermediation done by a monopolist.

In contribution closely related to Ramakrishnan and Thakor (1984), Millon and Thakor (1985) proposed a theory of information production under asymmetric information and moral hazard, where an information gathering agency acquires and processes reliable information for the purpose of certifying asset qualities, but do not get involved in funding.¹⁵⁷ In their model framework, several screening agents producing reliable information about true values of assets in question choose to form such information gathering agencies, because it enables them to diversify their risky payoffs from screening activities and allows intra-organizational information sharing without any further internal monitoring.¹⁵⁸

In a series of papers, Admati and Pfleiderer (1986, 1988, 1990) explored the economics of information selling and the payoff of a risky asset by a monopolistic information producer on a financial market with a perfectly competitive noisy rational expectations equilibrium.¹⁵⁹

Assuming that such an information producer has complete control over the distribution of his product and abstracting from incentive issues, Admati and Pfleiderer (1986) treated two cases of direct information selling: (1) selling the same information to all his clients and (2) selling personalized information to each. They showed that in any instance, the seller may prefer to sell noisier versions of the information he actually has or restrict the fraction of informed traders in order to overcome the dilution in the value of information due to its leakage through informative price setting. Moreover, to obtain higher profits, he may sell different information to different traders, so that the noise realizations do not affect equilibrium prices. One way of doing so is to sell identically distributed, but personalized information to each of a larger number of traders.

Admati and Pfleiderer (1988) discussed the option of selling information versus trading on own account in a model of imperfect competition. They found that if the seller is risk neutral, he never wishes to sell his information to any other and trades on own account, whereas if he is risk averse, it is generally profitable for him to commit not to trade himself and sell the information to other traders on the

¹⁵⁷Examples to such information gathering agencies are creditrating agencies, financial newsletters, investment counselors or credit bureaus, see Millon and Thakor (1985, p. 1403).

¹⁵⁸In order to operationalize this, Millon and Thakor (1985, pp. 1405–1412) supposed that there are firms that wish to sell new shares to raise their investment capital. Further, they assumed that the true aggregate value of such new shares depends on firm specific as well as market specific variables unknown ex ante, but precise information about them can be acquired by investing in information production. Therefore, screening agents are employed for information gathering, undertaking appropriate actions in order to obtain the variable values in question after some costly investigation subject to random shocks. The compensation scheme of the screening agent is based on a publicly observable, noisy, ex post indicator of information quality which induces them to produce reliable information.

¹⁵⁹They employed the model of Kyle (1984, 1985), in which traders submit market orders to a competitive, risk neutral market maker taking into account their effect on the price.

market. Furthermore, they examined another form of selling his information called "indirect information selling." Here, the information seller creates a mutual fund whose portfolio choices are a function of his private information, and the sale of information amounts to selling shares in this fund to traders. With identical traders, selling shares in a mutual fund seems to be better than selling information directly, but observing actual financial markets, the direct selling option dominates.

Admati and Pfleiderer (1990) compared two different strategies of selling information to traders on a financial market. One method involves the direct sale of information as analyzed in Admati and Pfleiderer (1986, 1988) and the other is indirect information selling as introduced in Admati and Pfleiderer (1988). They showed that when information is sold indirectly, it is never desirable to add noise, because adjusting the price for share in the mutual fund is a more effective way to control the usage of information, but may not allow as much surplus to be extracted as it is possible in the case of direct selling.

3.2.3.3 Certification and Guarantee Provision on Product Markets

Asymmetric information about product characteristics like quality can lead to severe market failures like bad-products-drive-out-good-ones-effects as shown in the Akerlof's (1970) well known "Lemon's Problem," that arises when product quality is not immediately observable. But introducing an intermediary certifying products being traded by him may be a viable solution in this context, as it will be shown following.

Marvel and McCafferty (1984) showed that resale price maintenance will be adopted when a manufacturer wishes to obtain quality or style certification for his products from reputable retailers. Moreover, they demonstrated that manufacturers opt for resale price maintenance even when they are permitted to shape the set of stores handling their products through refusals to deal.

Wernerfelt (1988) presented a signalling model in which a multi-product firm can use its reputation as a bond for quality by using a brand name of an established product when it introduces a new experience good. When it introduces the new product under such so-called "umbrella branding," it claims that the old and the new product are both of good quality, and invites consumers to pool their experience with the two products to infer the quality of both. He showed that in equilibrium, only multi-product firms with both goods being of good quality will do so, because if one of the products is of bad quality, it will harm subsequent sales of the other, too.

In a model similar to Wernerfelt (1988), Chu and Chu (1994) explored the case of renting reputation from one agent to another in order to signal quality. Specifically, they showed that in a separating equilibrium, a manufacturer of a high quality product will signal product quality by selling through a retailer with strong reputation, while a manufacturer of a low quality product will sell through retailer with no reputation. Furthermore, the reputable retailer has an incentive to truthfully represent the quality of the product it sells in order to protect his own reputation, so that he can be regarded as a quality certifying intermediary.¹⁶⁰

Biglaiser (1993) showed that introducing a monopolistic intermediary in a market environment suffering from adverse selection enhances efficiency, because this intermediary has a greater incentive to invest in quality inspection technologies enabling him to distinguish high-quality from low-quality suppliers than an individual customer since he trades more goods. Furthermore, he has an incentive to report accurately the quality of goods, because he wants to build up and maintain a good reputation in order to carry out more and more transactions of this kind. If suppliers and customers have the choice between direct and intermediated trade, high-quality products are traded through the intermediary and low-quality ones are traded without him. This separation of low- and high-quality products alleviates the Lemon's Problem in the intermediated market equilibrium.

Since intermediaries are usually trading more than one product, customers may rely on their reputation without further investigating all the suppliers of these products offered by them, too. Biglaiser and Friedman (1994) referred to this situation, where intermediaries serve as guarantors of product quality for their suppliers through warranties and contract terms. They came to the conclusion that an intermediary in this case trades again only in high-quality products, because selling only one low-quality product suffers loss of reputation. And this may come along with less customers not only for this but all other intermediated products, too. Moreover, if a supplier defects on quality, it will be dropped by the intermediary and he has then lower sales and earns lower profits afterwards when he must sell directly to the customers, because the intermediary is an important source of information for them.

Biglaiser and Friedman (1999) extended this analysis of adverse selection to allow for heterogeneous suppliers as well as customers and a free-entry competition between intermediaries, who can provide three services: (1) to certify quality of a good after costly inspection, (2) to provide information about good's quality to customers by presenting a bid price schedule to suppliers that sorts them in different markets and (3) to eliminate search costs by matching customers and suppliers.¹⁶¹

Li (1998) presented a general equilibrium version of Biglaiser (1993) emphasizing the moral hazard problem generated by the producer's ability to produce low-quality goods that cannot be immediately identified as such by the consumers.

¹⁶⁰According to Chu and Chu (1994), another example are manufactures with no brand reputation that rent it from a well-known manufacturer or merchandizer, where manufactures from other industries, non-manufacturing companies or associations license their brand names to products but carefully inspect their quality in order to protect their brand franchises.

¹⁶¹Further, they assumed the time horizon of the intermediaries to be infinite, whereas customers and suppliers live only one period. Therefore, suppliers lack any natural incentive to truthfully reveal the quality of their goods and customers, living only one period and buying only one unit, have little incentive to become expert judges of product quality. However, intermediaries can learn through experience and benefit from a reputation for truthfulness, because they live more than one period.

In this setting, intermediaries emerge endogenously by acquiring costly the ability to verify quality of goods.¹⁶² Doing this, they are able to sell goods at a higher price than the producers because consumers believe them to sell only high-quality goods, providing the Lemon's Problem is not too severe and the investment in a quality testing technology is not too high.¹⁶³

Lizzeri (1999) discussed the role of intermediaries as certificators who alleviate the causes of asymmetric information by extracting information from privately informed parties and then choose what to reveal to uninformed parties.¹⁶⁴ Doing this, the certificating intermediary sets a fee for his certification service and credibly commits to the disclosure rule to maximize expected profits. Following him, a monopolistic intermediary that certifies a privately informed seller has an incentive to disclose the minimum amount of information necessary to induce efficient exchange. Sometimes this information is a simple minimal quality standard: certification of products with quality above some minimal threshold. Introducing competition among certificators, he showed that there is always an equilibrium in which all information is revealed and the intermediary makes no profit.

Garella and Peitz (2000) showed that in a market suffering from asymmetric information regarding product qualities165, shared retailing (selling of goods through an intermediary serving several producers) can fully or at least partially replace costly certification of product quality¹⁶⁶ – even if the intermediary is not endowed with some special expertise or reputation effects of intermediation are absent. Since undergoing certification of product quality is a costly activity whereas shared intermediation signalling product quality only implies a redistribution of profits between the intermediary and the producers (and among them), the second is socially preferable to the first.

¹⁶²More specifically, they are producers who give up production of their goods and invest in a quality testing technology in order to buy goods from other producers and sell them to customers after inspection.

¹⁶³This is one of several equilibria in terms of severity of the private information problem and investment costs for a quality-testing technology welfare-improving role for intermediaries, which Li (1998) characterized. In another one with a more severe "Lemon's Problem," intermediaries may not find it profitable enough to trade high-quality goods only, so that they trade low-quality goods, too. Then, the intermediaries cannot be regarded as being trustworthy.

¹⁶⁴Examples to such certifying intermediaries are laboratories for industrial products or auditors.

¹⁶⁵More specifically, it is assumed that consumers are imperfectly informed about product quality (being either high or low) while producers retain complete information. Producers choose their qualities and share an incentive to misrepresent low quality, which raises a moral hazard problem. In Garella and Peitz (1999), an alternative modelling approach is discussed, where Nature chooses the state and the producers then decide whether, given the state of Nature, they want to sell.

¹⁶⁶The authors referred to a partially replacement of certification, if consumers only believe in shared intermediation as a signal of high quality goods with at least one producer undergoing costly certification and the other producers or the intermediary itself give explicit or implicit side payments to him. In the extreme case of fully replacement of certification, there is no certification conducted at all.

3.2.4 Delegated Expertise, Bargaining, Contracting and/or Monitoring

Apart from these active roles facilitating market transactions with or without buying and selling goods or information services on own account as shown above, an intermediary can just be an agent of someone else, conducting specific tasks on behalf of his principal. Reasons to this behaviour are at least fourfold:

Delegated expertise, where agents are hired by principals in order to obtain their expertise.
Delegated bargaining, where principals delegate authority to agents, who have to bargain or carry out competition with third-parties instead of them.
Delegated contracting, where principals employ agents to negotiate and close contracts with third-parties on their behalf.
Delegated monitoring, where principals employ agents to monitor the performance of third-parties for them.

3.2.4.1 Delegated Expertise

In a first attempt, Demski and Sappington (1987) extended the standard principal– agent model to study delegation of authority to an expert, who is considered to be uniquely qualified acquiring pertinent knowledge useful for a given decision problem. This expertise is both personally costly to acquire and prohibitively costly to communicate to the principal.¹⁶⁷ Assuming, that (1) the expert conducts two activities (planning followed by implementation)168, (2) there is no communication from the expert to the principal and (3) the productive outcome is publicly observed after implementation, they showed how best to motivate the expert to acquire and apply his expertise. In a situation of induced moral hazard surrounding the expert's planning activities, the most efficient way to control this may entail a spillover of motivation concerns into the implementation activities.

Wolinsky (1993) investigated expert markets of medical, legal or repair services, where the seller of such a service is at the same time an expert who determines how much of his service is needed. Although the success in performing the expert's service is observable, customers often cannot determine the amount of service that was needed and how much was actually done by the expert. This information asymmetry, of course, creates incentives for opportunistic behaviour on the expert's side. Emphasizing the role of costly search by the principals or taking reputation

¹⁶⁷According to Demski and Sappington (1987), this situation captures salient ingredients of a number of principal-agent relationships like a home owner and his building contractor, an investor and his financial advisor, a patient and his doctor or a company owner and his product-line manager. 168In their setting, planning refers to the costly acquisition of private information and implementation to its use with no personal costs for the expert.

considerations into account, there are several ways to discipline these experts. In the search setting, customers conduct costly search among experts and go to two or more experts who diagnose them and make binding offers afterwards.169 In equilibrium, the experts find it useful to specialize themselves on a distinct level of treatment which virtually removes all incentives to misrepresent their information. In the reputation setting, experts do not guarantee the cost in advance – customers entrust their problem to an expert who first fixes it and then presents the bill. Here, the experts might be kept honest by their need to maintain reputation and customers minimize their search costs by going only to one expert for diagnostic, but this comes along with higher prices on treatment service.

Taylor (1995) modelled expert markets for diagnostic and treatment services for durable goods subject to random failure, where a diagnostic check is needed prior to any type of treatment. Again, these markets have an asymmetric information structure, because typically an owner of such a durable good never discovers the veracity of diagnosis or even whether the treatment he authorized was performed. Furthermore there are costs of permanent maintenance and terminal costs equal to the loss in utility if the durable good fails. The risk-neutral owner of such a durable good is assumed to minimize the expected present value of all maintenance, diagnostic, treatment and terminal costs whereas the expert needed for all diagnostic checkups or treatment is a risk-neutral profit maximizer. In this setting, Taylor (1995) found support for real world practices in markets for diagnostics and treatment, like offering incentive-compatible warranty contracts, a heavy reliance on contracting after obligatory diagnostic checks, the prevalence of free diagnostic services, consumer procrastination in obtaining checkups and an informational demand for insurance cover.

3.2.4.2 Delegated Bargaining

There are several possible reasons, why a principal sends an agent as a delegated bargainer on behalf of him into a bargaining situation: (1) Agents might be chosen because they have special skills that make them better bargainers, (2) the principal may think, that the agent is more intelligent or more experienced than he and therefore again a better bargainer or (3) delegation may serve as a pre-commitment device.170 Whereas the first and second reason of delegated bargaining are quite straight forward, the last one, often subsumed under the term "strategic delegation," needs some more elaborated explanation.

According to Schelling (1956), bargaining has two aspects: an efficiency and a distributional one. The first aspect consists of exploring mutually profitable adjustments, whereas the second occurs in "situation[s] in which more for one means less

¹⁶⁹An alternative view proposed by Taylor (1995) is, that experts price diagnostic services at cost and they cannot commit to treatment prices before performing diagnostic checks.

¹⁷⁰Cf. Fershtman and Gneezy (2001).

for the other" (Schelling (1956, p. 281)). Focusing on this distributional aspect, he pointed out that bargaining power is more than simply skill in debate, obstinacy, misrepresentation, bluffing in connection with financial resources, physical strength or ability to withstand losses. Furthermore, it is the ability to make irrevocable commitments in advance to bind oneself in a way that convinces the other party to a negotiation which creates a real powerful instrument in bargaining.171 To establish such an irrevocable commitment, a 'bargaining agent' can be employed by the principal enabling him strategic delegation. In first instance, the principal may employ an agent and give him instructions visible to the counterparty that are difficult or impossible to change. Second, an agent can be brought in as a principal in his own right with his own incentive structure after transferring the principal's rights to him against some advance payment.¹⁷²

Katz (1991) showed, that unobservable agency contracts in which principals will give their agents incentives to match their own preferences, perfectly internalize the usual principal–agent externalities and therefore cannot serve as precommitments.¹⁷³ But in some special cases, unobservable agency contracts indeed affect the outcome of bargaining, especially when they serve to share risk, give the principal itself incentives to act or help to overcome problems of adverse selection.

Fershtman and Kalai (1997) explored the situation of unobserved delegation and distinguished there between (1) instructive delegation, where agents must follow a set of instructions and (2) incentive delegation, where the outcome of bargaining determines the agent's payoff. In the case of unobservable instructive delegation, the principal cannot benefit from an irrevocable commitment through delegation, but unobservable incentive delegation can benefit the principal.

In another recent contribution by Koc¸kesen and Ok (2004) conditions were analyzed, where delegation may arise solely due to strategic reasons even under unobservable contracts: To the extent that renegotiation is costly and/or limited, strategic aspects of delegation may play an important role in contract design, even if these agency contracts are completely unobservable.

Furthermore, Fershtman and Gneezy (2001) examined strategic delegation in a single ultimatum bargaining experiment, where a proposer gives a make-it-or-leave-

¹⁷¹Or better say in the words of Schelling (1956, p. 283): "[I]f the buyer can accept an irrevocable commitment, in a way that is unambiguously visible to the seller, he can squeeze the range of indeterminacy down to the point most favourable to him."

¹⁷²Cf. Segendorff (1998). Transport insurance claims may be a good example for this form of delegated bargaining, where an insurer regulates a loss or damage at once against transfer of claiming rights and then searches to recover his expenses by sueing the responsible party.

¹⁷³Assuming that the principal and the agent have equal capabilities of bargaining, Katz (1991) considers two forms of reaching an alignment of preferences: (1) the case of perfect control where the principal's information is detailed enough so that he can offer the agent a contract with a penalty scheme forcing him to play any desirable strategy and (2) the case of perfect delegation where the principal can offer a contract making the agent a residual claimant and therefore harmonizing both interests without any need to observe what actually happens.

offer to a responder who can either accept or reject it.¹⁷⁴ One of their results was that when either the proposer or the respondent use a delegate and this delegation is observable, they obtain a considerably higher outcome which shows again the importance of delegates as a precommitment device. Conversely, if delegation contracts are not observable, the proposer obtains a nonsignificant rise in outcome offering the proposal through his agent and the responder using a delegate makes him worse off, despite the average amount of the proposals rising considerably.

Moreover, Bond and Gresik (2006) showed that in a bargaining problem between an informed and an uninformed party, partial delegation (delegating control of bargaining responsibilities to an agent while retaining control of private information) with or without observable agency contracts enable the informed party to promote more efficient agreements with their counterparty while also endogenously improving their bargaining power.

Following Spulber (1999, pp. 332–335), the use of agents as precommitment devices can be further extended to delegated competition, where agents are contracted by their respective principals in order to act on their behalf in a market for goods or services. This situation of delegated competition is usually modelled as a two stage game, where in the first stage agents are contracted by principals who compete with each other in the second stage.¹⁷⁵ But up to now, these theoretical treatments of delegated competition seem to be too explorative to draw further conclusions referring to unobservable precommitments.

Last but not least, there is another aspect of third-party bargaining not reflected in all of these models discussed above: mediation and arbitration. According to Encyclopædia Britannica (2002), mediation is "a practice under which, in a conflict, the services of a third party are utilized to reduce the differences or to seek a solution"176, whereas arbitration is "a legal method of settling disputes between parties outside ordinary court procedures by deferring to a mutually agreed-upon third party who has the authority to determine an "award," a legally binding decision."¹⁷⁷ According to Johnson and Pruitt (1972), these two kinds of

¹⁷⁴More strictly speaking, a proposer and a responder negotiate upon the division of a pie, where the proposer offers the responder a share. If the responder agrees to the proposed share, the pie is divided accordingly and both get their shares, otherwise they both get zero.

¹⁷⁵See Vickers (1984), Fershtman and Judd (1987), Sklivas (1987), Fershtman et al. (1991) or Segendorff (1998) for such two-stage models.

¹⁷⁶Other definitions are as follows: In a contribution dedicated to labour mediation, Kolb (1983, p. 23–44) asserted, that mediators either play an active role as a "deal maker" or a rather passive one as an "orchestrator." According to Dunlop (1991), mediation typically seeks to persuade contending parties to agree, but has no authority to issue a binding award. Further, Friedman (1992) described the role of a mediator to serve as a link between opposing groups and to provide a buffer between them both to bypass and to maintain differences and conflict.

¹⁷⁷According to Encyclopædia Britannica (2002), the third party called arbitrator may be a single individual or an institution chosen by the parties and knowledgeable in the area of dispute settlement. In general, awards determined by the arbitrator are based on the existing laws related to the contracts or issues in dispute as they are understood and agreed upon in advance by the

intervention in bargaining situations differ at least in two dimensions: (1) the extent to which the intermediary's recommendations are binding and (2) the amount of information available to the intermediary about the issues negotiated. Following them, a mediator's decisions are usually non-binding and he typically has less access to information than an arbitrator serving normally as a "means of the last resort," whose decisions are always binding.

Bazerman et al. (1992) examined the impacts of delegated bargaining as well as mediation on negotiated prices and the likelihood of impasse in a real estate context by conducting student's experiments. ¹⁷⁸ Their results showed that the selling price of a property is higher if an agent is employed than if no intermediary is involved. However, the selling price was not affected by the use of a mediator. Furthermore, the use of an agent tends to increase the rate of impasse, but the empirical evidence was rather insignificant.

3.2.4.3 Delegated Contracting

In the base case of delegated contracting, a first agent (the intermediary) is contracted by a principal in order to establish a contractual relationship with a second agent (or in short the agent).¹⁷⁹ Doing this, the principal wants to make sure that the terms of this contract negotiated by the intermediary are those that would have emerged from direct contracting, so that he must offer him an incentive provision scheme. In order to end in a legally binding contract with the agent, the principal furthermore has to grant his intermediary either explicit or implicit authority. ¹⁸⁰ The duties of an intermediary usually incorporated in such an agency contract can illustrate his contracting role: At least, he has to (1) be loyal, (2) protect confidential information, (3) obey instructions, (4) inform the principal, (5) take care and (6) account accurately. In turn, the minimum duties of the principal are (1) reimburse the agent for his service as well as reasonable expenses and (2) indemnify the intermediary against liability. All these legal provisions are usually well defined for various forms of agency contracts in Commercial Codes like the HGB or the UGB in countries with a codified law or generally fixed in the Law of Agency in

parties concerned and by him. According to Dunlop (1991), arbitration is most extensively used in industrial relations or labour disputes between labour and employer's organizations.

^178

Other experimental studies comparing mediation and arbitration in the context of labour disputes are e.g. Johnson and Pruitt (1972), Grigsby and Bigoness (1982) or McGillcuddy et al. (1987).

^179

See Spulber (1999, pp. 324–325) for this and the following. For ease of exposition, the first agent will always be called the intermediary while the second one is just called agent. But one has to keep in mind that there exists a three-tier principal–agent relationship structure: (1) between the original principal (the principal) and the first agent (the intermediary) followed by (2) the principal–agent relationship between the first agent as the second principal against the second agent (the agent).

^180

In Germany as well as Austria, there are legal provisions given for such delegation of authority in §§ 48–58 HGB or §§ 48–58 UGB, defining explicitly "Prokura" and "Handlungsvollmacht".

Anglo-Saxon case law. In addition to this, GT&C's offer viable solutions for other forms of agency contracts as shown in Sect. 2.1.2.

Although there seems to be a need to theoretically reflect this bulk of legal provisions, delegated contracting in isolation has not been explicitly addressed in economic models of agency, as Spulber (1999, p. 325) noted. A reason to this might be the sheer complexity of legal provisions coming from practical experience in every-day business. But considering delegated contracting in connection with delegated monitoring, quite a few theoretical treatments like McAfee and McMillan (1995), Melumad et al. (1995, 1997), Laffont and Martimort (1998), Macho-Stadler and Perez-Castrillo (1998), Baliga and Sj ´ ostr ¨ om (1998), Faure-Grimaud et al. ¨ (2003) or Mookherjee and Tsumagari (2004) can be found as described in the preceding Sect. 3.2.4.2.

3.2.4.4 Delegated Monitoring

Monitoring problems in principal–agent relationships are a well known feature of PAT already addressed in Sect. 3.1.3.2 under the label of "moral hazard." And since monitoring is costly, some principals wish to outsource this task to an intermediary as a hired supervisor. In first instance, this intermediary might be more skilled in monitoring tasks, have lower opportunity costs and/or needs to spend less effort relative to the principal.¹⁸¹ Another aspect of delegated monitoring is hierarchical contracting, where a principal (employer, purchaser, customer) contracts with an intermediary (manager, first-tier supplier, contractor) who then writes a contract with an agent (employee, second-tier supplier, subcontractor) and supervises the agent's performance on behalf of his principal. Moreover, diversification of risk by delegated monitors provide important reasons to intermediated exchange in comparison with direct one, which will be developed further in the case of financial markets.

Third-Party Monitoring

It is straightforward to see that delegation of monitoring tasks to a supervisor is more efficient, if the principal has lower skills, higher opportunity costs or has to spend more effort doing it by own means. But as Strausz (1997) showed, a principal still prefers to delegate monitoring tasks to a supervisor even when these efficiency gains cannot be reached since signals from monitoring are private information and monitoring itself is non-verifiable and therefore non-contractable. For him, delegated monitoring is then still beneficial for two reasons. First, since monitoring is non-contractable, the principal must create incentives to induce both monitoring and agent's effort. By hiring a supervisor as a third-party monitor he

¹⁸¹Cf. Spulber (1999, p. 337).

has an extra contract at his disposal and can therefore better regulate incentives for monitoring and effort, leading to an incentive-effect of delegation. Second, delegated monitoring comes along with an increased possibility of commitment. As the evidence obtained by monitoring is private information and can therefore be concealed, a profit maximizing principal will never reveal information which forces him to reward the agent. When he employs a supervisor for monitoring tasks, such rewards can be made credible and commitment to reward the agent is possible, leading to a commitment-effect of delegation.

Hierarchical Contracting

According to Spulber (1999, p. 335), the cost of writing contingent contracts and monitoring performance can create advantages for hierarchical contracting over centralized one done by the principal.182 Notably, this situation occurs within hierarchical organizations as well as in an interorganizational context.

McAfee and McMillan (1995) explored organizational diseconomies of scale, where people in a hierarchy are able to exploit bargaining power that their private information gives them. In their model, only the agent has private information with a supervisor designing the contract. The supervisor extracts information rents in return for reporting the agent's information to the principal. As a result, adding layers causes information rents to increase exponentially with the length of the hierarchy. Moreover, because of the costs of incentives the principal's marginal cost of producing output also increases with the length of the hierarchy, thus reducing the overall output of the organization.

Melumad et al. (1995) pointed out that hierarchical contracting can be a means of efficient information processing among members of an organization. But the advantages of overcoming bounded rationality come at the cost of control.¹⁸³ Comparing a centralized contracting of two agents with several three-tier hierarchical contracting relationships similar to McAfee and McMillan (1995), they come to the conclusion that hierarchical contracting can serve as a second-best solution and is maybe preferable, if the principal wants to communicate and contract with only one agent. To reach such a second-best solution by hierarchical contracting, the principal must then be able to design a favourable sequence of contracting and monitor the intermediary's contribution to the joint product of him and the agent.

Melumad et al. (1997) showed, that hierarchical contracting may be better than centralized contracting of two agents when formulating contracts is costly, so

¹⁸²Originally, Spulber (1999, p. 335) used in this context the term "delegated contracting" instead of "hierarchical contracting," which is in contradiction to his statements made earlier in Spulber (1999:325). Therefore throughout this section, the term "hierarchical contracting" is employed to indicate, that delegated monitoring issues are added to the original delegated contracting.

¹⁸³Especially, if the principal cannot monitor the intermediary's contribution, he will be able to take advantage of his monopoly power, creating a problem that resembles that of double marginalization of rents usually occurring in vertical relationships (see Tirole (1988, pp. 174–176)).

that the resulting contracts are incomplete in the sense that they are not covering all contingencies. To them, a flexibility gain obtained by hierarchical contracting always outweights the loss of control, if the principal can align his preferences with those of the intermediary in the way as described in Melumad et al. (1995).¹⁸⁴

It was then Laffont and Martimort (1998) who enriched the discussion about centralized versus hierarchical contracting by allowing collusive behaviour but limited ability of communication between the agents. Under this circumstances, delegation of contracting takes advantage of the conflict between the agents coming from their asymmetric treatment and it dominates a centralized contracting of both agents by the principal.

Another rational for hierarchical contracting was given by Macho-Stadler and Perez-Castrillo (1998), who showed that centr ´ alized contracting is only superior, when the principal is able to contract with the agents simultaneously and can monitor their collusive behaviour of side contracting at the same time. Otherwise hierarchical contracting provides a first-best solution of allocating contracting capacity with the principal appointing agents to contract others who cannot be centrally monitored by him.

Considering a model where agents work in sequence one after another on a project under limited liability (like an architect and a builder), share information not available to the principal and can collude against him, Baliga and Sjostr ¨ om (1998) ¨ showed again that the optimal distribution of surplus can be always achieved by hierarchical contracting.

Recently, Faure-Grimaud et al. (2003) considered an organization with an agent privately informed on his productivity and a risk-averse supervisor getting signals about the agent's type. Comparing contracting both agents simultaneously with a hierarchical solution, where the supervisor is contracted by the principal and then sub-contracts the agent, the latter leads again to the best possible resource allocation. This outcome is again caused by the fact that in the case of centralized contracting both agents are able to collude against the principal.

And last but not least, Mookherjee and Tsumagari (2004) studied three alternative ways a principal can contract two agents supplying distinct inputs allowing him to form a valuable output. Furthermore, the two agents have private information about their costs and can collude with one another in order to establish side contracts not observable by the principal. They showed, that delegating to one agent the right to subcontract with the other always results in lower profits for the principal compared with the centralized contracting solution. Moreover, delegation of contracting to an intermediary who in turn contracts both agents and has superior information about their costs does not improve the situation for the principal, as long as the agent's inputs are substitutes. However, this sort of hierarchical contracting

¹⁸⁴More concrete, Melumad et al. (1997) explicitly mention a number of conditions, including (1) risk neutrality, (2) an appropriate sequencing of contracts, (3) the absence of collusion between the intermediary and the agent and (4) the ability of the principal to monitor the intermediary's production contribution.

can be beneficial, if the agents produce complements and the loss of control from delegation is not too high.

Delegated Monitoring on Financial Markets

Another stream of literature copes with delegated monitoring on financial markets, where multiple principals as lenders delegate monitoring of debt contracts to intermediaries who may achieve economies of scale of monitoring agents as borrowers by producing information more efficiently.¹⁸⁵

Diamond (1984, 1996) argued that commercial banks are intermediaries raising funds from many lenders (depositors) by promising them a given riskless pattern of returns in order to lend them to borrowers. Furthermore, they spend resources on monitoring and enforcing their loan contracts which are less costly than those available without monitoring. This task of monitoring and the payments received from the borrowers are both non-observable by the depositors.¹⁸⁶ From a lender's point of view, direct monitoring of borrowers is very costly and so it seems to be efficient to delegate this task to a specialized intermediary, although after this delegation a problem of monitoring the monitor occurs. But lenders to the intermediary reduce their total monitoring costs if the costs of monitoring the intermediary are lower than the monitoring costs of lenders lending directly to borrowers. Reasons to such lower intermediaries' monitoring costs can be found in economies of scale, reputational effects or gains from diversification by subdividing independent risks.

Combining Diamond's (1984) approach to delegated monitoring with the ideas of Stiglitz and Weiss (1981) about credit rationing, Williamson (1986) illustrated how intermediation on loan markets and credit rationing may be interrelated phenomena. In equilibrium, credit rationing may emerge because costly state verification adds a premium to loan rates and intermediation arises simultaneously as a way to minimize this premium and thus minimize rationing by economizing on monitoring costs.

Furthermore, Krasa and Villamil (1992), Yanelle (1997) and Hellwig (2000) extended both model frameworks of Diamond (1984) and Williamson (1986) in different directions. Krasa and Villamil (1992) showed that in an economy with a finite number of lenders and borrowers and costly state verification, two-sided debt contracts referring to the intermediary's contracts with lenders and borrowers occur, given a sufficient number of borrowers. Moreover, such two-sided debt contracts strictly dominate all other types of contracts in this environment. Yanelle (1997) added Bertrand-style price competition between intermediaries on deposit as well as loan rates leading to market asymmetries which induces the intermediaries to corner

¹⁸⁵Cf. Spulber (1999, p. 337).

¹⁸⁶Notably, the depositors are indeed not interested in knowing to whom the intermediary actually lends money, because they get a risk-free rent for their deposits.

one market as an attempt to achieve a monopoly position on the other. Hellwig (2000) studied the impacts of introducing risk aversion and concluded that all in all, Diamond's (1984) results are still valid.

3.2.5 Conclusions and Some Applications

After this rather lengthy review of theoretical treatments and some empirical evidence of MMT and ITF, findings are wrapped up now. As already stated at the beginning of Sect. 3.2, intermediaries provide at least six specific advantages over direct exchange and referring to their specific role-sets, they can be divided into four distinguishable categories as follows:

Market makers are intermediaries taking title on goods traded and operating in dyadic relationships. They usually reduce transaction costs, pool and diversify risks. Furthermore, they alleviate adverse selection as well as mitigate opportunism, when there is information asymmetry present. Other, more special services of market makers are immediacy and/or liquidity provision allowing consumption smoothing.
Matchmakers do not take title on goods and operate in triadic relationships in order to match prospective contracting parties. Doing this, they lower costs of searching and matching for market participants as well as other transaction costs of contracting. If there are asymmetric information structures present, they again help to alleviate adverse selection and mitigate opportunism.
Information producers, guarantors and certificators economize on asymmetric information present before contracting and seek to alleviate adverse selection, which tends to result in transaction as well as search cost reduction.
Agents employed for specific tasks like delegated expertise, bargaining, contracting and/or monitoring offer diverse advantages to their principals, with supporting commitments through delegation and mitigating moral hazard as the most crucial two. Furthermore, they can help to reduce transaction costs, costs of searching and matching or even help to diversify risk associated with contractual relationships. Moreover they may help their clients to alleviate bounded rationality, limited capacity and collusive behaviour or to build up and maintain efficient incentive and organization structures.¹⁸⁷

Although MMT and ITF is an evolving field of research as shown in the last sections, the overwhelming number of applications are coping mostly with financial, real estate, wholesale or retail markets up to now. Referring to the context of transport and/or logistics, this strict micro view of market structure and the role of intermediaries therein is rather non-existent. Notable exceptions in this direction are early works of Morash (1986, 1987) or Westfall (1987) describing US-American

¹⁸⁷Cf. Spulber (1999, pp. 342–343).

freight forwarders as specialized marketing channel participants and Strandenes (2000) or Curti (2003) with recent contributions about ship brokerage issues. Furthermore, Wu (2004) explored the theory of supply chain intermediation based on bargaining theory and gave an outline for further research in this direction.

Morash (1986, 1987) observed freight forwarders as well as other transport intermediaries in the United States from a logistics channel point of view and their interaction with the transactional and promotional channels of their clients. Furthermore, he suggested, that all these transport intermediaries should facilitate reduction of overall transport costs while preserving desired service quality at the same time. More specifically, he considered freight forwarders to be highly involved in transport efficiency functions like assembly, consolidation, distribution, break bulk, sorting, storage and containerization of shipments as well as selecting low cost carrier modes, whereas transport capacity retailing as well as service quality functions of service aggregation, expediting and tracking of shipments or coordination, control and administration tasks are underdeveloped in comparison to transport brokers and shipper's agents.

Westfall (1987) analyzed freight forwarder operations contracted by the United States Air Force in order to determine which freight forwarder service components provide the best measures to evaluate the effectiveness of independent members in an international channel of distribution. His ambitious, well structured measurement approach ended rather inconclusive, maybe caused by the small sampling frame size of 126 freight forwarders resulting in 31 testable observations.

Considering shipbrokers as intermediaries matching sellers and buyers of vessels or transport services, Strandenes (2000) analyzed their contribution to market efficiency. According to her, they (1) help to speed up searching and matching processes, (2) contribute to establish favourable ask-bid prices in bargaining situations and (3) provide risk-reducing expertise in deals with asymmetric information. Furthermore, she pointed out that efficient searching and matching are important in spot freight markets, whereas quality assessments are more important in secondhand and timecharter market. All in all, shipbrokers are important intermediaries for her, who add value to the ship brokerage business by economizing on their special market knowledge and experience.

Curti (2003) discussed the roles of brokers and agents as intermediaries on three markets with a special focus on their legal regulation in Germany. Considering, that shipbrokers, travel agents and real estate brokers all facilitate exchange on markets heavily impacted by information asymmetries, he asked how efficient the legal measures raised in German law to prevent them from opportunistic behaviour are. In the case of the shipping industry, he concluded, that ship brokers acting as agents for one party or on behalf and in the interest of both parties cannot behave opportunistically towards their clients and in addition to this, their business operations are highly regulated by the HGB.¹⁸⁸ Therefore their main task is

¹⁸⁸Despite ending in a correct conclusion, Curti's (2003:10–11) arguments seem to be somekind short-cut: Regarding vessels and cargos being highly homogenous commodities for all market

matching – not decreasing informational asymmetries – and market-based solutions are the preferred means to contract enforcement.

Employing a bargaining theoretical framework, Wu (2004) set out to examine why supply chain intermediaries exist, in which different forms they operate, and the way they influence supply chain efficiency. He considered such intermediaries to fall in two categories: (1) transactional intermediaries improving efficiency in certain supply chain transactions and (2) informational intermediaries, who alleviate inefficiencies borne from informational asymmetry. Referring to the first category, he considered each supply chain member to be an intermediary between upstream suppliers and downstream customers, who broadly serve the functions of a market maker as already outlined above.¹⁸⁹ Concerning the informational intermediary, he regarded him to avoid adverse selection by administrating coordination mechanisms to create a trusted institution thereby reducing the needs for direct negotiation and to synthesize dispersed information in order to reduce information asymmetries – intermediary functions who are highly correlated with the role-sets of the other three non-market making intermediaries as discussed in Sect. 3.2.5.

Furthermore, he introduced the role of a supply chain integrator as the leader of a more or less vertically integrated supply chain, who decides to use intermediaries or not, depending on intermediated trades are either more profitable than direct exchange or offer some value-added in terms of transaction efficiency and/or information asymmetry reduction. Then he developed a basic theoretical framework for supply chain intermediation based on four basic models with self-interested, profit seeking and risk neutral buyers, suppliers and supply chain intermediaries in between: (1) bilateral bargaining under complete information, (2) bilateral bargaining under incomplete information, (3) multilateral trade with vertical integration and (4) multilateral trade with matching markets. In this settings, a transactional intermediary improves operational efficiency by serving as an intermediate supply chain member or a third-party service provider, whereas an informational intermediary mainly alleviates the effects of information asymmetry by serving as a broker or a delegated bargainer. With this strict intermediary perspective of supply chains, he proposed to expand the view of negotiation and coordination in supply chains and outlined huge research opportunities concerning supply chain coordination, design or microstructure. But it can be clearly seen, that this supply chain intermediation approach is in his very infancy state and further development is necessary.

participants, he considered asymmetric information structure to be negligible without further discussion after assuming, that brokers should ideally act as fair agents for all parties involved. But his merely legal point of view can be supported, if their specialized market knowledge and reputational effects stemming from long-term market presence are taken into consideration.

¹⁸⁹More specifically, Wu (2004, p. 71) noted that a transactional intermediary may reduce uncertainty by setting and stabilizing prices, reduce the costs associated with searching and matching, provide immediacy by holding inventory or reserving capacity and aggregating supply of demand to achieve economy of scale.

3.3 A Taxonomy of Intermediaries in Multimodal Transport and Logistic Chains

Having this theoretical framework of market micro structure and intermediation theory in mind, a new intermediary perspective of multimodal transport and logistic chains can be established. As such transport and logistic chains are caused by trade, commercial as well as financial aspects have to be taken into account to provide a full picture of cross-border activities going on between their source (e.g. an original producer) and sink (e.g. an end consumer). Therefore, the following taxonomy of intermediaries along a multimodal transport and logistic chain as shown in Fig. 3.3 includes several layers with different objectives, but the same goal: facilitating trade of goods and/or services in an international trade environment.

First, there must be drawn a distinction between middlemen and service providers. Middlemen facilitate market transactions of goods and/or services as brokers or agents without holding stock or even taking title on them whereas service providers actually engage in these market transactions or offer means of handling, transporting, storing and warehousing of physical goods being transacted. However, there are several hybrid forms, which engage in both middlemen-like intermediation and actual service provision in order to market this as a bundle to their clients.

Second, looking at the international trade environment, several layers of economic activity can be observed. First, commercial intermediaries are focused on

Fig. 3.3 A taxonomy of middlemen and service providers

trading of goods and/or services. Next, financial intermediaries allow to overcome simple exchange of goods and offer mitigation of risks coming along with such trading operations. Because goods usually have to be moved physically between spatially dispersed transaction partners, transport middlemen play an organizing role with transport service providers actually moving goods from source to sink. Furthermore, spatially dispersed economic activities and intertemporal deviations between production and consumption of goods raise an increasing demand for logistics middlemen and service providers. In addition to this, infomediaries and information service providers serve all participants of the international environment by offering trade-related information, brokerage, consultancy and/or their special expertise on all levels of economic activity.190 But as these intermediaries fitting to this last group are myriad and as they can at the same time be regarded to play a rather ancillary role, they will be excluded from further observation.

This grouping of economic activity in several layers clarifies the diverse functions of middlemen and service providers and shows their contribution to marketing channels of cross-border active companies as shown in Fig. 2.2 as well as to the international trade environment as a whole.

3.3.1 Commercial Middlemen and Service Providers

In indirect modes of export–import operations, third-party commercial intermediation is a common feature, because it allows firms of all sizes an easy and reliable way to engage in cross-border business activities without getting too much involved in daily matters in the marketing channels of negotiation, promotion, ordering and documentation. For ease of explication, in the following a focus is set on export operations of an original producer from a country of origin (called home) to a country of destination (called abroad) with the other way around declared in the footnotes, if necessary.¹⁹¹

First, a producer may sell his goods directly to an export merchant at home, who takes title on goods and actually exports them at own risk and on own account.¹⁹²

¹⁹⁰Examples for such infomediaries are trade consultancy firms, advertising agencies, trade associations, chambers of commerce, credit rating companies, translators, goods surveyors and many other private as well as governmental organizations supporting a "trade-friendly" environment. Further, information service providers establishing and maintaining information and communication infrastructure all around the world are getting more and more important.

¹⁹¹Of course, the actual country of origin status of a good traded can be different from the country where the trade transaction actually takes place and goods even can be traded at remote third countries, directly exporting them to the country of destination. But in this context it makes no difference in exposition, so that such situations of transit or third country trade are not explicitly taken in consideration.

¹⁹²Conversely, an export merchant can buy goods from suppliers abroad in order to resell them at home after importation.

Alternatively, a producer might export goods to the country of destination at own risk in order to sell it directly to an import merchant abroad for further distribution.¹⁹³ In both instances, their relationships are usually long-term coming along with some exclusivity and these intermediaries can be regarded as pure market makers like merchant wholesalers and retailers as heavily reflected in the US-American marketing channel literature as well as the German "Handelslehre".194

Another export mode is assigning commercial agents for performing business activities abroad, which are third-parties working as agents on account and behalf of an exporter or importer on a permanent basis.195 Actually their obligations can vary widely, from being (1) just a broker-like agent searching for potential matches with buyers abroad as a delegated bargainer, (2) a sales representative as a delegated contractor abroad to (3) a factor-like agent, acting as a delegated contractor with holding consignment stocks for his client abroad.196 Furthermore, commodity brokers at home as well as abroad can be engaged in facilitating goods transactions between buyers and sellers on commodity markets as real matchmakers.

In addition to these easily distinguishable middlemen, there are several hybrid modes of commercial intermediation.

First there are commission merchants, who buy or sell goods abroad on behalf and account of their clients, but in their own name.¹⁹⁷ Doing this, they can then be considered to be a subcontracted market maker holding inventory, but actually not taking title on it.¹⁹⁸

Second, a considerable share of world trade is organized by export trading companies (ETCs) with a long tradition in most European countries (especially Germany, Sweden, France, the Netherlands and the United Kingdom) and general

¹⁹³Conversely, an import merchant can also buy goods from suppliers abroad after they arrived at the country of destination.

¹⁹⁴According to Spulber (1996b, pp. 137–138), retailers provide commercial services like pricing, marketing, holding inventory, certificating quality to their customers in the business-to-consumer segment, whereas wholesalers act as intermediaries in business-to-business transactions likely in the same manner by distributing goods, managing inventories, communicating prices and product information, certifying quality and providing credit. Further see e.g. Rosenbloom (1991, pp. 45-52 and 61-65), Kleer (1991, pp. 52–56), Stern and El-Ansary (1992, pp. 547–559), Bowersox and Cooper (1992, pp. 40–42), Bowersox and Closs (1996, pp. 115–117) or Pfohl (2003, pp. 214–216) with a more detailed treatment of typical merchant wholesaler as well as retailer functions.

¹⁹⁵Other labels are "wholesale agents" or "merchandise agents" – for this and a more detailed description of their obligations see Rosenbloom (1991, pp. 52–54), Bowersox and Cooper (1992, pp. 42–43) or Triebel et al. (1995, pp. 166–180).

¹⁹⁶The same holds true with a broker-like agent, purchase representative or factor-like agent acquiring goods for supply abroad.

¹⁹⁷Cf. Rosenbloom (1991, pp. 53–54), Bowersox and Cooper (1992, p. 43) or Bowersox and Closs (1996, p. 117). In export operations, they usually hold consignment stocks whereas in import operations, they run international procurement offices on behalf of their clients.

¹⁹⁸This seems to be a very strange hybrid not really fitting into this conceptual framework. But actually, a commission merchant e.g. acquires physical possession of goods to be sold with transfer of title being postponed until they are sold.

trading companies (GTCs) or "sogo shosha" (Jap.) from Japan or Brazil, Thailand, Taiwan, China and South Korea.¹⁹⁹ Both engage in cross-border trade as commercial intermediaries with a range from simple export/import brokerage over export/import merchant activities to highly diversified operations by integrating into other fields of international business like production, finance, countertrade, insurance, and even warehousing and shipping.200 Furthermore, they specialize themselves by trade direction, product and/or region or are engaged in various business activities all over the world, even in third-country trades.²⁰¹ A constituting feature of ETCs or GTCs is that they are not only present either at home or abroad, but in both countries (at least through dedicated representatives or agents) in order to have better control of their cross-border shipments.

From an intermediary point of view, the main functions of ETCs and GTCs, albeit from their actual commercial intermediation activity as a matchmaker or market maker, lay in their ability to reduce search, negotiation and transaction costs significantly for their clients, especially in export-import operations coming along with considerable high risks.²⁰² Furthermore, especially large GTCs engage in countertrade and financial intermediation in a quasi-banking manner or through affiliated commercial banks.203 Moreover, as both are engaged in their fields of specialization abroad on a long-term basis, they can permanently monitor markets in focus and accumulate useful business knowledge, which can be easily interpreted as a form of information production.²⁰⁴

A third hybrid form are export management companies (EMCs), which are traditional third-party commercial intermediaries in the United States, organizing virtually all task needed for marketing and exporting physical goods abroad on behalf of their clients with sometimes even taking title on goods traded, if necessary.205 Therefore, they can be at best described as delegated experts in export

¹⁹⁹See Stern and El-Ansary (1992, p. 547), Denis and Mallette-LaFreniere (1987), Mattsson (1986) ` Balabanis and Baker (1993), Peng and Ilinitch (1998), Jones (1998), Ellis (2003, pp. 1684–1686) and Albaum et al. (2005, pp. 282–292) for a general overview and Sarathy (1985), Cho (1984), Kim (1986), Kojima and Ozawa (1993) with a focus set on GTCs.

²⁰⁰Cf. Thomchick and Rosenbaum (1984) or Casson (1998). Notably, this seems to be the main reason why in more recent contributions like Balabanis (2000, 2001), Peng and York (2001) or Trabold (2002), all these commercial intermediaries were subsumed under the label "export intermediaries".

²⁰¹See e.g. Cho (1984), Jones (1998), Casson (1998) and Balabanis (2001).

²⁰²For this and the following see Ricker Kelly and Lecraw (1985), Kim (1986, pp. 35–36), Kojima and Ozawa (1993) or Casson (1998).

²⁰³See especially Thomchick and Rosenbaum (1984, pp. 98–99), Kojima and Ozama (1993, pp. 149–151) or Casson (1998, p. 37–44) for a more detailed discussion of countertrade and financial intermediation issues.

²⁰⁴See Kojima and Ozawa (1993, pp. 151–152) or Jones (1998, pp. 16–19) for a more elaborated treatment of this topic.

²⁰⁵See Brasch (1978, 1981), Williamson and Bello (1984), Bello and Williamson (1985a) or Haigh (1994). With the Export Trading Company Act of 1982 US-American legislation removed most of the legal restrictions which inhibited up to then a development of ETCs and GTCs as mentioned

marketing and management acting on account and in the name of their clients abroad.

3.3.2 Financial Middlemen and Service Providers

Following Cecchetti (1999), financial intermediaries of the twenty-first century generally (1) give access to the payment systems, (2) provide liquidity, either in form of demandable deposits or stand-by line of credits, (3) (re)package and (re)sell risks as part of their traditional banking function, (4) provide information, including quality-certification credit review and possible follow-up and (5) remain a conduit for government guarantees like deposit insurance or governmental bailments.206 In addition to these basic financial intermediation and banking functions as already discussed in Sect. 3.2, financial intermediaries commonly called commercial banks offer other dedicated services for cross-border business operations like documentary collection or documentary credit, which can be regarded as viable means to synchronize financing, ownership and documentation flows in international marketing channels.²⁰⁷

Furthermore, there are more specialized financial middlemen on the international trade arena like (1) confirming houses²⁰⁸ as agents specialized on financial and documentary flow matters and (2) a multitude of non-banking financial intermediaries (specialists, traders or brokers as discussed in Sects. 3.2.1 and 3.2.2) providing access to financial markets where all kinds of derivatives employable for financial hedging needs are traded. Moreover, highly specialized financial service providers (1) actually process financial and documentary flows like SWIFT209, IATA-CASS210

above but with rather little success, see Thomchick and Rosenbaum (1984), Bello and Williamson (1985b), Howard and Maskula (1988), De Noble et al. (1989) or Perry (1992).

²⁰⁶Another contemporary view of financial intermediation was provided by Collin (1997), who contrasted financial intermediation through markets with the banking system. To him, banks are "financial intermediaries with organizational boundaries that create property rights to allocative competence, making offers that are credible as a result, first of internal monitoring by organizational hierarchy and the residual; second, of external monitoring by government and insurance companies; and, third, through signalling, e.g. by conspicuous investments" (Collin (1997, p. 184)).

²⁰⁷See e.g. Lorenz (2004, pp. 315–325) or Kummer et al. (2010, pp. 233–235) with further references.

²⁰⁸Cf. Triebel et al. (1995, pp. 181–182) or Albaum et al. (1995, p. 288).

²⁰⁹SWIFT is the abbreviation for the "Society for Worldwide Interbank Financial Telecommunication," running a closed data communication network for financial transfer operations among their member companies, see Kummer et al. (2010, pp. 247–248) or http://www.swift.com for further information.

²¹⁰Cf. Sect. 2.2.3.1 about IATA-CASS.

and other clearing houses or (2) provide dedicated non-banking financial services like factoring²¹¹ by factors or forfaiting²¹² by forfaiters.

Another field of specialization is the provision of dedicated insurance coverage for physical goods in transit as well as financial flows in connection with them by specialized insurance companies. Concerning the physical goods in transit, transport insurance cover against loss or damage can be provided by several private insurance companies upon request directly or through their own middlemen-based distribution systems.213 In the case of financial flows, export credit insurance provision is usually offered through export credit insurance companies empowered and backed by national state, like the Euler Hermes Kreditversicherungs AG (EHK) in Germany and the Oesterreichische Kontrollbank AG (OeKB) in Austria.214

3.3.3 Transport Middlemen and Service Providers

Transport middlemen and service providers are intermediaries dedicated to move goods physically. As they are already described in detail in Sect. 2.2, there seems to be no use to do this a second time but to restate their functional roles from an intermediary point of view.

Regarding transport middlemen as matchmakers, they can be further divided into two basic intermediary roles: agents and brokers. First, transport middlemen offer agency services to the market on behalf and account of their clients vis-`a-vis their customers like (1) shipper's agents for shippers, (2) ship agents, shipping agents or liner agents for shipping companies and (3) IATA cargo agents or general sales agents for airlines. Second, others are engaged in the transport brokerage business like (4) transport brokers for inland transport operations, (5) ship brokers in the ocean shipping industry and (6) *air freight brokers*in the cargo segment of the airline industry, offering their service to both sides of the market. In addition to this, some of these transport middlemen engage heavily into information production, too, and

²¹¹Following Encyclopædia Britannica (2002), factoring is the selling of accounts receivable on a contract basis by their original holder to an agency known as a factor in order to obtain cash payment of the accounts before their actual due date. The factor then assumes full responsibility for credit analysis of new accounts, payments collection, and credit losses. Factoring differs from borrowing in that the accounts receivable and the responsibility for their collection are actually sold rather than merely offered as loan collateral. Further, see e.g. Kummer et al. (2010, pp. 239–241) with a more elaborated treatment.

²¹²Forfaiting is the purchase of credit instruments like letters of credit, bills of exchange, promissory notes, or other freely negotiable instruments on a non-recourse basis. The forfaiter deducts interest (in the form of a discount), at an agreed rate for the full credit period covered by the forfaitor, see Kummer et al. (2010, pp. 246–247) for this and further references.

²¹³See e.g. Lorenz (2004, pp. 234–246) or Kummer et al. (2010, pp. 382–384) with further references.

²¹⁴See http://www.hermes-kredit.com, http://www.oekb.co.at, Branch (2000, pp. 231–237) or Kummer et al. (2010, p. 384–386) for further information.

offer then their market expertise to the public in form of market surveys and/or newsletter services.

In first instance, transport service providers like trucking companies, railways, airlines and shipping companies are operating carriers actually moving goods physically. But as always relying solely on own transport capacity is too costly in most transport operations, they often are contractual carriers at the same time by subcontracting other transport service providers. Looking at contemporary market practice like liner consortia or strategic alliances in container shipping as well as interlining agreements and cargo alliances in the airline industry already described in Sect. 2.2.3.2, there seems to be a more and more fuzzy line between transportation by own means or contracting it out to other carriers. From an intermediary point of view, they serve the role of a market maker on transport markets buying, transforming and selling transportation service as a non-physical commodity. Furthermore, as service quality seems to be important at least for some shippers, they engage in certification and/or guarantee provision by e.g. offering time-definite transport services.

3.3.4 Logistics Middlemen and Service Providers

While the last group of intermediaries is mainly engaged in spatial movement of physical goods, logistics intermediaries care for goods movement in time as well as other more or less related transformation operations. On the logistics service provider's side, there are terminal operators at seaports or ground handlers at airports which have already been described in Sect. 2.2.4. Other, more focused intermediaries are public or private warehouses and export packaging firms with special expertise in protective packaging of goods, stuffing or stripping containers. A common feature of all of these logistics service providers is, that they are working either on their own site or at the client's premises.

Third-party logistics providers (3PLs or system providers) integrate all these logistics processes mentioned above in one hand in order to offer them to their clients on a one-stop-shopping basis.²¹⁵ According to Sink et al. (1996, p. 40), "third party logistics services are multiple distribution activities provided by an external party, assuming no ownership of inventory, to accomplish related functions that are not desired to be rendered and/or managed by the purchasing organization." Furthermore, they listed 25 logistics activities normally associated with contract logistics, grouped in six distinct logistics functions of transportation, warehousing, inventory management, order processing, information system provision and packaging.²¹⁶

²¹⁵See e.g. Sheffi (1990), Rendez (1992), Virum (1993), Rao et al. (1993), Laarhoven and Sharman (1994), Stank and Maltz (1996), Uhlig (1996), Diehl (1996), Peters et al. (1998), Berglund et al. (1999), Engelsleben (1999), Laarhoven et al. (2000), Murphy and Poist (2000), Logistics Europe (2002), Bolumole (2001, 2003) or Love (2004).

²¹⁶Cf. Sink et al. (1996, p. 41).

Hertz and Alfredsson (2003) went beyond this enumeration of logistics functions and classified 3PLs along their general problem-solving abilities and their ability to adopt to customers' needs in four subgroups: (1) standard third-party logistics providers just offering highly standardised contract logistics services to the public, (2) service developers offering dedicated logistics as well as advanced value added services to their customers, (3) customer adapters usually taking over their customer's existing logistics facilities in order to run them as a subcontractor, and (4) customer developers taking over their customers' whole logistics operations. As they are acting merely like a consultant, the last mentioned subgroup of customer developers is quite similar to the business of fourth-party logistics discussed in the following.

Furthermore, there are fourth-party logistics providers (4PLs or system integrators), which can be considered as asset-free logistics middlemen offering expertise for the establishment and control of complex logistic systems, including logistics consulting and the organization of information infrastructure, transport, logistics as well as financial services needed.217 Originally, the term 4PL is a trademark of Accenture (former Andersen Consulting) and was defined as "an integrator that assembles the resources, capabilities and technology of its own organization and other organizations to design, build and run comprehensive supply-chain solutions" (e.g. Bumstead and Cannons (2002, p. 19)). But up to now, there is an ongoing more or less fruitless debate about the definition or the role a 4PL should play in supply chains, coming along with an ever lasting redefinition of the topic without further insights.²¹⁸ In addition to this, 4PL in practice seems to be rather fiction than real business practice.²¹⁹ Perhaps the most advanced 4PLs are (1) logistics management departments having the ability to offer their in-house services to external clients, (2) transport integrators or postal services widening their range of value added services or (3) so-called lead logistics providers (LLPs), which are, in effect, freight forwarders heavily engaged in dedicated contract logistics backed by heavy usage of modern information processing technology.220

Concerning their intermediary functions, both can be regarded as agents taking over all kinds of logistics functions for their clients. Whereas 4PLs focus merely on the organizational side of logistics and supply chain management without getting too much involved in physical goods movement but informational flows, 3PLs actually provide physical and informational infrastructure to perform transport and

²¹⁷Cf. Gattorna (1998), Zadek (2001), Baumgarten and Zadek (2002a,b), Baumgarten et al. (2002), K¨uhner (2002), Emmermann et al. (2003), Love (2004), Albers et al. (2005) or Fulconis et al. (2007).

²¹⁸See e.g. Foster (1999), Nissen and Bothe (2002, pp. 17–18), Emmermann et al. (2003) or Win (2008).

²¹⁹Cf. Thorby (2001c), Peters (2002), Baumgarten et al. (2002), Love (2004) or Fulconis et al. (2007).

²²⁰Cf. Baumgarten and Zadek (2002a), Baumgarten et al. (2002), Nissen and Bothe (2002, pp. 19–20), Emmermann et al. (2003) or Albers et al. (2005). Further, Hertz and Alfredsson (2003) regarded them just as a special form of 3PL.

logistics services. As a high degree of service quality is demanded in such usually long-termed outsourcing relationships, they heavily engage in certification and/or guarantee provision in order to feed trust and reputation towards their clients.

3.3.5 Freight Forwarders as a Strange Hybrid

In this conceptual framework, the freight forwarder (as well as customs house brokers for import and NVOCCs or OTIs for export operations in the United States) may be described best as a strange hybrid providing transport and logistics services as well as financial intermediation on behalf and account of his clients with the following role-set:²²¹

As organizing shipment of physical goods being somehow the core of freight forwarding business (at least from a legal point of view as shown in Sect. 2.1.1), he can be regarded in first instance as a transport middlemen between shippers and carriers.222 There, his main intermediary role is to be a matchmaker between shippers seeking for transport service and carriers seeking for cargo offered for transport.
When dispatching and transporting goods on own account as a contractual or even operating carrier, he is a transport service provider - even from a legal point of view (see again Sect. 2.1.1) and his intermediary role is being a market maker on transport markets.
As spatial cargo movement is often closely coupled with other logistics activities like packaging, handling, temporary storage or warehousing, a freight forwarder offers them as a logistics service provider, too. Moreover, these logistic activities allow him to offer consolidation of cargoes and groupage service, where an analogy to wholesaling or retailing activities of commercial middlemen can be drawn.
Offering transport or warehouse insurance coverage, he engages in financial intermediation as a dedicated middlemen on behalf of insurance companies taking risks. At the same time, he serves his client at least temporary credit functions for freight, charges, fees or duties payable at once and offers documentary collection or cash-on-delivery services upon request.

²²¹In addition to this, freight forwarders may engage into commercial intermediation, too. But today, commercial activities as an export, import or commission merchant described e.g. by Vadnai (1921, p. 60–61) seem to be very rare.

²²²This, of course, includes contracts of affreightment and completion of documentary paperwork needed for cross-border shipments.

• Last but not least, as already indicated above, a freight forwarder can be regarded as a logistics middlemen, if he offers full service logistics packages in a onestop-shopping manner as a LLP.²²³

3.4 Summary and Some Criticism on the Intermediary Perspective

As shown at the beginning of this part of the book, NIE give theoretical wellfounded insights for various phenomena in economic relationships by clarifying the role of property rights, transaction and agency costs in law and economics. In addition to this, MMT and ITF help to understand, how markets are structured and inter-organizational relationships (with and without intermediaries) are working. For TCT, PAT and intermediation theory, at least some applications were identified within the field of transport and logistics or supply chain management, and a taxonomy of intermediaries in multimodal transport and logistic chains was established. The two dimensions, namely middlemen and service providers and four layers (commercial, financial, transport and logistics) of this theoretical framework gave an impression of the inherent specialization of actors therein. However, some hybrid forms of intermediaries offering more then one kind of service could be identified with the freight forwarder somewhere in between, focused on transport and logistics matters either as middleman or actual service provider.

But especially the theoretical concepts TCT and PAT raised some heavy criticism, which can be grouped along four statements discussed in the following:

There is an overall narrow focus on single phenomena in isolation.
The way of modelling economic situations seems to be biased.
There is often a failure in quantification and validation.
A limited scope on a few relationships is evident.

A Narrow Focus Set on Single Phenomena

Especially in TCT and PAT, there is always a narrow focus set on single economic phenomena in isolation. TCT is centered around transactions and their characteristics which lead to a set of transaction cost minimizing forms of governance for contractual relations.²²⁴ In PAT, research objects are principals, agents and their

²²³See Steininger (2000), Lindner and Benz (2002), Rudolph (2002), Peters (2002), Emmermann et al. (2003) or Love (2004).

²²⁴Cf. Johanson and Mattsson (1987, pp. 42–43), Barney and Hesterly (1996, p. 123), Holmstrom¨ and Roberts (1998, pp. 91–92), Casson and Wadeson (1998, p. 5–6) or Sect. 3.1.3.2.

actions under the presence of asymmetric information structures.²²⁵ To some extent, this narrow focus seems to be set in order to keep things simple as in the case of normative PAT: finding a phenomena, raising questions to motivate a formal model while securing mathematical tractability to end in clear-cut conclusions.²²⁶ But looking at real-world phenomena, it is evident that (1) the context of a situation and its surrounding environment has to be taken into account, (2) inter-temporal dynamics matter and (3) there is not always a straight-forward, unique and stable equilibrium solution available.

MMT and ITF put the focus of investigation a step further in exploring establishment of market structures and various ways markets allocate goods and/or services as well as the important economic role played by intermediaries on them. But relying to a great extent on existing concepts like NIE, IOT as well as NCE, most of these contributions as described in this part of the book have again a rather narrow focus in giving theoretical explanations to specific problems raised in isolation.

A Biased Way of Modelling Economic Situations

The second statement is a logical consequence of the first: As the focus is set on single phenomena, modelling economic situations can be quite parsimonious but biased. Many critics of TCT as well as PAT argue that both theoretical approaches heavily highlight the negative, selfish side of human behaviour, which results in a merely "undersocialized" view of things:227

In TCT, bounded rationality, asset specificity and an ever present threat of opportunism support a mechanism of economic ordering which results in specific governance structures meeting adapted to specific situations. Doing this, it abstracts away issues like market power, resource dependence, cultural forces and/or social embeddedness.228
PAT builds on the fundamental behavioural assumption that principals cannot trust their agents shirking their duties and always behaving opportunistically if given the chance. The general solution proposed is then to invest in monitoring

²²⁵Levinthal (1988), Saam (2002, pp. 52–54) or Sect. 3.1.3.1.

²²⁶This statement originates out of personal experience as the author holding a master degree in economics from Humboldt-University always heard at the end of a typical lecture a sentence like this: "[. . . ] the solution [of our rather endless acrobatics in mathematics] is 1/3 – what does it mean?"

²²⁷Cf. Granovetter (1992), Nilakant and Rao (1994, pp. 650–651), Sheppard and Tuchinsky (1996, pp. 334–336), Barney and Hesterly (1996, pp. 123–129), Gobel (2002, pp. 343–357) or Richter ¨ and Furubotn (2003, pp. 209–210).

²²⁸Cf. Johanson and Mattsson (1987, p. 43) or Shelanski and Klein (1995, p. 228) as well as Barney and Hesterly (1996, p. 123).

them and/or design "optimal" contract mechanisms offering some monetary rewards.²²⁹

However, theoretical concepts of mutual trust and/or cooperative behaviour seem to be of great importance in every-day economic transactions, which in turn are heavily negated in TCT, PAT, MMT as well as ITF at the same time.

Moreover, both problem-solving approaches function in well specified environments designed for, but fail in others, so that they cannot be considered as one-fit-all explanations for real-world economic behaviour.230 Taking Williamsonian TCE as an example and confronting it with real-world economic behaviour, more than the black and white picture of a dichotomy of markets versus hierarchies and a few hybrid exceptions as shown in Sect. 3.1.3.1 can easily be identified:²³¹

Richardson (1972) regarded firms not to be "islands of planned co-ordination in a sea of market relations." Looking at actual business practices in manufacturing and retail industries, he considered firms to coordinate their activities in three ways: by direction, by cooperation or through market transactions.232 For him, "planned co-ordination does not stop at the frontiers of the individual firm but can be effected through co-operation between firms" so that the dichotomy between markets and hierarchies seems to be misleading.
Blois (1972) discussed the impacts of uneven size and power of contracting partners in customer–supplier relationships, where the stronger party seeks to extent its influence over the counterpart. Calling this situation "vertical quasiintegration," he asked then how firms are gaining the advantage of vertical integration without assuming the risks or rigidity of ownership.
MacMillan and Farmer (1979) observed an increasing trend in the 1970s towards closer vertical relationships between buyers and sellers on intermediate industry markets which they called "collaborative dealing." Believing that problems of interorganizational coordination cannot be only solved by vertical integration or exclusive dealing contracts, they see considerable mutual gains through

²²⁹Cf. Perrow (1986), Petersen (1993, pp. 288–289), Nilakant and Rao (1994, pp. 655–656), Barney and Hesterly (1996, pp. 128–129) or Saam (2002, pp. 54–55). Some of them add the claim, that PAT has an inherent bias towards the principal's interests.

²³⁰Cf. Barney and Hesterly (1996, p. 123) indicating that transaction cost considerations seem to be of second order or Nilakant and Rao (1994, p. 656–659) showing that situations characterized either by (1) a conflict over outcomes and complete knowledge about effort–outcome relationships or (2) knowledge about effort–outcome is incomplete and multiple preferences about outcome exist are outside of the PAT research domain. Moreover, Saam (2002, pp. 57–59), claimed that models of PAT are always of little robustness to changes in their underlying assumptions.

²³¹Cf. Bradach and Eccles (1989), Powell (1990), Siebert (1991, pp. 292–294), Sydow (1992, pp. 62–74), Elg and Johansson (1993, pp. 30–32) or Holmstrom and Roberts (1998). ¨

²³²According to Richardson (1972, p. 890), direction is employed when business activities are consolidated in an organization under a single control and one coherent plan, cooperation occurs when two or more independent organizations agree to match their related plans in advance and market transactions lead to a spontaneous coordination of business activities based on successive interacting decisions taken in response to changing profit opportunities.

long-term, collaborative relationships between otherwise independently operating firms.

While not fully questioning the dichotomy of markets versus hierarchies, Ouchi (1980) extended the Williamsonian organization failures framework with a third organizational dimension. He distinguished between markets, bureaucracies and clans as three distinct mechanisms of mediating transactions between individuals and described them along their normative and informational requirements.²³³
Conducting a field study in the construction industry, Eccles (1981) observed there a common contracting mode somewhere between the relational contracting modes of bilateral and unified governance, where a general contractor and his specialized subcontractors form a stable organizational unit called "quasifirm."

This list of organizational forms being neither markets nor hierarchies can be extended *ad finitum*234, but already these early contributions from Richardson (1972) to Eccles (1981) showed, that Williamsonian TCE lacks explanatory power in many situations. Moreover, Hill (1990) came to the conclusion, that over time the invisible hand of the market may favour actors who behave less opportunistic but more cooperative so that the general market failure proposed by Williamson tends to be a short-term phenomenon. Saam (2002:51) argued this missing dynamic view, too, which prevents PAT to cover long-lasting and sustainable principal–agent relationships.

A Failure in Quantification and Validation

Another line of criticism is a failure in quantification and validation, which is a main goal of empirical studies confronting theoretical model constructs with realworld situations. Although there are quite a lot of applications to cope with various transport and logistics issues, most of them employ TCT, PAT and intermediation theory solely to support their line of argumentation in a normative way. Looking at the various topics treated in these empirical studies, it seems to be quite evident, that these empirical treatments are often borne out of the chance to observe a phenomena fitting to the theoretical framework and to collect enough concrete

²³³Following Ouchi (1980, pp. 137–139), normative requirements "refer to the basic social agreements that all members of the transactional network must share if the network is to function effectively, without undue costs of performance auditing or monitoring" and informational requirements are prerequisites needed for decision making. Argumenting along these lines, markets need only a norm of reciprocity and prices as an informational prerequisite to work efficient whereas bureaucracies as a first form of hierarchy rely on a set of rules instead of prices and need, in addition to reciprocity, a norm of legitimate authority to function properly. Clans, however, are a second more socialized form of hierarchy governed by traditions as well as common values and beliefs in addition to legitimate authority and reciprocity.

²³⁴For more contributions of this kind, see Part 4 discussing network forms of economic organization in more detail.

data around it at the same time in order to conduct a case study.235 Moreover, the empirical findings thereof are often not too convincing as David and Han (2004) or Carter and Hodgson (2006) showed after systematic evaluation of empirical studies based on Williamsonian TCE. The same is valid in the case of PAT as Arrow (1985, p. 49) noted: "in some cases where principal–agent theory seems clearly applicable, real-world practice is very different from the model" and some critics even raised the claim that problems of PAT are not empirically important.236

Indeed, attempts to estimate the amount of transaction, agency or intermediation costs are rather rare. Maybe a simple reason to this is, that these costs cannot be measured at all, because their theoretical constructs rest on rather vague, nonmeasurable conceptions.²³⁷

A Limited Scope on a Few Relationships

The last point of criticism is one of a limited scope: Whereas TCT and PAT are mainly restricted to dyadic relationships238, intermediation theory extents its scope to triads, but not beyond.²³⁹ This delimitation seems to either lie in the underlying concept of relationship treatment and/or is again a matter of securing mathematical tractability. But thinking of specific real-world situations and setting the scope solely on relationships of two or three distinguishable actors, all these models shown above are rather stylized and merely tell only an episode of the whole story. Looking at the topic of this work, multimodal transport chains including a leg by sea or air and their participants as presented in Sect. 2.2, it is quite obvious, that they cannot be treated as a conglomerate of isolated bilateral or trilateral relationships. This shows a need for a more integrative approach to multimodal transport chains, which takes explicitly the striking network structure into consideration.

²³⁵See e.g. the reviews by Shelanski and Klein (1995) or David and Han (2004) on TCT and Eisenhardt (1989) or Petersen (1993) on PAT.

²³⁶Cf. Petersen (1993, p. 288) relying on Perrow (1986) or Nilakant and Rao (1994, pp. 659–663) and Saam (2002, pp. 41–52).

²³⁷See e.g. Masten et al. (1991, pp. 2–8) or Shelanski and Klein (1995, pp. 338–341) for a more detailed discussion of estimation issues in TCT and Schneider (1987), Schmidt (1987) as well as Saam (2002, p. 55) for PAT.

²³⁸See e.g. Johanson and Mattsson (1987, p. 43) or Elg and Johansson (1993) for TCT and Saam (2002, pp. 50–51) for PAT. Notable exceptions are Wathne and Heide (2004) using TCE to explore interfirm governance in double-dyadic relationships between contractors, apparel companies and retailers forming a vertical supply chain network or Agrell et al. (2004) employing a minimum principal–agent model to analyze three-tier supply chains in the telecommunication industry.

²³⁹Again with the notable exception of Mookherjee and Tsumagari (2004) already discussed in Sect. 3.1.3.2.

Chapter 4 Models of Network Structure and the Network Perspective

One of the first explicit definitions of networks in the social sciences can be found in Moreno (1934), summing up his results of a structural analysis of groups: "We have found that these [psychological] currents which break group lines, and even community lines, are not lawless. They are related to more or less permanent structures which bind individuals together into complex lines of transportation and communication, large 'networks'." (Moreno (1934), (1953, p. 441)). According to Casti (1995, 5), all kinds of networks "[...] have in common a set of objects [...] tied together in a connective structure by links [...]. What this adds up to is that we can abstractly regard a network as nothing more (or less) than a system: network D objects C connections D system."

A viable way to visualize such network systems consisting of objects and connections between them is the graph theory. As a research subject in its own right it was first formulated in a monograph by Konig (1936), but it took until 1950 ¨ for his pioneering work to be recognized by the international scientific community, after it was republished in the United States.537 As graph-theoretical descriptions of networks will be heavily employed in the following, a set of basic definitions from graph theory will be outlined in advance:538

⁵³⁷In addition to this, it was the only monograph dedicated to the subject of graph theory until Berge (1958), see Gallai (1990). Authors like Barnes (1969, p. 217), Deo (1974, p. 10) or Jungnickel (1994, 17–18) date the first graph theoretic treatment back to the mathematican Leonard Euler, who discussed the problem of crossing all seven bridges over the River Pregel at Konigsberg in ¨ 1736 without crossing each bridge more than once. See e.g. Konig (1936, pp. 24–25), Deo (1974, ¨ PP. 23–30), Nollemeier (1975:12–13), Jungnickel (1994, pp. 31–33) or Casti (1995, pp. 8–10) for a more elaborated discussion of this bridge-crossing problem.

⁵³⁸Cf. Konig (1936, pp. 1–18), Moreno (1953, pp. 719–720), Fla ¨ ment (1963), Harary et al. (1965), Freeman (1979), Knoke and Kuklinski (1982, pp. 37–42), Schenk (1984, pp. 42–46), Scott (1991, 67–71), Ahuja et al. (1993, pp. 23–31), Casti (1995), Johnson (1995), Domschke and Drexl (1995, pp. 1–16), Jansen (2003, pp. 94–97), Brandes and Erlebach (2005) or Knoke and Yang (2008, pp. 45–51).

A graph as a visualization of a network structure consists of a non-empty set of N points $p_i$ with $i=1,\ldots,N$ standing for actors or facilities (sources, sinks or transhipment points) and a non-empty set of M lines $q_m$ with $m=1,\ldots,M$ called arcs, edges, links, relations or ties connecting pairs of points.
The strength of a link between a pair of points (p_i, p_j) with i ≠ j can be denoted by z_ii = z(p_i, p_j) for all i ≠ j.
If all possible lines between all points are present and therefore z_ij ≠ 0 for all i ≠ i, a graph is complete.
If the direction of the link between a pair of points (p_i, p_j) with i ≠ j, matters so that (p_i, p_j) is not virtually the same as (p_j, p_i), then a graph is called a directed graph (or digraph) and the linkages between points are expressed as a set of directed lines called choices or nominations. In a digraph, three different types of linkages between two points can occur: (1) mutual, with both points linked to each other, (2) asymmetric in which one point directs a line toward another that is not reciprocated and (3) null, in which no link in either direction between a pair of points exists.
In a signed graph, a linkage between two points can have a positive or negative sign, to represent a positive or negative affection between two actors connected.
In a valued graph, the strength of a link can have different magnitude standing for intensity, capacity or costs incurred.
In a binary graph, a link between pairs of points (p_i, p_j) with i ≠ j either exists or not, so that the strength of all links consists of dichotomous relations and therefore z_ij ∈ {0; 1} for all i ≠ j. Each graph can be transformed into a binary graph just by setting z_ij = 1 for all i ≠ j if there is a link between p_i and p_j and 0 otherwise, regardless of strength and direction.
When two points p_i and p_j are directly connected by an edge they are called adjacent, and their link can be denoted by a_ij = a(p_i, p_j) for all i ≠ j.
A walk is an alternating sequence of points and edges where each edge is linked to both the preceding and succeeding point. If every point and edge in a walk is distinct (different) it is a path.
Associated with each path is a distance d_ij = d(p_i, p_j) for all i ≠ j equal to the number of edges in that path.
A cycle is a closed path which begins and ends in the same point.
Given an unordered pair of points (p_i, p_j) where i ≠ j, one point is reachable from another if and only if they are adjacent or they are connected through a path including one (or more) intermediary point(s) p_k between p_i and p_j, like (p_i, p_k) and (p_k, p_j) in a path with d_ij = 2.
When every point p_i is reachable from any other point p_j and i ≠ j, the graph is called connected.
The shortest path linking a given pair of points (p_i, p_j) with i ≠ j is called a geodesic g_ij = g(p_i, p_j) for all i ≠ j.
Points p_k falling on the geodesic or on all geodesics linking a given pair of points (p_i, p_j) with i ≠ j are said to stand between these end points.

As already indicated in the above mentioned definition by Moreno (1934), transportation and communication flows between individuals form networks. Therefore,

it seems to be a quite straight forward extension of multimodal transport chains to treat them as networks. In the following, six theory-driven streams of modelling network structures are discussed one after another, before they are compared and their usability in the context of this work is assessed. Of course, in the wide research fields of economics, law and other social sciences there are far more modelling approaches as e.g. Sydow (1992), Grandori and Soda (1995), Powell and Smith-Doerr (1994), Araujo and Easton (1996) or Oliver and Ebers (1998) showed in their extensive reviews on network literature. But only these six approaches are of special interest here, because they seem to be directly applicable to multimodal transport chains including a leg by sea or air.

4.1 Network Modelling in Operations Research and Management Science

According to Pfohl (2003:1-7), all sorts of logistics systems are basically systems of transformation objects between space and time which in turn can be characterized as a sequence of transport and storage processes. Furthermore, these processes can be visualized as a set of points for storage processes, connected with lines as transport processes forming a network structure.⁵³⁹ A first and foremost approach of modelling such network structures – apart from just visualizing them as a graph - is to quantify these points and their connections in order to describe them mathematically. Historically, this approach to network modelling is closely related to the rapid development of Operations Research (OR) and Management Science (MS) as a distinct field of scientific research starting around the World War II, which will be addressed first, before applications to specific transport and logistics networks of interest in this context are discussed.

4.1.1 Historical and Theoretical Background

Following Koopmans (1951), Brothers (1954), Solandt (1955), Dantzig (1963, pp. 12–14), Glover et al. (1992, pp. 2–3), Fortuin and van Beek (1996), Thomas (1996), Vertinsky (1998), Dantzig (2002) or Weir and Thomas (2008), the subject of OR / MS originates in dedicated operations research departments established by the military forces of the United Kingdom and the United States around World War II. Their purpose was to improve all kind of military operations from artillery fire effectiveness or bombing accuracy over vehicle and crew scheduling, radar usage to staff, supply and maintenance planning by means of mathematical methods.

⁵³⁹See Simon (1993:20-31) discussing structural as well as procedural aspects of logistic networks and Jaggi (2000, pp. 88–92) for a de ¨ tailed typology of network definitions in transport and logistics.

During their work on problems of military operations research, first mathematical programming techniques were founded, which draw heavily on graph theory in connection with matrix algebra, combinatorial analysis, and methods for solving linear equation systems.⁵⁴⁰

After World War II, many of these researchers employed in military operations research departments returned to non-military assignments and developed these mathematical programming techniques further to solve practical problems in business activities like production planning, inventory control and physical distribution.⁵⁴¹ In the 1950s and 1960s, initial contributions to OR/MS modelling and mathematical programming approaches were published like monographs by Morse and Morse and Kimball (1951) and Churchman et al. (1957) on operations research in general, Ford and Ford and Fulkerson (1962) on network flow problems or Dantzig (1963) on techniques in linear programming.542 In addition to this, a steady stream of articles appeared, with initial contributions e.g. on the warehouse problem (Cahn, 1948 or Charnes and Cooper, 1955), the optimal movement of ships in ballast (Koopmans, 1949; Koopmans and Reiter, 1951; Flood, 1954 or Dantzig and Fulkerson, 1954), the optimal allocation of aircrafts to routes under uncertain demand (Ferguson and Dantzig, 1955, 1956), the warehouse-location problem (Baumol and Wolfe, 1958) or the truck dispatching problem (Dantzig and Ramser 1959).543 Furthermore, transportation problems already assessed before by Tolsto˘ı (1930, 1939), Kantorovich (1939, 1942) or Hitchcock (1941) were treated with the transhipment problem (Kantorovich and Gavurin, 1949 or Orden, 1956) and the traveling-salesman problem (Dantzig et al., 1954; Flood, 1956; Ford and Fulkerson, 1956 or other modelling approaches reviewed by Bellmore and Nemhauser (1968)) as the most prominent ones.⁵⁴⁴ During this short time, most standard approaches of mathematical programming in the field of OR/MS like the Simplex Algorithm (Dantzig 1951; 1963, pp. 94–253), the Branch-and-Bound Algorithm (various authors, see a review by Lawler and Wood, 1966), the Primal-Dual Algorithm (Ford and Fulkerson (1957, 1962, pp. 26–30) or the Hungarian Method (Kuhn, 1955) were introduced. Today, the whole subject of OR/MS covers a wide-spread field of theories including mathematical programming, combinatorial

⁵⁴⁰See Koopmans (1951) or Dantzig (1963, pp. 20–24, 2002) for further explanation about the origins of mathematical programming.

⁵⁴¹Cf. Dantzig (1963, pp. 18–20), Thomas and Biddle (1966a) or Fortuin and van Beek (1996).

⁵⁴²Some of these early contributions to OR/MS relied on much former research findings subject to a strict military confidentiality. But later on, they were released after being declassified, like in the case of Morse and Kimball (1951), Flood (1954) or Harris and Ross (1955) with the last one downgraded recently upon request in 1999, see Schrijver (2002, p. 441).

⁵⁴³See Avondo-Bodino (1962), Berge and Ghouila-Houri (1962) or Dantzig (1963) for a more detailed overview of these early contributions.

⁵⁴⁴See Avondo-Bodino (1962), Ford and Fulkerson (1962, pp. 93–172), Dantzig (1963, pp. 299– 351), Glover et al. (1992, pp. 6–7), Ahuja et al. (1993, pp. 19–20) or Schrijver (2002) for further discussion about early treatments of these transport problems.

analysis, queuing theory, network plan technique545 and simulation546, providing problem-solving techniques for various applications not only in the business and military sector, but also e.g. in sports, politics and library management.⁵⁴⁷

From a network point of view, the objectives of OR/MS modelling can be divided into two broad categories: (1) a tactical and an operational level with actual network flow problems in given network structures and (2) a strategic level with network design problems focusing both on prospective network flow and network structure simultaneously.⁵⁴⁸ Following Ahuja et al. (1993, pp. 4–9), network flow models can be classified further in (1) shortest path (or longest path) problems, (2) maximum flow (or minimum cut) problems and (3) minimum cost flow problems, with the latter consisting of transportation problems, assignment problems, circulation problems and many more generalized forms covering topics like non-linearity, loss or gains in transit or multicommodity flows.⁵⁴⁹ The category of network design models mainly consists of minimum spanning tree or Steiner tree problems and network location problems.⁵⁵⁰

4.1.2 Applications

As already indicated before, problems from transport and logistics operations are somehow a core field of applications in OR/MS. So, various theory-driven models of network structure can there be found, as it will be shown in the following

⁵⁴⁵Network plan techniques are methods for scheduling, risk assessment, control, capacity and cost planning in projects, see e.g. Noltemeier (1976, p. 219). Well known procedures are the Critical Path Method (CPM), the Programm Evaluation and Review Technique (PERT), the Meta Potential Method (MPM) and the Graphical Evaluation and Review Technique (GERT).

⁵⁴⁶Cf. Heinzl and Brandt (1999).

⁵⁴⁷See Fortuin and van Beek (1996), Vertinsky (1998) or Domschke and Drexl (1995, pp. 7–9). Regarding these last mentioned rather unusual applications of OR/MS see short reviews by Ladany (1996), Hess and Wong-Martinez (1996) or Reisman and Xu (1996) with further references.

⁵⁴⁸Cf. Magnanti and Wong (1984), Magnanti (1996), Diruf (1999, pp. 378–380) or Crainic (2000).

⁵⁴⁹Strictly speaking, assignment (as well as matching problems) generally belong to the group of combinatorial (or discrete) optimization techniques, but there is a close relationship between them and the other network flow models mentioned above, see e.g. Iri (1969, pp. 212–238) or Ahuja et al. (1993, pp. 461–501). Further, authors like Iri (1969, pp. 3–5) or Deo (1974, pp. 400–408) considered network plan techniques as network flow models in activity networks, too. But their basic objective is project planning and scheduling of activities, so that they are excluded in this context.

⁵⁵⁰See Tansel et al. (1983b), Magnanti and Wong (1984), Dejax and Crainic (1987), Teodorovic´ (1988), Glover et al. (1992), Eiselt and Sandblom (2000), Crainic (2000) or Minoux (2003) for extensive overviews of OR/MS network modelling approaches. Further Noltemeier (1976, pp. 133– 199) or Ahuja et al. (1993, pp. 510-536) focusing on minimum spanning tree and Steiner tree problems and Tansel et al. (1983a), ReVelle (1996), Domschke and Drexl (1996, pp. 41–161) for network location problems, including the warehouse-location or hub-location problem.

with a special focus set on treatments in the airline as well as ocean shipping industry.⁵⁵¹ Concerning network structures in the field of transport and logistics, three ideal types of network configurations as well as some hybrid forms can be distinguished:552

Linear networks which consist of several walks, paths and/or cycles with some points of interchange forming a network. They were a common network configuration in the airline industry from the early beginning of scheduled transport operations up to the time e.g. of the air transport deregulation in the United States.⁵⁵³ Furthermore, industrial as well as liner shipping companies often operate in this fashion with several subsequent ports of call at both legs of a ship's journey.
Grid networks consist of a complete network with direct point-to-point connections between all sources and sinks. Examples are overnight transport networks on the main haul in groupage service, courier, express, mail or parcel operations on a national scale basis, where virtually every depot is connected with each other due to a tight time schedule.
Hub-and-spoke networks are networks with one or more hubs as transhipment points through which all net flows are routed. They are a dominant network configuration for airline networks in passenger transport operations, where at least one airport serves as a "home base" and transhipment point for each airline. In comparison to the "pure" forms of linear or grid networks, the hub-and-spoke network configuration always comes along with a considerable rise in transport frequency and volumes shipped with less vehicles, ships or aircrafts needed.
Hybrid forms are a mixture of these above mentioned forms of network configuration. Some real-world examples are: (1) linear networks in connection with additional hub-and-spoke network structures at both legs of a journey as in the case of the contemporary container shipping industry, where main ports of call serve as hubs for feeder-services, (2) grid networks with one or more additional hubs to consolidate small shipment volumes between depots as widely common in freight forwarder groupage service operations or (3) hub-and-spoke-networks with feeder-routes and/or stopovers as in the case of integrator or other express service operations.

⁵⁵¹Of course, OR/MS modeling is not restricted to these two fields solemnly, as e.g. Bramel and Simchi-Levi (1997), Schmidt and Wilhelm (2000) or Daganzo (2005) showed in their mathematical programming approach to various logistics problems, including the formulation of production planning, inventory and distribution modeling. But they are not of special interest in this context and so excluded in the following.

⁵⁵²See Kuby and Gray (1993), Jaggi (2000, pp. 105–110), Button and Stough (2000, pp. 50–58), ¨ Lohatepanont and Barnhart (2004) and Hanlon (2007, pp. 123–129).

⁵⁵³See Graham (1995, pp. 12–13), Wells (1999, pp. 37–40), Button and Stough (2000, pp. 46– 47), Jaggi (2000, pp. 109–110) or Mayer (2001, pp. 6–11). In Europe, hub-and-spoke networks ¨ outdated linear networks much earlier because of geographical, political and historical factors, see Mayer (2001, pp. 31–34) with further references.

Regarding the airline industry, OR/MS modelling approaches got wide-spread application throughout the whole hierarchical process of planning and operations management in scheduled airline services:554

Beginning with the strategic level, there are first decisions to be drawn about the network design, including network configuration, fleet planning as well as service level definitions.555 In the airline industry, network configuration, as already indicated above, depends heavily on service level defined and objects transported (passenger, cargo, mail, express or parcel service) so that quite different hub location and routing problems have to be addressed at this stage.556 After a particular network design evolved, initial flight schedule construction, fleet assignment and routing decisions have to be settled.
At the tactical level, airline scheduling is executed, followed by resource assignments. There, the goal is to find optimal work assignments for a given network design and the initial flight schedule. During the process of airline scheduling, the initial flight schedule is evaluated, given the resources (e.g. aircrafts, personnel and maintenance capacities) expected to be actual available. Then the resource assignment phase starts with aircraft assignment and routing, maintenance planning and crew scheduling, resulting in an operational plan for actual flight operations.
At the operational level, OR/MS applications are employed as decision support systems for flight operations, including aircraft re-scheduling or fleet re-routing in the case of irregular operations.557

In addition to this hierarchical structured planning and operations management, fields of OR/MS modelling in the airline industry are yield management serving as a revenue controlling tool mainly employed in passenger transport operations or efficient cargo load planning for aircrafts and ULDs as well as more general corporate planning problems not explicitly mentioned above.⁵⁵⁸

In the case of ocean shipping, OR/MS applications seem to be of less interest in the scientific community, with recent advances in connection with container terminal operations and container fleet management as notable exceptions.

⁵⁵⁴See Etschmaier and Mathaisel (1985), Teodorovic (1988), Rushmeier et al. (1995), Ryan (1996), ´ Mathaisel (1997), Rushmeier and Kontogiorgis (1997), Yu (1998), Lohatepanont and Barnhart (2004), Barnhart and Cohn (2004) or Snowdon and Paleologo (2008) for this and in the following. Relying on Suhl (1995), Mayer (2001, pp. 48–57) mapped the hierarchical planning process for scheduled airline services in four subsequent phases, starting with product planning at the strategic level, followed by production planning at the tactical level and resource planning as well as operations management at the operational level.

⁵⁵⁵See reviews by Lederer and Nambimadom (1998), Bryan and O'Kelly (1999), Mayer (2001, pp. 107–257) and Barnhart et al. (2002) focusing on OR/MS network design models in the airline industry. Further Jaggi (2000) with a resource-based approach to airline network design. ¨

⁵⁵⁶Cf. Kuby and Gray (1993), Hall (1989), Mayer (2001, pp. 107–257) or Barnhart et al. (2002).

⁵⁵⁷Cf. Mathaisel (1996) or Ball et al. (2007).

⁵⁵⁸See. e.g. Richter (1989), Amiouny et al. (1992), Smith et al. (1992), Klampferer (1996), Ryan (1996), Thomas et al. (1998) or Snowdon and Paleologo (2008).

In a series of review articles covering about 120 contributions from 1954 to 2003, Ronen (1983, 1984, 1993) and Christiansen et al. (2004) showed OR/MS modelling for ship routing and scheduling problems may not be so common as e.g. in the case of the airline industry.⁵⁵⁹ For them, main reasons are: (1) low visibility, because there are only a few companies operating ships, (2) ship scheduling is a very complicated task due to e.g. fleet variety and different operating environments, (3) shipping operations suffer from much more uncertainty, (4) the shipping market is volatile, international, capital intensive and subject to free entry and (5) the ocean shipping industry has a long tradition, is therefore somehow more conservative and builds on personal expertise.⁵⁶⁰ Moreover, other differences supporting this can be added like (6) trip or voyage lengths are much longer, (7) destinations may often change while a ship is under way, (8) affreightment on vessel from other shipping companies as well as chartering of additional vessels is quite common as well as (9) a combination of ownership, ship and operations management is quite unusual.⁵⁶¹

Focusing on liner shipping, there is a quite similar hierarchical process of planning and operations management as in the airline industry, too, but it is less formalized.⁵⁶² It starts at the strategic level with the network design, or better say a long-term decision making in route, schedule and fleet planning. At the tactical level, resource assignment (ship deployment, routing and scheduling) is done after the network design is validated and negotiations are finished with other third-parties like terminal operators involved in the shipping process. Finally, every-day business decisions about ship's sailing dates, cruising speed selection and actual service prices are drawn at the operational level with little use of OR/MS applications.

In the container shipping industry, planning and operations management of container fleets, especially for empty container allocation is another evolving field of OR/MS modelling worth to mention.⁵⁶³ Concerning terminal operations in container shipping, reviews by Vis and de Koster (2003) and Steenken et al. (2004) showed a rapidly growing interest in a wide-spread employment of OR/MS methods, because they reached such a degree of complexity that it cannot be judged any longer just by

⁵⁵⁹All in all, they found in industrial shipping most evidence for OR/MS applications with 24 treatments, followed by liner shipping (19), shipping in multimodal transport operations (16) and tramp shipping (11) with the remaining contributions covering various specialized shipping operations including e.g. fishing, cost guard and naval shipping. Furthermore, an update to this can found in Christiansen et al. (2007). In addition to this, Ernst (1982) and Ordemann (1996) are dissertations in German focusing on ship routing and scheduling in the container shipping industry, too.

⁵⁶⁰Cf. Ronen (1983, 1984).

⁵⁶¹See Christiansen et al. (2004) or Sect. 2.2.4.1 and 2.2.4.2.

⁵⁶²Cf. Pesenti (1995) or Christiansen et al. (2007).

⁵⁶³See early treatments by Ramcke (1973), Huch (1973), Jarke (1981) or Ernst (1982) as well as overviews by Dejax and Crainic (1987), Macharis and Bontekoning (2004), Christiansen et al. (2004), Maloni and Jackson (2005) or Olivo et al. (2005).

operational expertise.⁵⁶⁴ Again, a distinction between three levels of planning and operations management can be drawn:565

At the strategic level, investment decisions about container terminal layout, deployment of handling equipment and their way of operations in the transhipment process are more and more backed by simulation and mathematical programming methods.
At the operational level, the overall complexity of transhipment operations at container terminals need planning support: berth allocation of vessels, crane allocation for ship-to-shore handling, stowage planning onboard the vessels, storage planning on the container terminal as well as scheduling of on-site container handling equipment and workforce are typical applications of OR/MS in this field.
At the tactical level, assisting systems play an eminent role today for the organization and the optimization of work flow at contemporary container terminals, especially if on-site container movement is at least partly automated. There, decision problems about container handling equipment moves, scheduling and routing as well as optimal stacking policies have to be solved.⁵⁶⁶

4.1.3 Critical Comments

Concluding this section, it can be clearly seen that there is an ongoing progress in OR/MS research in the airline as well as the container shipping industry with various treatments from strategic network design to operational planning and control. But in the context of this work, a serious problem for this kind of OR/MS modelling approach to network structures is that solely functional aspects are caught with institutional ones considered as being ex ante fixed and regarded not to be of special interest. Furthermore, most decision problems assessed are looked at from a carrier's or a terminal operator's point of view, but not from a freight forwarder's

⁵⁶⁴Vis and de Koster (2003) reviewed about 55 contributions and the comprehensive literature overview by Steenken et al. (2004) contains about 200 citations, most of them being published within the last decade. Further with Jehle (1981), Kunzmann (1988), Schott (1989), Kasprzak (1997) and Pumpe (2000) there are some dissertations in German about this topic not explicitly mentioned in Vis and de Koster (2003) or Steenken et al. (2004).

⁵⁶⁵See Meersmans and Dekker (2001), Meersmans (2002, pp. 7–27), Vis and de Koster (2003), Steenken et al. (2004), G¨unther and Kim (2005) or Maloni and Jackson (2005) for a more thoroughly overview with further references. Further, Macharis and Bontekoning (2004) reviewed OR/MS modelling issues in intermodal freight transport operations with a focus on road, rail and inland navigation, where they distinguished between drayage, terminal, network and intermodal operator decision problems on a strategic, tactical or operational level.

⁵⁶⁶In addition to this, container loading problems are adressed e.g. by Dyckhoff and Finke (1992), Bischoff and Ratcliff (1995), Ratcliff and Bischoff (1998), Bortfeldt and Gehrig (1998) or Bortfeldt (2000).

one. In addition to this, modelling of network structures comes along with a high abstraction from the underlying real-world environment in order to manage the inherent computational complexity.

4.2 Strategic Networks: Focal Companies with Cooperative Relationships

Apart from this OR/MS treatments in planning and operations management, in the 1980s a network perspective evolved in strategic management. For a long time, large companies were usually hierarchically structured, vertically integrated and widely diversified with a strict demarkation line between own business operations and the market outside. But during the 1970s and 1980s, deregulation and globalization of markets paired with the revolutionary adoption of new information processing technologies resulted in a tremendous change of the business environment.567 Many of them downsized their administrative overhead, concentrated their operations around their core competencies and sought for less formal but more adaptive forms in intra- as well as inter-organizational relationships. As one outcome of this restructuring process, more and more neither market nor hierarchy forms of economic organization could be observed.568 Expanding the Williamsonian governance framework as shown in Sect. 3.1.3.1, a theory of "strategic networks" evolved originating in the works of J. Carlos Jarillo and Jorg Sydow. ¨

4.2.1 Theoretical Background

Defining them as "[...] long-term purposeful arrangements among distinct but related organizations that allow those firms in them to gain or sustain comparative advantage vis-`a-vis their competitors outside the network" Jarillo (1988) considered the strategic network concept as a distinct form of organization between hierarchies and markets. By categorising different modes of organizing complex economic activities along the two dimensions of legal form and kind of relationship, he proposed four prototypes of organization as shown in Fig. 4.1. 569

First, "classic markets" (many players interacting on a spot market basis) as well as "bureaucracies" (hierarchical organizations with antagonistic labourmanagement relationships) are organizational arrangements plagued by transaction costs and therefore driven by a competitive, zero-sum game approach to relationships. Second, "clans" (as defined by Ouchi, 1980) are cooperative organizational

⁵⁶⁷Cf. Snow et al. (1992, pp. 6–11) or Hinterhuber and Levin (1994).

⁵⁶⁸Cf. Thorelli (1986), Powell (1990, pp. 314–322) or Snow et al. (1992, pp. 11–14).

⁵⁶⁹Cf. Jarillo and Ricart (1987) or Jarillo (1988).

Fig. 4.1 Four modes of organizing economic activity (Source: adapted from Jarillo and Ricart, 1987)

arrangements which can be best characterized as long-term relationships based on relational contracts within the formal environment of an organization. Finally, in "strategic networks," a focal firm has special cooperative relationships with other members forming a network of closely interrelated, but still independent organizations. Like "clans," their relationships are long-termed and based on relational contracts upon relatively unstructured tasks. But at the same time, they can be regarded as some kind of an investment, because "there is always a certain "asset specificity" to the know-how of, say, dealing with a given supplier instead of a new one" (Jarillo, 1988).

The logic behind such a strange hybrid is as follows: A company can be regarded as a collection of value-adding activities where some can be outsourced while others are conducted in-house. It is then quite rational that activities in which one has some comparative advantage will be kept internally while others are contracted out to the most efficient supplier unless transaction costs are not too high. In order to keep them low, these inter-organizational relationships are established in a more collaborative way in order to end in a self-reinforcing system, where "[...] every characteristic of the network builds on the others: adoption, specialization, trust, long-term outlook, internal consistency vis-`a-vis the outside world, etc., which provide efficiency and flexibility. This delivers a good competitive position, justifying the existence of the network and reinforcing its characteristics" (Jarillo 1993, p. 143). At the end, this kind of network structure reduces behavioural uncertainty and enables reaching higher efficiency at the same time.

It was then Sydow (1992, 1993) who restated these initial, rather loosely formulated considerations and explored further the circumstances under which a strategic network arrangement lowers transaction costs. For him, strategic networks are the result of either a "quasi-internalization" or a "quasi-externalisation" which means an in- or outsourcing of value-adding activities in connection with a networklike coordination approach. However, Sydow (1992, p. 296, 1993, p. 234) concludes that Williamsonian TCE – mainly relying on (transaction)cost considerations – can be at best a partial approach to explain the evolution of strategic networks.

Therefore, Sydow et al. (1998, 1995) introduced a theory of structured interorganisational networks based on the theory of structuration by Giddens (1984) which considers actions and structures as related in structuration processes. Following Sydow et al. (1998), two theorems are central to this theory: (1) the duality of structure and (2) the recursiveness of life. The basic idea of the duality of structures is that "[t]he constitution of agents and structures are not two independently given sets of phenomena, a dualism, but represent a duality" (Giddens 1984, p. 25). Therefore, actions as well as structures of networks should always be investigated in respect to three dimensions: (1) signification, referring to the cognitive aspect of social practice in social systems, (2) legitimation referring to the normative aspect inherent in any social practice and (3) domination, as allocative and authoritative resources referred to intervene in ongoing network practices.570 The second theorem finally points to the fact that the history of interaction matters, because outcomes of structured action are necessarily the input for further actions. Backed by this theory of structuration, they explored the recursive interplay between industry's structures and interfirm networks therein in several cases like financial services networks (Sydow et al., 1998, 1995) or strategic networks in franchising (Sydow, 1998).

4.2.2 Applications and Critical Comments

Looking at the field of transport and logistics a multitude of strategic alliances emerged within the last decades.⁵⁷¹ Especially in the airline and the container shipping industry, strategic alliances – among other neither market nor hierarchy forms of organization – are of high importance to actual market structures, as already described in Sect. 2.2.2.2 and 2.2.3.1. Furthermore, quite a few interesting contributions focusing on strategic network issues not initiated by carriers can be identified:

• While discussing corporate governance structures of German freight forwarder transport networks on a European scale, Stahl (1995a,b) observed a distinct organizational form in which "[...] legally autonomous transport firms coordinate their activities [...] in a cooperative fashion transcending organisational [sic!] boundaries, in order to jointly obtain strategic competitive advantages vis-`a-vis competing suppliers of international logistical services from outside the network" (Stahl, 1995b). Furthermore, he showed that type of cooperation agreements, coordinating mechanisms employed and degree of centralized decision making are precisely tailored to specific cooperation characteristics as well as to their monitoring and safeguarding requirements.

⁵⁷⁰Cf. Sydow et al. (1998) or Sydow (1998).

⁵⁷¹See e.g. Faller (1999), Jung (1999) and Kaspar (1999).

In Becker (1999), a focus is set on the institutional arrangements in strategic alliances of air freight forwarding companies teaming up in air cargo consolidation in order to reach critical shipment volumes to obtain lower freight rates. Backed by an empirical study among German freight forwarders acting as IATA cargo agents, he offered managerial implications for planning and implementation of such strategic alliances, which can be regarded at the same time as strategic networking on a horizontal level.
Furthermore, Lunnan and Reve (1995, pp. 357–368) considered the corporate governance structure of Laboremus, a small Norwegian company in industrial shipping, to be a strategic network. Rather than vertically integrating all activities in one single firm, it performs activities like network management and strategy, accounting and finance as well as ship procurement as its core competencies in-house with ship management, marketing and operations actually done by closely related but independent companies around.

While the first two contributions solely draw their theoretical background from Williamsonian TCT, the last one is based on a so-called "contract theory of strategic management" first formulated by Reve (1990) where he attempted to bring Porter (1980) competitive positioning model together with PAT and TCT considerations of organizational governance.⁵⁷² Common to both underlying theoretical concepts is that they provide some rational why such strategic networks are workable in the field of transport and logistics, but fail to explore further how such networks are structured.

This criticism is in line with Weyer (2000) discussing Sydow et al. (1998, 1995) theory of structured inter-organizational networks and concluding that networks are considered to be planned constructs of strategically acting actors, who coordinate their actions to obtain specific benefits. As the focus is set on a specific phenomenon – trustful cooperation – only field studies or historic-reconstructive, qualitative methods are available so that the inherent network structure is just verbally described. Another topic is assessed by Hirsch-Kreinsen (2002, p. 109) who claims strategic networks to represent only one extreme pole of the spectrum, as they are inter-organizational arrangements hierarchically coordinated by one or more dominant firms.

Summing up in a more precise way, the strategic network approach has two main deficits: (1) It is only centred around a focal firm, their organizational governance and the contractual relationships to others being adjacent and (2) as this firm is considered to be somehow dominant, the inherent presumption is that of a

⁵⁷²More specifically, Reve (1990) considered a firm to be a nexus of internal and external contracts offering organizational and inter-organizational incentives in order to bind core and complementary skills to it. The strategic core of a firm can then be defined as a bundle of skills which are of high asset specificity and therefore governed with in the boundary of the firm. Further, complementary skills are of medium asset specificity and can therefore be acquired by vertical, horizontal or diversification alliances forming a strategic network whereas other assets of low specificity are obtained in the market.

hierarchically ordered network structure where the firm is like a queen bee served by their subordinates. Looking at the freight forwarding business in multimodal transport chains as outlined in Sect. 2.2, it is quite evident that the strategic network approach is not too useful in this context.

4.3 Dynamic Networks: Temporary Virtual Companies for Specific Tasks

Another innovative management approach in connection with inter-organizational network structures was provided by Miles and Snow (1984, 1986). Building on their earlier findings, that all kind of organizations need a fit between business strategy, organizational structure and process mechanisms to be successful (Miles et al., 1978), they proposed a new organizational concept called "dynamic networks" as an alternative response to this changing business environment as already described in the last section.

4.3.1 Theoretical Background

For them, the main characteristics of such dynamic networks come along with (1) a strict vertical disaggregation of business operations, (2) external and internal brokerage, (3) employment of market substitutes for administrative purposes and (4) supportive full-disclosure information systems, so that:573

Business operations are performed within a group of more or less independent organizations, instead of being vertically aggregated inside the hierarchy of a single company.
At least one single broker is responsible to assemble all resources needed by employing these network members specialized in production of components and services to end in a joint product or service for external sale.⁵⁷⁴
Instead of hierarchical organization structures of plan and control, market mechanisms are employed to administer these resources inside this network of business operations.

⁵⁷³Cf. Miles and Snow (1984, pp. 26–27, 1986, pp. 64–65) or Rennings (1992, pp. 40–41). Further Siebert (1991, pp. 294–298) with a quite similar approach to characterize corporate networks along functional specialization, market mechanism efficiency, trust, cooperative conduct and information integration.

⁵⁷⁴Later on, Snow et al. (1992, pp. 14–19) discussed fundamental different functional roles of such an broker in establishing (the "architect"), operating (the "lead operator") and maintaining (the "caretaker") such dynamic networks.

• A full-disclosure information system is established as a substitute for lengthy transaction procedures between all network members in order to ease e.g. ordering or financial flows.

As such dynamic networks rely on a radical disintegration of business operations in favour of a network of resources, they are considered to be highly adaptive as they allow to quickly reshape their organizational structure whenever necessary in order to meet a complex and ever changing competitive environment. As Ochsenbauer (1989, p. 275) or Rennings (1992) noted, this network concept is closely related to Piore and Sabel's (1985, pp. 286–307) more global notion of "flexible specialization" as a "shift of the technological paradigm" away from mass production based on vertical integration and a rigid allocation of resources to produce low-cost standardized products towards a more customized, adaptive way of production.575 More specifically, they postulated the renaissance of the former artisan method of production as a path to prosperity in saturated markets with demand instability and increasing uncertainties in the business environment.

Having usually one face to their customers, such dynamic networks can be regarded at the same time as "virtual corporations," a term first coined by Davidov and Malone (1993, [1992]) in their rather visionary management reader. In the following, this term became an often cited metaphor for either (1) small company business backed by a much larger network of easily accessible external resources or (2) temporary business networks of otherwise independently operating companies joining their resources for a specific purpose.⁵⁷⁶

4.3.2 Applications and Critical Comments

While referring to the works of Piore and Sabel (1985), Miles and Snow (1986) and Ochsenbauer (1989), Rennings (1992) proposed the idea that freight forwarders can be regarded as a brokerage element in dynamic networks built on the principle of flexible specialization, but failed to go into more details. Furthermore, in a study focusing on the impact of information processing technology on transaction costs in multimodal transport chains for deep-sea container shipping, Kylaheiko et al. (2000) considered all participants involved in physical goods, documentary and/or financial flows in such transport chains to form a single virtual organization, or better say, a virtual corporation vis-`a-vis the seller, headed by the freight forwarder as the central broker.

Following these authors, a freight forwarder might be considered to be a single broker in a dynamic network assembling together different specific resources to

⁵⁷⁵See also Sydow (1993, pp. 185–188) with a critical review.

⁵⁷⁶Cf. Voss (1996), Christie and Levary (1998) or Corsten and Gossinger (2000, pp. 24–51) with ¨ further references.

bundle a final product in form of a tailored door-to-door service. If then a shipper wants him to expedite a shipment, he should act like a single virtual corporation assembling all resources needed for fulfilment.

But this illustration of the freight forwarding business seems to be somehow misleading, because most freight forwarders cannot be regarded to be such a simple broker only organizing tailored door-to-door shipments by employing temporary network relationships upon request. As already shown in Sect. 2.1.4, freight forwarders are usually engaged in much more fields of activity than such single transport brokerage at the same time like transport operations on own account, cargo consolidation and groupage service on a permanent basis or they offer longterm contractual agreements, including warehousing and value added services in logistics.

4.4 Entrepreneurial Networks: Specialized Intermediaries Economizing on Information Costs and Hierarchical Network Structures

In a series of contributions on international business issues, Casson (1990, 1993, 1997a) developed an information-based view of the economy as a system of structured information flows heavily influenced by entrepreneurial activity. For him, entrepreneurs specialized in information processing are the driving force to allocate resources properly on an international (as well as a national) scale. Free market economies backed by a legal framework of property rights as well as a moral framework of trust encourage them to intermediate effectively between factor input owners and prospective consumers based on their knowledged decision making. Doing this, they facilitate exchange of goods(and services) by enabling people to get in touch with each other, to communicate needs and wants, to explain what they want to offer, to negotiate prices and to monitor performance. To clarify his solely verbal reasoning on economic behaviour, he introduced a system of schematic diagrams as a flexible tool for comparative-static analysis, which can be also employed to survey network structures.⁵⁷⁷ Although this approach to network structure is rather isolated and not well reflected in mainstream economic literature, it seems to be another viable starting point for a network perspective of international freight forwarding under the label "entrepreneurial networks".⁵⁷⁸

⁵⁷⁷See e.g. Casson (1993; 1997a, pp. 6-9 and 124–139, 1997c, pp. 157–162) or Casson and Cox (1997, pp. 183–189) for a general introduction.

⁵⁷⁸Initially, he talked about "business networks", "inter-firm networks" or "intermediated networks", see Casson (1997a, pp. 117–145) or Casson and Cox (1997). Later on, Casson (1997b,c) sticked on the term "entrepreneurial networks", which should indicate that "[i]n the aggregate, entrepreneurs may work better as a cooperative network than as a collection of competitive individualists" (Casson 1997b, p. 811).

4.4.1 Theoretical Background

Drawing on and developing further seminal work on the economy as an information system by Hayek (1937), Richardson (1960) and Marschak (1974) as well as on the nature of entrepreneurship by Knight (1921) and Schumpeter (1931), Casson (1996, 1997a,c, 1999) questioned the main lines of thought in NIE as shown in Sect. 3.1.3 to a great extent. More specifically, he felt quite uncomfortable about their overstatement of bounded rationality, information asymmetries, opportunism and asset specificity problems as well as the overall disaggregated modelling approach in TCT and PAT emphasizing too much on "the adversial use of information rather than the use of information for creative purposes" (Casson 1997a, p. v). He wanted to provide instead an integrated approach along a concept of information cost economizing behaviour with a "theory of the firm centred on the entrepreneur as the founder and the prime-mover within it" (Casson 1997a, p. 114).579 Furthermore, he wanted to highlight the value of intermediation as a real value-adding activity, which is heavily negated by NIE, too.580

According to him, the information-based economy can at best be described as an information system, where handling of information flows relating to goods and services direct material flows (see e.g. Casson 1997a, pp. 3–9). Therefore, "the structure of institutions existing at any time can be interpreted as a rational response to the social needs to economize on information costs. As information costs change, so too [sic] does the institutional structure of the economy" (Casson 1997a, p. 3). Furthermore, he defined information costs as the costs of collecting, communicating, and memorizing of information as well as the costs of calculation and forming judgements required to make reasonable decisions.581 They do not occur only in connection with actual economic transactions or agency relationships, but evolve during the whole process of decision-making in a firm like forecasting market demand, monitoring of competitors or appraising investments.⁵⁸² So he considered his information cost approach as a more general concept than TCT or PAT coping with transaction costs and/or agency costs.583

⁵⁷⁹Following Casson (1996, p. 60, 1997a, p. 81), "[...] a firm may be defined as a specialized decision-making unit, whose function is to improve coordination by structuring infromation flow, and which is normally endowed with privileges [...]." Further, he considered (1) limited liability, (2) indefinite life and (3) favourable taxation rules as the main privileges of a firm, see Casson (1996, pp. 59–60, 1997a, pp. 80–81).

⁵⁸⁰According to Casson (1997a, pp. 5–6), 1999, pp. 79–80), NIE as well as the international business literature mainly emphasizes that the intermediary's role is just to organize production as the core function of a firm with raising mainly three issues: (1) The provision of technology, (2) the coordination of intermediate product flows and (3) the direction and supervision of labour involved in the underlying production process.

⁵⁸¹Cf. Casson (1997c, p. 151).

⁵⁸²Cf. Casson (1997a; 1997c, pp. 151–152, 1999, p. 80).

⁵⁸³Actually, Casson (1997a, 1999, pp. 85–86) mainly stressed on the difference between information costs and transaction costs, but the same arguments hold in connection with agency costs, too.

Another important factor in his framework is the information-intensive activity conducted by intermediaries, who usually buy goods in order to resell them later, perhaps after some transformation.⁵⁸⁴ Mark Casson called this intermediary activity "market-making" and distinguished between retailers intermediating in both physical good and information flows while market-makers in his sense intermediate only in information flows.⁵⁸⁵ Furthermore, he added another aspect of intermediation, namely the ownership of goods, which may change many times in a free market economy. For him, taking title on goods is a means to extract rents from information accumulated in advance, because ownership on goods gives power to realize benefits from information-improved decision-making.⁵⁸⁶ Following Casson (1997a, pp. 35–75, 1997c, pp. 155–157), these intermediaries specialized in (1) search, (2) specification, (3) negotiation, (4) completion and (5) enforcement tasks are the driving force of a sequential and localized coordination process to allocate resources efficiently. Doing this on a regular basis, they perceive not only considerable economies of scale in the actual process of intermediation but also lower transaction costs and create a chain of trust needed for successful business transactions.⁵⁸⁷ Moreover, this intermediating activity essentially captures the spirit of entrepreneurship as formulated by Schumpeter (1931) considering innovations not only to address technological issues, but also to encompass creating new products, developing new markets and tapping new sources of supply.588

Expanding this theoretical framework of an information-based economy to a network perspective, Casson (1997a, pp. 117–145, 1997b, 2000c) as well as Casson and Cox (1997) considered again that physical goods flows are being coordinated by information flows. Furthermore, they considered social bonds to lower information

⁵⁸⁴See Casson (1997a, pp. 15–16, 1999, pp. 80–84). However, despite it is explicitly mentioned, this is again valid for the intermediation in services, too, as already shown in Sect. 3.3.

⁵⁸⁵Notably, he used the terms "intermediation" and "market-making" with a slightly different meaning in contradiction to the strict definitions in Sect. 3.2. But all in all, he came to the same conclusions.

⁵⁸⁶According to Casson (1997a, p. 16), another reason of goods ownership in this context is that by taking title on goods, "the possessor of information insures potential users of the information against the consequences of its being false." This in turn forces the possessors of the information to ensure their quality for beneficial decision-making.

⁵⁸⁷Cf. Casson (1997a, pp. 87–89, 1997c, pp. 155–157).

⁵⁸⁸See Casson (1997b, pp. 815–816, 1999, p. 81, 2000c, p. 265) referring to Schumpeter (1934) and Casson (2000b, pp. 228–235) for a more detailed discussion about contributions by Joseph Alois Schumpeter and the Austrian School to the theory of entrepreneurship in the context of international business. Notably, there seems to be a deviation between Schumpeter (1931) and Schumpeter (1934) in connection with "new markets" and "new sources" in question, because Mark Casson implied them to be found only abroad, whereas Schumpeter (1931) believed them just being other markets or sources. Further, Langen and Welters (1999) researched innovation issues in the freight transport industry and came to the conclusion, that radical innovations in transport technology (they called it "hardware") seem to be of diminishing importance. But innovation in the organization of business practice (denoted as "orgware") are needed to cope with the dynamic environment.

costs, which leads to an increased sharing of information between entrepreneurs forming entrepreneurial networks. There are two crucial components which allow to establish and maintain such network structures:

First, there is a need for trust-worthy relationships between cooperating entrepreneurs. In this context, Casson (1997a, pp. 118–123) proposed a strict rational approach to trust: "It defines trust as a warranted belief that someone else will honour their obligations, not merely because of material incentives, but out of moral commitment, too" (Casson 1997a, p. 118). To him, trust is either generated through material incentives mechanisms backed by a legal system of property rights or it relies on emotional incentives mechanisms arising from social bonds and moral commitments.⁵⁸⁹ As the obstacles of such moral trust mechanisms are high, they tend to work best in tightly connected social groupings with a common moral commitment like a family, church, school, university, military service, clubs, professional associations and/or political parties.590
Following Casson (1997a, pp. 123–129), a second component of entrepreneurial networks is the dispersion in actual ownership of goods as well as resources employed. And as for him ownership is usually based upon competence due to better decision-making, a wide distribution of ownership implies a wide distribution of competence underpinning it. In addition to this, a relative uniformity of network members not only in terms of competence but also in terms of optimism and risk aversion support such network structures, because it reinforces trustworthy relationships, too.⁵⁹¹

Bringing these two components together, entrepreneurial networks can then be characterized as networks with high-trust relationships based on emotional incentives but subject to a diffused distribution of ownership and competence.⁵⁹² This allows again to make a topology with markets and firms being the other forms of organization along the dimensions of (1) intensity of trust and (2) distribution of ownership (including competence).593 Moreover, the issue of intermediation can

⁵⁸⁹In both instance, adequate investments have to be undergone to perceive warranted trust: either establishing a quite impersonal system of monitoring and sanction enforcement or building up and maintaining social ties with other network partners based on a system of common moral beliefs, see Casson (1997a, pp. 119–120, 1997b, pp. 814–815, 2000c, pp. 258–259) or Casson and Cox (1997, pp. 177–180).

⁵⁹⁰Cf. Casson and Cox (1997, pp. 180–182) or Casson (2000c, pp. 262–264).

⁵⁹¹Another aspect of this sort of uniformity among the participants of an entrepreneurial network is a standardized language and culture, which actually reduces communication costs, see Casson (1997c, p. 814, 2000c, p. 258) referring to Casson (1997a, pp. 26–27).

⁵⁹²Cf. Casson (1997a, pp. 123–124, 1997b, p. 815, 2000c, p. 259).

⁵⁹³Following Casson (1997a, pp. 123–124), "[t]he difference between the markets and the networks lies in the degree of trust involved, while the difference between the market and the firm lies principally in the distribution of competence instead." Further, he distinguished between high-trust firms (or clans) relying on internal networking as well as emotional incentives mechanisms and low-trust firms (or hierarchies) with material incentive mechanisms.

be easily introduced in this theoretical framework by considering intermediaries as multi-affiliated entrepreneurs in privileged positions between network members adjacent to them.594 More specifically, they intermediate in information flows instead of integrating in ownership forward and/ or backward for the purpose of an improved physical goods flow coordination. To fulfil these promises, an intermediary first needs some superior competence *vis-`a-vis*the others which allows him again a better decision-making. Furthermore, he should possess a wide-ranged personal network of high-trust relationships to often spatially dispersed individuals belonging to different social groupings as already described above, which gives him a business opportunity to mediate between them upon request.

Casson (1997b, 2000c) went a step further, considering that a special feature of entrepreneurial networks is that if there are several subnetworks of tightly connected social groupings with overlapping memberships existing, members of a higher-level subnetwork specialize in belonging to several lower-level subnetworks which leads to a hierarchical network structure. Then an intermediary being a member of a higher-level and a lower-level subnetwork at the same time can be considered as an information hub being able to mediate between different lower-level subnetworks by employing their high-trust relationships to other higher-level subnetwork members who are again members belonging to other lower-level subnetworks otherwise nonreachable.⁵⁹⁵ Furthermore, it can be asserted that "the lower the level of the group, the greater is the focus on supporting routine operations and the stronger is the consequent incentive to channel information through a hub" (Casson 1997b, p. 821). A higher-level subnetwork of entrepreneurs, however, tend to be more democratic and collegial, reflecting the dispersion of expertise among their spatially dispersed members.

4.4.2 Applications

Only quite a few direct applications of Mark Casson's entrepreneurial network theory can be found in literature, with a focus set on research around historical international business practices and long-term developments in international trade:

• In Casson (1997a), three cases are discussed: (1) a comparison of two industrial districts – namely the Italian textile industry of the 1970s in the province of

⁵⁹⁴See Casson (1997a, pp. 129–130) or Casson and Cox (1997, pp. 186–189) for this and the following.

⁵⁹⁵Moreover, such higher-level subnetworks can play an important role in synthesizing information, too, because "it is often cheaper to get information and advice from other people [...]. "Who you know" is often more important than "what you know" because the people that you know can plug the gaps in what you know. This, of course, depends on knowing the right people. Sometimes the right people are those who know a lot of facts, but more often they are people simply know a lot of other people who in turn know useful facts. These people can act as brokers, linking the decision-maker who demands the information to the person who ultimately knows the facts" (Casson 2000c, p. 257).

Prato and the iron industry at Merthyr Tydfil in South Wales between 1750 and 1850 with a focus set on the important role of intermediating entrepreneurs there, (2) free-standing firms during the age of imperialism as a distinctive form of an international business enterprise and (3) a case study showing how chartered trading companies became the dominant market making firms on colonial trade routes from the seventeenth to the nineteenth century.

Casson and Cox (1997, pp. 189–194) examined rather general strategical and operational issues connected with British international business networks in the nineteenth century.
Casson (1998) conducted an economic analysis of multinational trading companies, where he discussed topics like the difference between market making⁵⁹⁶ and brokerage in an international dimension and vertical integration of such companies in production, shipping, banking and/or finance.
Casson (2000b, pp. 235–249) showed the evolution of the international business system from subsistence to global economy of today in a sequence of schematic diagrams with a focus set on the British economy.
Popp (2000) observed clothing supply chains of a contemporary fashion market retailer and showed that intermediaries add value there by handling information.

By adapting this schematic diagram approach to entrepreneurial networks to the context of this work, a stylized model of hierarchical network structures in multimodal transport chains can be established. Players in this model are (in addition to transport middlemen *TM*x, transport service providers *TP*x, and logistics service providers LP^x as defined in Sect. 3.3) primary sector firms Px, industrial sector firms Ix, distributors Dx, retailers Rx and customers Cx, situated in two countries called h (home) and f (foreign) and indicated by the subscript x. Normally, primary sector firms are suppliers of raw materials and/or intermediate goods to industrial sector firms, who process them to final products ready for distribution in both countries through domestic outbound logistic chains including some distributors, retailers and customers at the end. The resulting entrepreneurial network is shown in Fig. 4.2, where the thick lines stand for transport service providers employed in physical movement of goods between nodes, and the boxes standing for different participants handling and/or transforming goods along their way from Px to Cx at home (^x ^D ^h) or abroad (^x ^D ^f ). For example, if a domestic distribution of final products takes place, the goods are moved downstream from the firms of the industrial sector Ih to their distributors Dh by some domestic transport service providers TP^h as performing carriers.⁵⁹⁷

⁵⁹⁶In this contribution, Casson (1998, pp. 24–29) used the word "reseller" instead of "market maker" in order to stress on the fact that these sort of intermediaries take title on goods in order to resell them later.

⁵⁹⁷Notably, these performing carriers are either third-parties specialized in physical goods movement or belong to primary sector firms, industrial sector firms, distributors, retailers or customers who move their physical goods on own account.

Fig. 4.2 Physical goods movements in multimodal transport chains

Due to the mostly multimodal character of cross-border good flows and the already mentioned additional problems in international transports operations, international distribution of final products is more complex. First, the logistics service providers LP^h and LP^f represent points of transhipment and/or consolidation between different modes of cross-border shipment (e.g. ocean shipping between home and abroad by *TP*hf ) and inland transportation (e.g. trucking by TP^h at home and TP^f abroad). Looking at the information flows in Fig. 4.3, a second deviation from domestic distribution patterns is quite evident: Transport middlemen are managing the cross-border transportation of physical goods by employing transport and logistics service providers for physical goods movement.⁵⁹⁸ They can be considered as intermediaries specializing on information flows (indicated by thin single arrowed lines) between shippers, transport and logistics service providers in different countries due to some superior competence in cross-border logistical processes (e.g. documentary paperwork and customs procedures) and a higherlevel entrepreneurial network with high-trust relationships to colleagues abroad (shown as thin double arrowed lines). This enables them to manage cross-border transportation of physical goods from door to door along a multimodal transport chain, which seems to be the main reason why transport middlemen, customs house

⁵⁹⁸Notably, all information flows at both ends of each transport operation performed by one of the many transport service provider as shown in Fig. 2.4 are omitted in Fig. 4.3. in order not to overload the diagram.

Fig. 4.3 Information flows in multimodal transport chains

brokers or NVOs as specialists and international freight forwarders as generalists are often regarded as being "the Architects of Transport" in international transport and logistic chains.

4.4.3 Critical Comments

Unfortunately, this stylized model framework cannot be quantified straight-forward, but travel trips of shipping company representatives as described by Miller (2003) and the traditional correspondence system of international freight forwarders as described by Fischer et al. (1930, pp. 310–311), Schlichting (1931, pp. 78–81), Mayer (1933, pp. 64–65), Kirchner (1950, pp. 78–86) or Fuhrmann (2000) as well as horizontal alliances of small and medium sized freight forwarders in traditional air cargo logistic chains (see Neiberger, 2007 and Sect. 2.2.4.3) strongly support such a view.

In addition to this, actual international freight forwarders' disintermediation threats can be clarified shortly. First, transport service providers on the crossborder main haul may feel able to vertically integrate forward and/or backwards in landbased operations (including at least provision of documentary paperwork, customs clearance and airline or carrier haulage on pre- and onward-carriage legs). Then they are able to offer door-to-door services directly to industrial sector firms as well as distributors, bypass other transport intermediaries and make them obsolete. At the same time, exporting industrial firms as well as importing distributors may seek to engage themselves more actively in their cross-border goods movement and attempt to organize them in-house. Then they again bypass the transport intermediaries and build up direct relationships with logistics and transport service providers at home and abroad in order to keep the physical goods moved along the logistic chain under their control. This is a viable approach for shippers with a considerable expertise in cross-border goods movements and constantly large shipment volumes like automobile manufacturers feeding assembly lines all around the world on a regular basis. In both instances, building up and maintaining relationships beyond the status quo as shown above may result in an increase of control of information flows which direct physical goods movements.

4.5 Industrial Networks: Interdependent Business Relationships Within a Network of Interrelated Actors, Activities and Resources

Starting again in the 1980s, a network view of industrial marketing and purchasing (IMP) gained considerable popularity, especially among Scandinavian researchers or other members of the IMP research group being founded in the mid 1970s to conduct a series of cross-country empirical studies on industrial markets. According to Easton (1992, p. 3), this industrial network approach

"[...] shares with other approaches a belief that the existence of relationships, many of them stable and durable, among firms engaged in economic exchange provides a compelling reason for using inter-organizational relationships as a research perspective. It differs from other approaches mainly in terms of its scope. It is concerned to understand the totality of relationships among firms engaged in production, distribution and the use of goods and services in what might best be described as an industrial system."

4.5.1 Theoretical Background

Following Johanson and Mattsson (1988), Easton (1992) or Mattsson (1995), the antecedents of this network approach lie in (1) US-American marketing channels research as already briefly discussed in Sect. 2.2.1, (2) internationalization process theories of industrial firms as developed e.g. by the Uppsala School and (3) a series of empirical studies on interaction and relationships in industrial markets conducted in the late 1970s.⁵⁹⁹ Probably the most direct predecessor is the last mentioned

⁵⁹⁹See e.g. Hertz (1993, pp. 33–63) for a thoroughly discussion.

"interaction approach" initiated by the first IMP research project, where a group of European researchers joined together in the IMP research group developing a dynamic model of buyer–supplier relationships in industrial markets (the interaction model) and illustrating its applicability through comparative studies of buyer– supplier relationships within and across a number of European countries (namely France, Germany, Italy, Sweden, and the United Kingdom).⁶⁰⁰ The main conclusion of these initial pan-European studies was that buying and selling of goods and services in industrial markets could not be understood as a series of isolated and serially independent market transactions. Instead, these transactions should only be examined as episodes in often long-lasting, complex, relatively balanced and quite informal business relationships between them.⁶⁰¹

Shortly after these initial empirical studies focusing on buyer–seller relationships, researchers of the second IMP research project began to explore the implications of such strong and stable inter-organizational relationships for the network structure of industrial markets in general.⁶⁰² Whereas markets were traditionally regarded as being either totally frictionless (NCE) or burdened with some transaction and agency costs (NIE), empirical evidence from the first IMP project demanded a new treatment of industrial market settings. Constantly referring to the notion by Cook and Emerson (1978) that networks can be defined as sets of connected exchange relationships, industrial markets were considered as networks of heavily interdependent business relationships.⁶⁰³ According to this "markets-asnetworks" approach, market coordination takes place through ongoing mutually oriented interaction between reciprocally committed counterparts closely interrelated with third parties or other business relationships.⁶⁰⁴

More specifically, each business relationship between two actors in an industrial network can be described along two dimensions, namely their (1) layers of substance and (2) levels of function. The first dimension characterizes present business

⁶⁰⁰See e.g. Johanson and Mattsson (1988, pp. 288–290) or Ford and Hakansson (2006) for a brief ˚ overview, the findings of this IMP research project are well documented in Hakansson (1982) or ˚ Turnbull and Valla (1986).

⁶⁰¹Cf. Hakansson and Snehota (1995, pp. 7–8). ˚

⁶⁰²See Hagg and Johanson (1982) or Hammarkvist et al. (1982) for early treatments, followed ¨ by Axelsson and Easton (1992) or Hakansson and Snehota (1995) presenting the findings of this ˚ second IMP project.

⁶⁰³Cf. Johanson and Mattson (1992, p. 209) or Axelsson (1992, p. 239) or Hakansson and Johanson ˚ (1993, p. 32). Notably, in this work Emerson and Cook (1978) is associated with network exchange theory (NET) as an extension of social exchange theory (SET) to a network level, see Sect. 4.6.1.3. 604Cf. Hakansson and Snehota (1995, pp. 25–26). According to H ˚ akansson and Snehota (1995, ˚ pp. 9–18), four interaction process characteristics (adaption, cooperation and conflict, social interaction and routinization) and at least five contexts of relationship interdependencies (available technologies, accumulated knowledge, social relations, administrative and systems as well as legal ties) were prevalent. Further, see Easton and Araujo (1994) for a similar description along the social and temporal context of market exchanges.

relationships as an entity of activity links, resource ties and actor bonds representing three layers of substance (Hakansson and Snehota (1995, pp. 26–27) ˚ 605:

"*Activity links*regard technical, administrative, commercial and other activities of a company that can be connected in different ways to those of another company as a relationship develops.
Resource ties connect various resource elements (technological, material, knowledge resources and other intangibles) of two companies. Resource ties result from how the relationship has developed and represents in itself a resource for a company.
Actor bonds connect actors and influence how the two actors perceive each other and form their identities in relation to each other. Bonds become established in interaction and reflect the interaction process."

As a consequence, an industrial network consists not only of actors and their mutual business relationships, but also of changing activities employing resources and bonds reflecting an endless chain of interaction. The second dimension goes a step further and describes more far reaching impacts of such business relationships: As an industrial network is considered to be heavily interdependent by nature, they probably affect and are affected by third parties or other business relationships. Therefore, Hakansson and Snehota (1995, pp. 27–28) distinguished at least three ˚ different, but closely interwoven functions of business relationships:

"Function for the dyad: This originates in the conjunction of the two companies; their activities, resources and actors. Activity links, resource ties and actor bonds in a relationship integrate various elements and thereby some unique outcomes and effects are produced.
Function for the individual company: A relationship has effects on each of the companies, on what it can do internally and in other relationships. These depend on how what is produced in the dyad can be connected to other internal elements of the company and its other relationships.
Function for third parties: Being a building element in the larger network structure, what is produced in a relationship can affect and is affected by other relationships that involve other parties. The effects on third parties and from third parties and their relationships on the relationship in any of the three layers of substance depend on how tight the connectedness of relationships is in the overall network."

Furthermore, these three levels of business relationship functions allow three different levels of analysis: (1) companies, single actors or other forms of organizations engaged in one or more business relationships, (2) the dyads resulting from

⁶⁰⁵See Hakansson and Snehota (1995, pp. 24–49) for this and the following. Other, much shorter ˚ overviews of this "markets-as-networks" approach can be found e.g. in Johanson and Mattsson (1987, pp. 34–41, 1988, pp. 448–453), Easton (1992, pp. 8–16), Hakansson and Johanson (1992, ˚ 1993) or Mattsson (1995, pp. 206–209).

Fig. 4.4 Analytical scheme for industrial networks (Source: adapted from Hakansson and Snehota 1995, pp. 43–45) ˚

establishing such business relationships and (3) their impacts on third parties (and vice versa) forming a network structure.⁶⁰⁶ Putting then both dimensions together, a conceptual framework for industrial network research emerges in form of a broad analytical scheme as shown in Fig. 4.4. There are three different levels of analysis (organization or single actor, business relationship and network structure) with each level having three layers of substance (actors, activities and resources). As indicated by bidirectional arrows, all these elements are considered to be connected by mutual interdependencies.

Following Hakansson and Snehota (1995, pp. 44–46), this broad analytical ˚ scheme can be used either "as a conceptual framework to analyse the effects of change in a relationship and/or identify the factors that effect the possibilities of development of a relationship" or "as a heuristic device in coping with relationships in business." But looking at Fig. 4.4, it is quite evident that other perspectives than "networks as relationships" (Easton 1992, p. 8) can be developed out of this conceptual framework. Especially Easton (1992, pp. 16–25) proposed "networks as structures," "networks as position" and "networks as process" to be three

⁶⁰⁶Notably, Hakansson and Snehota (1995) defined "company", "relationship" and "network" to ˚ be the levels of analysis. But for sake of internal consistency of this work, these levels of analysis were renamed in "organizations or single actors", "business relationships" and "network structure" without alternating their theoretical content.

alternative points of view arising from this industrial network approach apart from the relationship perspective as the most important one:

Networks as structures: As the elements of an industrial market system are interdependent rather than independent, they sum up to a network structure, which can be described along measures like structuredness (or general level of interdependency among actors and their relationships), homogeneity (the similarity of actors and their relative importance) and exclusiveness (the extent to which the actors in focus hold relationships to other networks outside).607
Networks as position: Focusing on an organization or a single actor as elements inside an industrial network structure, their position can be defined as a result of past and present activities both by them and by others.⁶⁰⁸ In this context, several concepts of position were proposed e.g. by Johanson and Mattsson (1988).609
Networks as process: First, as long as industrial networks are considered to be stable but not static, an ever ongoing evolutionary process of interaction shapes them over time, which is worth to be observed more closely. Second, coordination processes taking place in industrial networks can be conceived as being effected by three classes of mechanisms, namely market, hierarchies and networks with the last considered to be a specific heterarchical governance structure influenced by a balance between cooperative and competitive elements.⁶¹⁰

4.5.2 Applications

Although this network approach was initially dedicated to industrial markets mainly dominated by manufacturing companies, there are indeed some contributions with a focus on transport and/or logistics topics, which will be discussed briefly in the following.

Hertz (1993, 1996a,b) first employed the industrial network approach to develop a dynamic network model of internationalization in the transport industry based on a series of event studies tracking the internationalization processes of three large Swedish freight forwarding companies and their resulting transport network

⁶⁰⁷See Johanson and Mattsson (1988, pp. 294–294) or Easton (1992, pp. 16–19) referring to an earlier work by Mattsson (1986).

⁶⁰⁸Cf. Johansson and Mattsson (1988, pp. 292–295, 1992, pp. 211–214) or Easton (1992, pp. 19–21).

⁶⁰⁹More specifically, Johansson and Mattsson (1988, p. 293) distinguished two types of positions: "A micro-position is characterised by: (1) the role the firm has for the other firm; (2) its importance to the other firm; and (3) the strength of the relationship with the other firm. A macro-position is characterised by: (1) the identity of the other firm with which the firm has direct relationships and indirect relations in the network; (2) the role of the firm in the network; (3) the importance of the firm in the network; and (4) the strength of the relationships with other firms."

⁶¹⁰Cf. Easton (1992, pp. 21–25), Axelsson (1992, pp. 239–242), Hakansson and Johanson (1993, ˚ pp. 44–49) or Hakansson and Snehota (1995, pp. 19–21). ˚

transformation over a period of almost fifty years. Her study showed typical patterns of internationalization in the freight forwarding industry and complications occurring in connection with the adoption of their transport systems after crossborder expansion of operations.

Furthermore, in this already in Sect. 3.1.3.1 mentioned treatment of freight forwarder transport networks, Stahl (1995a, pp. 122–134) discussed the industrial network approach as an additional theoretical background for his purely transaction cost-driven treatment of organizational governance. For him, the main reasons not to follow fully this industrial network perspective was, that he wanted to generate and prove hypotheses by an empirical survey but not conducting a longitudinal study of strategic alliance formation.

It was then Ludvigsen (2000), who underwent a longitudinal case study about a Nordic freight forwarder network called the E-1 Alliance from the early beginnings as an Inter-Nordic logistic alliance to an expansion of their operations on the European continent in order to derive a verified model of alliance evolution based on industrial network methodology.

Recently, Skjøtt-Larsen (1999a,b), Mattsson (2003), Halldorsson et al. (2004, ´ 2005, 2007) as well as Gadde and Hulthen (2009) discussed t ´ he applicability of the industrial network approach for third-party logistics outsourcing and/or supply chain management treatments. The main conclusions of them were, that it is useful especially when the focus of investigation is set on the social context and dynamic aspects of long-term relationships.

4.5.3 Concluding Comments

The industrial networks approach first contrasts sharply with other more economic views of business relationships in network structures as shown up to now, especially those that regard such inter-organizational relationships as formal cooperative agreements for a competitive advantage like in the strategic or dynamic network approach. Indeed, it shares some concerns with Williamsonian TCE (e.g. most of the basic assumptions and what accounts for the existence of different governance structures), but rejects its comparative-static framework methodology, as well as the assumption of ever present actor opportunism.⁶¹¹ Furthermore, the concrete contractual form of business relationships is regarded to be of less importance, because they are not considered to be a series of rather isolated transactions being either market-related or hierarchically structured but emerge through an ongoing interaction process of establishing, maintaining and terminating them as a governance mode of its own with boundaries between individual organizations tending to be rather unclear. Moreover, different levels of analysis (including

⁶¹¹See especially Johanson and Mattsson (1987, pp. 43–46), Easton and Araujo (1994), Hakansson ˚ and Snehota (1995, pp. 379–381) or Skjøtt-Larsen (1999a, 1999b, 2000) for this and the following.

different perspectives for actors, dyadic relationships, networks and the ongoing development process) are possible, whereas in TCE solely a pure relationship perspective is offered. Taking all the intellectual roots and ancedents of the industrial networks approach as shown above into consideration, Easton and Araujo (1994, p. 82), drew the conclusion that

"[i]ts positioning is half-way between economics and sociology. [...] It is concerned with how different types of actors transform, consume, create and use economic resources in industrial systems and is prepared to formulate complex models of micro-behaviour taking into account insights generated by sociological theory."

4.6 Social Networks: Network Structure, Positions and Relationships

After exploring diverse economic arguments of network structure, this last section is about social networks of economic action. The idea, that social phenomena can be interpreted as networks of relations has been recurrent since Georg Simmel's formal sociology and his studies of relations between actors in triads and dyads in Simmel (1908).⁶¹² In the following, researchers originating from various disciplines in the social sciences, including psychology, sociology, polity, anthropology and (to some extent) economics were attracted to conduct research on social networks and their contributions summed up to an exorbitant bulk of literature about this truly multi-disciplinary approach to network structures.⁶¹³ But it took until a breakthrough article by Granovetter (1985), promoting a new approach to economic sociology to overcome an inherent oversocialized view of economic action common to most works of sociologists, anthropologists, political scientists and historians in this field.⁶¹⁴ After heavy criticism of both over- and undersocialized conceptions of man in sociology and economics neglecting ongoing structures of social relations, he suggested an embeddedness concept of economic action.⁶¹⁵ This paper was a starting point for intensive research efforts under the label of New Economic Sociology (NES) centred around the issue of embeddedness of actors in social network structures and social capital as the value of connections therein.616 In the following, this theoretical core of NES is briefly discussed, before two alternative

⁶¹²Cf. Schenk (1984, pp. 12–17) or Freeman (2004, pp. 10–30) for a more elaborate discussion.

⁶¹³See Sect. 5.1 for more details.

⁶¹⁴Cf. Wrong (1961), Coleman (1988), Swedberg (1991), Smelser and Swedberg (1994).

⁶¹⁵More specifically, Granovetter (1985, p. 485) considered that "both have in common a conception of action and decision carried out by atomized actors; in the undersocialized account, atomization results from narrow utilitarian pursuit of self-interest; in the oversocialized one, from the fact that behavioral patterns have been internalized and ongoing social relations thus have only peripherial effects on behavior."

⁶¹⁶Cf. Swedberg (1991, 1997) or Borgatti and Foster (2003).

analytical approaches to social network structures are shown and some applications from the field of transport and logistics are presented.

4.6.1 Theoretical Background

As already indicated above, NES offers (1) a new sociological way to think about economic action taking place within a network of social relations (the issue of embeddedness) and (2) the sum of these social relations itself is valuable (the issue of social capital). Both offer a theoretical background to catch the inherent structure of social networks.

4.6.1.1 Embeddedness and Social Capital

Granovetter's (1985, p. 487) initial arguments about embeddedness were that "[a]ctors do not behave or decide as atoms outside a social context, nor do they adhere slavishly to a script written for them by the particular intersection of social categories that they happen to occupy. Their attempts at purposive action are instead embedded in concrete, ongoing systems of social relations." In Granovetter (1992, p. 33) he put this issue of embeddedness further as he referred it "to the fact that economic action and outcomes, like all social action and outcomes, are affected by actors' dyadic (pairwise) relations and by the structure of the overall network of relations. As a shorthand, I will refer to these as the relational and the structural aspects of embeddedness."

Furthermore, Zukin and DiMaggio (1990, pp. 15–20) argued that apart from Mark Granovetter's "relational" and "structural embeddedness" there is a cognitive, cultural and political dimension to be considered, too. For them, cognitive embeddedness refers to "the ways in which the structured regularities of mental processes limit the exercise of economic reasoning," a limited ability of actors well-known in the field in NIE under notion of bounded rationality. Cultural embeddedness is then considered to show "the role of shared collective understandings in shaping economic strategies and goals."617 Finally, by political embeddedness they meant "the manner in which economic institutions and decisions are shaped by a struggle

⁶¹⁷Zukin and DiMaggio (1990, p. 17) went into more detail when explaining how economic behavior is culturally embedded: "Culture sets limits to economic rationality: it proscribes or limits market exchange in sacred objects and relations [...] or between ritually classified groups. Moreover, [...] culture may shape terms of trade. [...] Culture, in the form of beliefs and ideologies, taken for granted assumptions, or formal rule systems, also prescribes strategies of self-interested action [...] and defines the actors who may legitimately engage in them. [...] Culture provides scripts for applying different strategies to different classes of exchange. Finally, norms and constitutive understandings regulate market exchange". Further see Dequech (2003) for a closer look at cognitive and cultural embeddedness considering them to be intertwined.

for power that involves economic actors and nonmarket institutions," such as the legal framework of the state. Common to these three aspects of embeddedness introduced by them is that they capture the social environment and not the network structure so that they are not too useful at the moment.⁶¹⁸

The other stream of literature in the field of NES copes with the value of connections itself in a social network structure, called "social capital". Although this term was mentioned several times before,619 the first systematic contemporary analysis can be dated back to Bourdieu (1980, 1983, 1985), where he distinguished three forms, namely economic, cultural and social capital.⁶²⁰ For him, social capital is "[...] the aggregate of actual or potential resources which come along with the possession of a durable network of more or less institutionalized relationships of mutual acquaintance or recognition" (Bourdieu 1985, p. 248) as cited by Portes 1998, p. 3). Alternatively, it can be regarded as resources which determine membership to a certain group (Bourdieu Bourdieu, 1983, pp. 192–193).

In another well cited formulation, Coleman (1988, p. S98) distinguished social capital from physical and human capital and defined it rather vague by its function: "It is not a single entity but a variety of different entities, with two elements in common: they all consist of some aspect of social structures, and they facilitate certain actions of actors - whether persons or corporate actors – within the structure. Like other forms of capital, social capital is productive, making possible the achievement of certain ends that in its absence would not be possible. [...] Unlike other forms of capital, social capital inheres in the structure of relations between actors and among actors." Therefore, social capital can be considered as a useful resource for individuals, like thrustworthiness in the social environment, existence of information channels or effective norms in communities.

Whereas Bourdieu (1980, 1983, 1985) regarded social capital as a resource resulting from relationships forming a social structure and Coleman (1988) defined it as a function of social structure and a resource for actions, Burt (1992, 1997, 1998, 2000) focused on the network structure effect of social capital in offering opportunities.621 He considered financial and human capital to be distinct in from social capital two ways: (1) they are the property of individuals as they are owned in whole or in parts by them and (2) they get invested to create opportunities. Therefore, he sees social capital as "[...] friends, colleagues, and more general contacts through whom you receive opportunities to use your financial and human capital" (Burt, 1992, p. 9). Later on, he asserted that "[s]ocial capital is the contextual complement to human capital. The social capital metaphor is that the people who do better are somehow better connected. Certain people or certain

⁶¹⁸Cf. Uzzi (1996, 1997).

⁶¹⁹See e.g. Woolcock (1998, p. 155) or Anderson and Jack (2002, pp. 194–195).

⁶²⁰The initial treatment by Bourdieu (1980) was in French, the second (Bourdieu, 1983) in German and the third (Bourdieu, 1985) in English, which was concealed in the pages of a text on the sociology of education, as Portes (1998, p. 3) noted.

⁶²¹Cf. Walker et al. (1997, p. 109–112) or Burt (2000, pp. 348–349).

groups are connected to certain others, trusting certain others, obligated to support certain others, depend on exchange with certain others. Holding a certain position in the structure of these exchanges can be an asset in its own right" (Burt, 2000, p. 347).

Although these three approaches to social capital are different at first sight, they have much in common: All emphasize its highly intangible character (in comparison to other forms of capital) and its value as a collective or group-level phenomenon identifying an informal social context that constrains and influences the behaviour of individuals.⁶²²

4.6.1.2 Social Network Structures

After discussing the basic notions of embeddedness of actors and their social capital, the focus switched on the structure of social networks surrounding them. Already before Granovetter (1985), initial works emerged during the 1970s and 1980s, as a group of sociologists around Harrison White (including him) got heavily interested in topics of economic sociology from a network point of view.623 In the following, two distinct streams of argumentation appeared concerning the importance of ties between actors: one about the strength of weak ties by Granovetter (1973, 1982) and Burt (1982, 1997, 1998) and the other by Coleman (1988, 1990) and Krackhardt (1992) emphasizing the strength of strong ties.

Granovetter (1973, 1982) explained the importance of weak ties in bridging subgroups (weak tie argument). For him, the strength of a tie can be defined as "a (probably linear) combination of the amount of time, the emotional intensity, the intimacy (mutual confiding), and the reciprocal services which characterize the tie" (Granovetter 1973, p. 1361). His argument is shortly as follows: Acquaintances with weak ties are less likely being socially involved with one another than close friends with strong ties.⁶²⁴ As the degree of overlap of two individual's friendship networks varies directly with the strength of their ties to one another, a weak tie between them is valuable, because its a bridge between both friendship networks.

Burt (1992, 1997, 1998) refined this weak tie argument of Granovetter (1973, 1982) and showed the benefits of bridging ties from the average strength of these ties leading to a structural view of social capital. There, social capital is regarded as a function of the information and control advantages of an actor who stands in

⁶²²See e.g. Lin (1999), Paldam (2000), Adler and Kwon (2002), Li (2007) or Lee (2009) for this and other facets of social capital not addressed in this work.

⁶²³Examples are Granovetter (1973) analyzing the role that information networks play when people get jobs, Baker (1984) analyzing buyers and sellers in a national securities market or Mintz and Schwartz (1985) as well as Palmer (1983) conducting studies on interlocking directorates.

⁶²⁴Or more precisely described in the words of Granovetter (1982, p. 105):"[T]he set of people made up of any individual and his or her aquaintances will constitute a low-density network (one in which many of the possible relational lines are absent), whereas the set consisting of the same individual and his or her close friends will be densely knit (many of the same possible lines present)."

between otherwise unconnected actors, which gives him the opportunity to exploit this "structural hole".⁶²⁵ Furthermore, Burt (2000, p. 353) noted, that it is "an opportunity to broker the flow of information between people, and control the form of projects that bring together people from the opposite side of the hole."⁶²⁶

Coleman (1988, 1990) extensive personal bonds (strong tie argument), as it is derived from a theory of rational action leads to a relational view of social capital, where it is a function of the inseity and extensiveness of personal bonds among actors. In this perspective, strong ties among actors, intense relationships characterized by mutual trust, a sense of obligation, common norms and expectations create social capital on which the respective actor can draw.627 Finally, employing Granovetter (1973) definition for tie strength, Krackhardt (1992) explored the affective character of strong ties playing an important role especially in cases of severe change and uncertainty.

To end this discussion which form of tie (strong or weak) might be more important for the formation of social capital, some authors like Taube (2002, pp. 72– ¨ 74, 2004) offered a rather Solomonian way out. To him, there are simply two distinct sorts of social capital: (1) strong ties generate support capital and (2) weak ties as well as structural holes offer leverage capital. ⁶²⁸ This distinction is in line with Adler and Kwon (2002) identifying three different views of social capital: (1) bonding, internal capital focusing on collective actors' internal characteristics (Coleman, 1988, 1990), (2) bridging, external capital of external linkages of focal actor as well as (3) a neutral view of social capital in respect of this external/internal dimension. Moreover, they recommend further research to address both external and internal dimension simultaneously.629

All these findings about embeddedness and formation of social capital so far were derived from personal to intra-organizational settings. But they can be easily extended to an inter-organizational context if entrepreneurial activity is observed in a market environment:

• Podolny (1994, 2001) explored ways of organizations to overcome market uncertainty. Podolny (1994) found support for the strong tie argument in investment banking because with increasing market uncertainty, they engage in exchange

⁶²⁵Cf. Burt (1997, 1998) or Ebers (2001).

⁶²⁶Therefore, the benefits of such a brokerage role are – apart from simply exploiting its power of control – having access to information beyond one can process alone, getting information early and obtaining referrals to get own interests in a positive light at right time and place, see Burt (1998). 627Cf. Coleman (1990) or Ebers (2001).

⁶²⁸Following Taube (2004, p. 34), "[ ¨ s]upport capital is typically generated in informal groups with dense relations that are the result of a high frequency of interaction between socially similar persons. [...] Leverage capital comes into existence when an increasing differentiation of a group (as a result of its size) leads to growing distances with less frequent interactions between some of the group members. With widening gaps between smaller groups of high internal density, structural holes [...] come into existence allowing actors that bridge such gaps to exploit weaker relations for their own sake."

⁶²⁹Cf. Adler and Kwon (2002, p. 35).

relations with those they have transacted in the past and with those of similar status. In Podolny (2001) an analytical distinction between two types of market uncertainty was drawn: ego-centric, which refers to a focal actor's uncertainty regarding market opportunities and the set of resource allocation decisions that will best enable him to realize these opportunities, and alter-centric, which denotes the uncertainty confronted by his (her) exchange partners regarding the quality of him (her). Observing venture capital markets, he found out that the value of structural holes increases with ego-centric uncertainty, but not with altercentric uncertainty. In contrast, the value of status increases with alter-centric uncertainty, but declines with ego-centric uncertainty.

Uzzi (1996, 1997) observed how embeddedness and network structure affects economic action in New York's apparel industry. His findings suggest that embeddedness is a logic of exchange, that promotes economies of time, integrative agreements, allocative efficiency, and complex adaption. These positive effects rise up to a threshold, however, after which embeddedness can reduce economic performance by making entrepreneurs vulnerable to exogenous shocks or insulting them from information that exists beyond their network. Therefore, both sorts of social capital are important.
Walker et al. (1997) tested empirically whether social capital (as outlined by Bourdieu (1980, 1983, 1985) or Coleman, 1988, 1990) or Burt (1992) structural holes are more important in the formation of relational networks in high growth technology-intensive industries, taking intercompany collaboration of biotechnology firms as an example. Their analysis showed clearly that network formation is significantly influenced by the development and maintenance of social capital and not by opportunities for entrepreneurs exploiting structural holes which in turn may be applicable more to networks of market transactions.
Fafchamps and Minten (1999) documented the role that personal relationships play in economic exchange while observing agricultural traders in Madagascar. While regularity of supply and demand and the sharing of risk appeared to be of particular importance, larger and more prosperous traders were those with qualitatively better relationships and family ties played little role in business beyond of assistance at start-up.
Rangan (2000a,b) showed that social networks are most significant in those spheres of activity, where search (or identifying potential exchange partners) and deliberation (assessing their quality and intension to deal honestly) are important but problematic. If potential exchange partners are spatially dispersed and when the object of the exchange does not trade at high margins or in high volumes then social networks will tend to matter in the search process as entrepreneurs embedded in a broad social network are better positioned to discover them. Furthermore, in situations where uncertainty coupled with large downside risk which cannot be cost-effectively levied by insurance, deterrence, prediction or internalization, social networks will tend to matter, too. Examining speed and elasticity with which US-American imports respond to the US-dollar's real exchange rate and comparing US-American multinationals with industry
average, he found support for both of his social network explanations about search and deliberation.
Anderson and Jack (2002) and Jack and Anderson (2002) discussed the effects of embeddedness on the entrepreneurial process or, better said, whether the nature of social capital is a glue to form network structures or a lubricant that facilitates the operation of networks. Backed by a considerable number of ethnographs, they found social capital not to be a thing but a process that creates a condition of social capital with both aspects as dimensions of this process. Moreover, they discovered that there exist successful etiquettes of social capital formation, shared values or paradigms that allow a common understanding of appropriate ways of interacting which can be regarded as a third, cognitive dimension of social capital.⁶³⁰
Liao and Welsch (2003) investigated entrepreneurial growth aspiration of US-American business start-ups. Their results suggested that technology-based entrepreneurs benefit more from relational embeddedness coming along with freer and greater exchange of non-redundant information which supports Coleman (1988, 1990) strong tie arguments. In contrary, non-technology-based entrepreneurs benefit more from structural embeddedness or the extensiveness of their social networks consistent with Burt (1992, 1997, 1998) structural holes and Granovetter (1985) weak tie arguments. Therefore, support or bonding capital as well as leverage or bridging capital provide certain benefits, which depend on entrepreneurs' suitable growth strategies.

Common to all of these studies (and many more not explicitly stated above) is, that they showed clearly, in which way the issues of embeddedness and social capital are not only programmatic statements but contribute to a better understanding of entrepreneurship in different market environment settings.

4.6.1.3 Analytical Concepts

As already indicated above, in this wide field of research on social capital and embeddedness issues in social networks at least two methodologically distinct streams of analytical approaches can be identified, namely (1) social network analysis (SNA) and (2) network exchange theory (NET).631

First, since Moreno (1934, 1953) initial efforts in the structural analysis of groups, an ever increasing body of mathematical calculus was employed to describe sets of actors and their relationships between them forming a social network structure in very different contexts. Whereas early rather descriptive treatments of social networks like the famous Hawthorne Studies (see Roethlisberger and Dickson, 1939 and Homans, 1950) or diverse field studies in social anthropology

⁶³⁰See Anderson and Jack (2002) referring to Nahapiet and Ghoshal (1998).

⁶³¹Cf. Cook (1982), Cook and Whitmeyer (1992) or Hirsch-Kreinsen (2002, p. 107).

from African communities to next door neighbourhood relations in European cities (see Barnes, 1954; Bott, 1957; Gluckman, 1955 or Mitchell, 1969) rested simply on graph theoretic expressions and aggregated measures of social network structure by counting actors and relationships, the information processing revolution enabled researchers to apply more sophisticated mathematical methods based on matrix algebra to analyse social network structures empirically.⁶³² According to Wellman (1988, p. 47),

"[s]tructural analysis has become self-conscious and organized. Intellectually, it has moved from a minimalist position, where 'network analysis' was seen as a useful supplementary method, to a maximalist, paradigmatic position, where its central concept – that all social phenomena are best studied through methods designed to uncover basic social structure – is seen as an important new approach to social inquiry. In addition to its critiques of other sociological approaches, structural analysis has now developed a coherent set of characteristics and principles backed by a sizeable body of empirical work."

Recently, Freeman (2004, p. 3) regarded modern social network analysis to be "an organized paradigm for research" with a strong commitment in four features, which together define the field: It is (1) motivated by a structural intuition based on ties linking actors, (2) grounded in systematic empirical data, (3) drawing heavily on graphic imagery and (4) relying on the use of mathematical and/or computational models.

Second, authors like Richard M. Emerson, Karen S. Cook, Toshio Yamagishi, James S. Coleman, Peter V. Marsden, David Willer or Barry Markovsky extended social exchange theory (SET) based on merely micro-sociological works from Thibaut and Kelley (1959), Homans (1961), Emerson (1962) or Blau (1964) to a network level.⁶³³ Following Cook and Whitmeyer (1992, p. 110), these authors

"[...] advance[d] a basic image of social structure as a configuration of social relations among actors (both individual and corporate), where these relations involve the exchange of valued items (which can be material, informational symbolic, etc.). Exchange theory increasingly has involved explicit consideration of social structure, as both product and constraint, typically in the form of networks of social relations."

⁶³²See Sect. 5.1 or merely methodological contributions like Mitchell (1974), Kahler (1975), ¨ Shulman (1976), Burt (1980a, 1982), Lincoln (1982), Knoke and Kuklinski (1982), Schenk (1984), Weiman (1989), Marsden (1990), Scott (1991), Wasserman and Faust (1994), Diaz-Bone (1997) or Jansen (2003).

⁶³³See overviews by Cook (1982), Cook and Whitmeyer (1992), Willer (1992), Sydow (1993, pp. 193–199), Yamaguchi (1996, pp. 308–322) or Jansen (2003, pp. 163–181) for further details. Following them, there are at least three schools of thought to distinguish: (1) the power-dependence theory stream originated in the works of Emerson (1962, 1972a,b) with a series of insightful laboratory experiments like Stolte and Emerson (1977), Cook and Emerson (1978), Cook et al. (1983), Yamagishi et al. (1988), Cook and Yamagishi (1992), (2) the exchange theory of purposive action as first formulated by Coleman (1972, 1973) and further extended by Marsden (1981, 1982, 1983) in order to observe brokerage behavior in networks with restricted access, and (3) a merely graph-analytic approach with contributions like Markovsky et al. (1993, 1988), Patton and Willer (1990), Lovaglia et al. (1995) or Willer (1999, 2003).

As already indicated in Sect. 4.5.3, the NET approach to social network structure has much in common with the industrial network approach with the notable exception that not only actors and their relationships between them are of importance, but also activities by them as well as resources employed.634 Both SNA and NET conceptualize social network structures as a configuration of social relationships and actor positions, but their treatments are very different635:

Being deeply rooted in the SET tradition of modelling social exchange, NET is much more theory-driven than the mostly empirical-driven SNA just attempting to analyse observed structures of social networks.
Whereas NET considers per definition only mutual relationships of social exchange, SNA principally allows to analyse all kind of mutual as well as asymmetric communication, information, negotiation, financial or goods flows as relationships in a given network structure.
While NET considers all relevant exchange relationships in a social network to be of interest (including potentially available exchange alternatives), SNA takes only actually realized connections into consideration.
NET draws a distinction between positive (cooperative, conditional nonexclusive or substitutable), negative (competitive, conditional exclusive or complementary) and mixed connected social network structures, whereas in SNA, there are no such special conditions laid on the configuration of relationships between actors.⁶³⁶

To sum up, it can be concluded that SNA and NET are complement approaches to assess social networks structures: While the first just analyses observable network structures of actors and their relationships as a whole, the latter focuses more on relationships of social exchange between the actors therein. Or in other words: whereas NET focuses on relational embeddedness, SNA deals more with structural embeddedness issues of social networks.

4.6.2 Applications

Until recently, only quite a few works making use of social network theory in the field of transport and logistics could be identified like Cisic et al. (2000a,b) exploring flows of documentary paperwork in maritime transport chains or Carter et al. (2007a) surveying the relationships of participants in a cross-functional

⁶³⁴Cf. Easton (1992, pp. 5–6), Axelsson (1992, pp. 239–240), Hakansson and Johanson (1993, ˚ pp. 35–36), Araujo and Easton (1996, pp. 99–102) and Batt (2008, pp. 488–489).

⁶³⁵Cf. Cook (1982), Cook et al. (1983, pp. 275–279), Cook and Whitmeyer (1992) or Weyer (2000, pp. 14–18).

⁶³⁶Notably, this has a decisive impact on the identification of the most powerful or prominent actor in a social network structure, see e.g. Cook et al. (1983), Bonacich (1987), Yamagishi et al. (1988), Yamaguchi (1996) or Mizruchi and Potts (1998).

environmental and safety initiative of a high-technology firm. Furthermore, contributions employing citation analysis (Carter et al., 2007b; Phillips and Phillips, 1998) or author collaboration studies (Autry and Griffis, 2005) as classical applications of SNA can be mentioned, where a network structure of journals, authors or their affiliation as actors and their relationships, denoted as citations or coauthoring are surveyed in a social network context. However, considering that supply chains are in fact supply networks (Lamming et al., 2000 or Harland et al., 2001), authors like Omta et al. (2001), Lazzarini et al. (2001), Choi et al. (2001) promoted social network analysis in their rather programmatic statements to overcome an overall present myopic view in supply chain management research focusing mostly on single relationships and dyads. This resulted in a steadily increasing stream of contributions importing concepts from SNA to the supply chain management domain, as noted e.g. by Carter et al. (2007a) or Borgatti and Li (2009).

4.6.3 Concluding Comments

To conclude this section, (1) a well grounded sociological theory about personal interaction, weak and strong ties, structural holes, social capital and embeddedness issues offers opportunities to explore network structures from a sociological point of view and (2) social network analysis as well as network exchange theory allows to explore and/or quantify relationships of actors in social network structures.

During the last two decades, NES got popular across multiple research disciplines so that authors like Paldam (2000), Adler and Kwon (2002) or Batt (2008) even suggested that social capital should get a "broad umbrella concept" for all social sciences. Accordingly, Richter and Furubotn (2003, pp. 209–210), Richter (2005) or Williamson (2000) stated that NES is more a complement than a rival to NIE.⁶³⁷ In addition to this, several authors like Easton and Araujo (1994), Welch and Wilkinson (2004) or Batt (2008) saw a third wave of IMP research extending the research agenda of industrial networks towards more sociological notions. At least several examples showing how authors adopted the issues of embeddedness and social capital in order to overcome an undersocialized view of economic theory can be mentioned:

Elg and Johansson (1993) as well as Jones et al. (1997) mandated to integrate TCE and structural embeddedness issues in order to get a better understanding how industrial or network governance emerges and sustains.
Easton and Araujo (1994) and Halinen and Tornroos (1998) elaborated the ¨ concept of embeddedness in industrial network research. While Easton and

⁶³⁷Especially Williamson (2000) drew an extensive four level hierarchical structure of NIE with a social embeddedness level at the top (represented by NES), followed by an institutional environment level (represented by PRT), governance level (represented by governance approach of TCT) and the lowest level coping with resource allocation and employment (represented by PAT as well as some contributions from NCE).

Araujo (1994) focused on the structural and temporal context of exchange relationships, Halinen and Tornroos (1998) defined six types of embeddedness ¨ characteristic to business networks, namely temporal, spatial, social, market, technological and political embeddedness with the latter further developed by Hadjikhani and Hakansson (1996), Hadjikhani and Ghauri (1998) or Welch and ˚ Wilkinson (2004).

Autry and Griffis (2008) as well as Min et al. (2008) discussed the role of social capital in supply chains. Whereas Autry and Griffis (2008) introduced structural and relational embeddedness concepts to supply chain management in order to establish a theoretical framework of supply chain capital and its impact on supply chain knowledge as well as performance, Min et al. (2008) tried to explain the supply chain phenomenon through the concepts of social identity salience (the extent to which a firm senses that it belongs to a supply chain) and social capital (the sum of formal and informal ties of supply chain members).
While introducing a concept of triads as the smallest unit to study the embeddedness of relationships in a network structure, Laage-Hellman (1989), Blankenburg (1992), Pedersen (1996), Havila (1996) as well as Havila et al. (2004) sought to overcome the inherent "dyadic reductionism" in the industrial network approach.638 Albeit not directly referring to these contributions, recent works from Choi and Kim (2008), Li and Choi (2009) and Choi and Wu (2009) go into this direction, too. Recognizing that dyadic buyer–supplier relationships sometimes are influenced by relationships with other buyers or suppliers (Wu and Choi, 2005), they explored triads of buyer–supplier–supplier as well as buyer– supplier–buyer relationships applying the concepts of embeddedness, social capital and structural holes as well as balance theory to the field of supply networks.639
Finally in the case of international trade intermediaries, Ellis (2003) proposed that transaction cost reduction as well as alleviation of adverse selection and moral hazard problems cannot fully explain their market making activity. But information gaps between markets which create opportunities for them to negotiate cross-border exchange can explain their existence which are in line with the core principles of Burt (1992) structural hole theory.

4.7 Comparisons and Evaluation

After presenting six distinct modelling approaches to network structure, it is now time to compare and evaluate them, which can be done along several dimensions as shown in Table 4.1.

⁶³⁸Reviewing this stream of literature, Holmen and Pedersen (2000) even proposed the concept of "serial tetrads" in order to avoid "triadic reductionism".

⁶³⁹Balance theory itself, as developed by Heider (1946, 1958), is part of the sociometric stream of social network analysis, see Sect. 5.1.1.

Table 4.1	Models of network structure Network modeling in MS OR/	Strategic networks	Dynamic networks	Entrepreneurial networks	Industrial networks	Social networks
main Original do	Planning and operations ment manage	ment manage Strategic	ment manage Strategic	International business	Industrial marketing	sociology mic Econo
objec- Theory	network design and/or ms of Solve proble	Offering managerial solutions	Offering managerial solutions	entreprenuerial Explaning	me Broad anlytical sche for industrial	Analysis of social network
Drivers tive	mizing available certain restrictions resources under network flow Opti	In-and Outsourcing of value-added activites	whenever needed Teaming up resources	information costs and personal networks ment of mizing on activity mploy Econo e	links, resource ties m of activity and actor bonds networks A syste	Embeddedness and social capital of structures actors
Actors	Focus laid on objects of network flow, not on actors	mpany and their related partners Focal co	organizations and less independent central broker(s) Group of more or	information flows intermediating in makers Entrepreneurs as market-	Independent but closely interconnected actors	representing their mpanies Individuals co
Linkages	Transport and storage processes	ments contractual Long-term arrange	relationships market-like Short-term business	Information flows, goods flows and trustworthy reltionships	business relationships Long-lasting heavily interdependent	Personal interaction or information flows
ments Argu	min) max- mic Strictly econo reasoning (	mic Strictly econo reasoning	mic Strictly econo reasoning	mic as well as social aspects considered Econo	mic as well as social aspects considered Econo	Mainly social considered aspects
Network gover- nance	Considered as being ex ante given, not of special interest	structure around a ordered network central actor Hierarchically	virtual corporation Market-like network organization or	levels of sub-networks structured in several Hierarchically	business relationships Markets as networks of highly intedependent	Undetermined
Analytical instru- ments	programming and graph theory matical Mathe	mnly verbal reasoning Sole	mnly verbal reasoning Sole	matic diagrams Verbal reasoning supported by sche	matic diagrams Verbal reasoning and sche	expressions and graph theory matical Mathe

The foremost objective of the OR/MS approach to network modelling is just to solve problems of network design and/or flow, where available resources as objects of network flow are optimized under certain constraints and restrictions. This is in sharp contrast to other approaches to network structure which capture functional as well as institutional issues to offer explanations for network formation and ongoing processes therein.

Both deeply rooted in the strategic management literature, the strategic and the dynamic network approaches have much in common: Both stick on solely verbal reasoning of strictly economic arguments to promote managerial solutions to an ever changing business environment. But the drivers of network formation are different: Whereas in case of the strategic network approach, a central actor wants to gain competitive advantage by in- or outsourcing of value-added activities on a long-term commitment basis, in the dynamic network approach at least one central broker teams up resources from a group of more or less independent organizations whenever needed which gives him a high amount of flexibility.

In the entrepreneurial network approach, entrepreneurial activity in an information-based economy is explained where network formation occurs by actors as market makers economizing on information costs and employing their personal network of group memberships in different sub-networks allowing them to establish trustworthy relationships with distant colleagues. Its verbal reasoning supported by schematic diagrams can be regarded as a further development of findings from MMT and ITF as described in Sect. 3.2 in the wider context of network structures.

Although the industrial and social network approach both share common social values with the closest ties between the industrial network approach and the network exchange theory, their approaches to network structure are very different. The industrial network approach with its system of activity links, resource ties and actor bonds offers a broad analytical scheme for networks of independent but closely interconnected actors through long-lasting, heavily interdependent business relationships whereas in social network theory, a sociological view of economic action sets the focus on actors embedded in a network structure of personal interaction or information flows which sums up to social capital on a personal as well as company level if these actors are regarded as representatives of their companies. As already shown, NET and SNA are two complementary analytical concepts: Whereas NET focuses on networks of social relationships or relational embeddedness, SNA deals more with social network structure or structural embeddedness.

Furthermore, undersocialized as well as rather oversocialized network approaches can be identified with the OR/MS approach of network modelling being clearly the most undersocialized and the social network approach as the most socialized one. Being somewhere in between, the entrepreneurial as well as the industrial network approach partly build on sociological notions, whereas the strategic and the dynamic network approach deny social contacts between network members in favour of strictly economic arguments.

Common to all network approaches as described in the last sections is that in all one or more actors in a network structure under observation are considered to play specific roles, which can easily be described in terms of being an intermediary, middlemen, broker and/or agent as already thoroughly discussed in Sect. 3.3. Their differences lie in the model assumptions made with the strategic network approach ruling out agent positions with its focus on focal companies, the dynamic network approach supposing brokerage action and the others being virtually open to all possibilities.

From an overall institutional point of view these six network approaches form a continuum of network governance from strictly hierarchically ordered to marketlike network structures. Towards the hierarchical end of this continuum, both the entrepreneurial as well as the strategic network approach can be located, whereas the dynamic and the industrial network approach propose merely markets-as-network structures with the OR/MS approach and the social network approach being in first instance undetermined and therefore open to all forms of network governance.

Last but not least, the applicability of all of these six approaches to network structures has to be assessed, as their comparison unveils considerable differences in their theoretical core. This in turn has a decisive impact on their usefulness in the given context, because different approaches to network structure support different structural properties and network governance. Therefore, an one-fits-all approach to network structure is not recommended as outlined in the following:

First, the OR/MS approach to network modelling demands well defined problems of network design and/or flow, where objects of network flow are subject to subsequent transport, handling and storage processes which can be optimized under certain restrictions. As already discussed in Sect. 4.1.3, this approach to network structure is not useful, regarding the purpose of this work.
The strategic network approach offers concrete managerial solutions to specific network structure constellations around a focal actor. Such forms of network governance are prevalent in real-world supply chains of big scale automotive or high-tech industry where an original equipment manufacturer (OEM) has to orchestrate other members of his supply chain as discussed e.g. by Dyer (1996), Lambert et al. (1998) or Pfohl and Buse (2000). In addition to this, this network approach can be employed to conduct studies of merger and acquisition behaviour or strategic alliance building e.g. in the maritime or airline industry because there, dominant carriers like Maersk, Wallenius Wilhelmsen or Lufthansa can be easily identified as focal actors in a strategic network. As the object of interest are freight forwarders being one of many players in multimodal transport chains, the strategic network approach does not seem to be appropriate in this context.
Dynamic networks, as outlined before, are clusters of small or medium sized companies forming market-like network organizations or virtual corporations for specific tasks. This specific network governance is very common in supply chains of the toy or apparel industry with low-tech value-adding activities and short production cycles as well as of quickly evolving industries like biotechnology and microelectronics (Miles and Snow, 1994 or Pfohl and Buse, 2000) where companies like the Li & Fung Group (Magretta, 1998 or Flextronics Int. (Ferry,
2004) serve as a central broker. Furthermore, Bitran et al. (2007) recently presented their "maestro model" of third-party co-ordination in supply chain governance, which can be regarded as a further development of these dynamic network considerations. In concordance with this, some authors like Rennings (1992) or Kylaheiko et al. (2000) considered that a freight forwarder can be regarded as serving such a central broker role, too. However, striking reasons against such a view are given in Sect. 4.3.2, which prevent the dynamic approach to be applicable in this work.
The entrepreneurial network approach itself focuses on intermediary aspects in an information-based economic environment, which is often a topic in cross-border transport chains e.g. organized by freight forwarding companies or intermediaries of the international trade sector with far reaching business contacts supporting transport, logistics as well as supply chain management. In Sect. 4.4.2, a short example was already outlined, where the freight forwarder serves as a transport middleman in multimodal transport chains due to crossborder logistical processes and a higher-level entrepreneurial network with high-trust relationships to colleagues abroad. But the overall descriptive nature of the entrepreneurial approach prevents to draw further conclusions beyond steadystate observations.
The industrial network approach seems to be of special interest, if interdependencies between resources and/or actors are high and there is a market environment present. These conditions can be found in the land-based groupage service sector, third-party logistics provision or other not too much concentrated industry clusters with the first well reflected e.g. in Hertz (1993) or Ludvigsen (2000) and the second e.g. in Skjøtt-Larsen (2000) or Halldorsson et al. (2007). ´ Referring to freight forwarding in multimodal transport chains, it might be questionable whether interdependencies of valuable resources are ever present. But principally, this approach to network structure seems to be applicable if the focus is laid on social context and long-term business relationships.
However, the social network approach is highly relevant in all kind of transport, logistics or supply chain management studies, if there is a need (1) to model a full set of actors and their linkages in a network, (2) to develop purely social aspects and/or information flows in network structures beyond strictly economic considerations and (3) to explore and/or quantify them directly. No further presumptions are laid on network governance.

Finally, analytical instruments offered by these network approaches as shown in Table 4.1 can serve as a guide for the appropriate research methodology. The social network approach allows operationalizing network structure directly but comes along with a lot of mathematical calculus. It is common to the strategic, dynamic, entrepreneurial and industrial network approach that they all show an inherent descriptive nature to assess institutions and their relationships which demands a case study methodology where network structures are initially described and then compared in order to discover common patterns and to end in more or less general conclusions.

To conclude this section, the social network approach with SNA as a measurement concept seems to show the highest usability in the context of this work, as it (1) allows to model a full set of possible actor roles as well as network governance without bias, (2) develops purely social aspects and/or information flows in network structures beyond strictly economic considerations and (3) allows to explore and/or quantify them directly.

| • | | | |---|--|--| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |

Chapter 5 A Closer Look on Contemporary Social Network Analysis

In the following, roots, levels of analysis and relevant measurement issues of SNA are discussed. The main purpose of these sections is to show (1) which are the main streams of development towards contemporary SNA, (2) which are possible levels of analysis and (3) which are relevant analytical instruments to capture social network structures as well as social capital issues.

5.1 Roots of Contemporary Social Network Analysis

As already indicated above, Georg Simmel's formal sociology and his studies of relations between actors in triads and dyads in Simmel (1908) are somehow the beginning of social network analysis. These initial efforts were further developed by two groups of social scientists, namely social psychologists from Austria and Germany deeply rooted in Gestalt theory of Kohler (1921) postulating that a total view ¨ on the "Gestalt" gives every element a meaning and US-American as well as British social anthropologists heavily influenced by the structural-functional anthropology of Radcliffe-Brown (1940, 1957).1 Following Schenk (1984, pp. 1–29, 138–162), Mitchell (1979), Scott (1991, 7–38), Hummon and Carley (1993), Galaskiewicz and Wasserman (1993, pp. 6–12), Wasserman and Faust (1994, pp. 10–18), Jansen (2003, pp. 37–49) and Freeman (2004), contemporary social network analysis can be tracked along three main lines of more or less convergent developments: (1) sociometric analysis and graph theory, (2) US-American community and industrial sociology and (3) the Manchester School of social anthropology. In addition to this, methodological breakthroughs in the mid 1970s like blockmodelling analysis

¹According to Schenk (1984, pp. 2–3) or Freeman (2004, pp. 100–103), Alfred Reginald Radcliffe-Brown was the first introducing the term "social structure" as a metaphor for networks of actually existing relations.

and models of social typology led to contemporary social network analysis as shown in Fig. 5.1.

5.1.1 Sociometric Analysis and Graph Theory

The roots of this first stream of development can be dated back to the 1920s and 1930s, when a group of Austrian and German researchers working on the topics in cognitive and social psychology heavily influenced by the Gestalt theory of Kohler ¨ (1921) emigrated to the United States. These emigrated researchers, namely Jakob Moreno, Kurt Lewin and Fritz Heider, stressed the role played by the structure of interpersonal relations in the construction of cognitive schemes, and their studies dealt with information flows and the information of shared ideas.

Being the founder of *sociometry*2, Moreno (1934, 1953) pioneered sociometric analysis techniques by constructing sociograms as pictorial representations in order to map social linkage patterns. These sociograms are two-dimensional diagrams consisting of a set of points representing actors and a set of lines which stand for linkages connecting actors to each other. When a pair of actors is linked to each other, a line is drawn between them. Recognition, that sociograms could be employed in studies of social structures led to a rapid development of more analytical techniques.

Lewin (1936) early work in what he called "topological psychology" outlined his view that group behaviour has to be regarded as being determined by a "field" of social forces, a social space in which the group and its surrounding environment is located. Furthermore, he argued that the environment of social groups matters to the extent it is perceived by their members. According to Lewin (1951), structural properties of this social space can be analyzed by employing mathematical techniques of topology and set theory. The aim of his field theory is to determine the interdependence between group and environment in a system of relations. But this theoretical framework was not developed further in the field of sociometric analysis, although the late work of Lewin (1951) published after he died suddenly and unexpectedly in 1947 provided a basis of mathematical modelling of group relations.³

Another contribution to social psychology was the work on group dynamics by Heider (1946, 1958). His initial work was on cognitive psychology of attitudes and

²Following Moreno (1953, pp. 15–16), "[s]ociometry is the mathematical study of psychological properties of populations, the experimental technique of and the results obtained by application of quantitative methods." Further, to him "[s]ociometry aspires to be a science within his own right" (Moreno (1953, p. 48)). The term itself can be regarded as a combination of "socius" (Lat.), meaning social or companion and "metrum" (Gr.) or "metron" (Lat.) standing for measurement.

³Main reason to this may be, that "field theory is in place indeterminate and resists formalization" (Martin (2003, pp. 42–43)) and therefore no progress into this direction occurred. Martin (2003) itself provides a good review of further development and discussion of contemporary field theory.

Fig. 5.1 Roots of social network analysis

perceptions and he was especially interested in the cognitive balance of a person's attitudes towards others. This concept of balance theory was further developed by Newcomb (1961) extending it to an objective approach of structural balance in a group. With his work, a shift occurred from a concept of cognitive balance of individuals to that of interpersonal balance in groups so that it was from now on possible to build models of systematic interdependence between the attitudes held by different persons within a group.

Furthermore in the 1950s and 1960s, Dorwin Cartwright, Frank Harary and Claude Flament were among the first to quantify these structural balance propositions in group dynamics mathematically by employing mathematical instruments from graph theory. In addition to this, the attempt to apply mathematics in order to model the structure of groups was of course no new idea (see e.g. Zeleny, 1940a,b, 1941, Bavelas, 1948, 1950, Seeley, 1949 or Katz, 1953), but contributions by Cartwright and Harary (1956), Flament (1963) and Harary et al. (1965) provided a path-breaking methodological framework for powerful models to show cohesion, social pressure, cooperation, power and/or leadership in social groups.

In the following, researchers like Rapoport (1963) and Coleman et al. (1966) developed contagion models in order to study diffusion of diseases, information and inventions in networks with heavy usage of statistical methods because these traditional methods of sociometric analysis were not found quite suitable in the analysis of large populations.

Another stream of research in the tradition of balance theory is triad census analysis. In this context, authors like James A. Davis, Paul W. Holland and Samuel Leinhardt made basic contributions writing a series of papers in the 1970s, which are worth to mention. Davis (1967) e.g. discussed which types of triads including three actors and all observed relations between them should and should not arise in empirical research. Based on this work, Holland and Leinhardt (1970, 1971, 1972) and Davis and Leinhardt (1972) refined in a series of papers their concept of triad census analysis by introducing a wide variety of randomly directed graphs distributions in order to test their hypotheses about various structural tendencies.

5.1.2 US-American Community and Industrial Sociology

A second stream of development originates in the community and industrial sociology tradition of US-American researchers around the anthropologist William Lloyd Warner and the psychologist George Elton Mayo at Harvard University who conducted a series of important anthropological studies during the 1920s and 1930s to collect ethnographic data in several communities as well as industrial companies. The aim of this first Harvard structuralist group was to uncover social structures of interacting individuals and groups of actors.

One of the most famous outcomes of this extensive research programme were the Hawthorne studies(reported e.g. by Roethlisberger and Dickson, 1939), which mark a starting point for the Human Relations Movement in industrial sociology.4 Among other topics not related to this context, it was central to this study to investigate the relationship between informal group structure and work performance.

Combining Moreno (1934) sociometric analysis techniques with matrix formulation, Homans (1950, 1961) re-examined the data of Roethlisberger and Dickson (1939) later on, but finally he failed to get further insights mainly due to the sheer complexity of this task and a lack of computational capacity at that time.

5.1.3 Manchester School of Social Anthropology

The third stream of development as indicated above is closely related to the Manchester School of social anthropology with classical field studies e.g. by British social anthropologists like Max Gluckman, John Barnes, Elizabeth Bott and Clyde Mitchell.5 This group of researchers was even more influenced by the structural-functional anthropology of Radcliffe-Brown (1940, 1957) than their US-American colleague at Harvard University. For them, conflict and power were integral elements of any social structure and therefore, in contrast to a traditional emphasis on social integration and cohesion, they developed network analysis concepts to study conflict and power relations in social groups.

Max Gluckman was the central figure at the Manchester School who combined his interest in complex African societies with the development of a structural approach recognizing the important part played by conflict and power in both the maintenance and the transformation of social structures (see e.g. Gluckman, 1955). Furthermore, Barnes (1954) reported a field study of social behaviour in a Norwegian fishing village conducted by him. This field study seems to be one of the first contributions to social network analysis, where the term "social network" was explicitly defined. Another well cited early field study about kinship relations and social environment of families in London was conducted by Bott (1957) who is often regarded to be the first who developed measurement concepts for personal, egocentered networks. All these three classical field studies from African communities to next door neighbourhood relations in European cities offered examples of a viable approach to the discovery of informal relations between members of social groups. In addition to this, measurement concepts like density, connectedness, reachability, as well as parameters related to the intensity and strength of ties were first developed by them.

⁴See Rogers and Agarwala-Rogers (1976, pp. 36–48) or Freeman (2004, pp. 43–64) for a historical overview.

⁵Notably, Max Gluckman was the only resident in Manchester with the other three being regular visitors working at London School of Economics, see Freeman (2004, pp. 103–105). Moreover, anthropologists from other countries did also field studies of this style, see Mitchell (1974, 1979) or Wolfe (1978) for a more broad review.

Meanwhile, Siegfried Nadel, an Austrian psychologist, got in contact with them and attempted to join their findings together with the other two streams of development to form a theoretical basis for a structural sociology backed by formal network analysis. To him, social structure was an overall system network or pattern of relations, which can be analyzed without observing single actions of individuals involved therein. Moreover, he asserted that social structures are structures of role occupants together with their corresponding role-sets which should be the object of further analysis employing matrix algebra. Unfortunately, his sudden death in 1956 prevented him from developing further his ideas he had signposted in Nadel (1957).

In the following, Mitchell (1969) and Barnes (1972) tried to realize this research agenda of Nadel (1957) by developing a full bag of graph theory related measurement instruments, but their exclusive concern on informational relations mainly restricted to personal, ego-centered networks and its merely descriptive approach have contributed to a decline of this stream of development during the 1970s.

5.1.4 Methodological Breakthroughs

By this time, a first breakthrough in analytical development of social network analysis was made by a second Harvard structuralists group around Harrison C. White. They combined graph theoretical methods with algebraic models of group theory and founded blockmodel analysis as a way to formulate explicit models of social structure driven by matrix algebra. Doing this, they developed an algorithm which allows to draw conclusions about position and role structures in social networks having relational data on the individual level and this even in large networks of social structure. Examples of early works in blockmodelling are Lorrain and White (1971), White et al. (1976) or Boorman and White (1976).

Lorrain and White (1971) did foundational work by expressing the theoretical concept of social role as a formal mathematical procedure called structural equivalence. For them, individuals who have identical relationships to and from all others in a network are structural equivalent. White et al. (1976) and Boorman and White (1976) extended this to a self-consistent search procedure in order to partition networks into sets of actors called blocks being structural equivalent with respect to other such sets across several types of relationships.

The second breakthrough in analytical development was in a steady stream of contributions by Ronald S. Burt summing up to the coherent framework of social typology in order to describe a system of actors forming a network. In Burt (1976, 1977a), he determined network positions of actors in terms of social distance and employed cluster analysis and structural equation modelling to work out structural equivalent positions. In the following, he specified e.g. three general conceptualizations of actor's power within a network in Burt (1977b) and defined a concept of structural autonomy in Burt (1980b) only to name two other early contributions of his extensive research agenda.

Now, social network analysis can be regarded as an established stream of research dominated by US-American researchers.⁶ Widespread application of social network analysis in form of empirical studies as well as theoretical treatments are occurring in the fields of anthropology, sociology, politics, geography, communication and organization studies as cited e.g. in Hummon and Carley (1993), Jansen (2003, pp. 48–49) or Freeman (2004).

5.2 Levels of Analysis in Contemporary Social Network Analysis

According to network data, Knoke and Kuklinski (1982, p. 16) distinguished between an attributional and a relational level of analysis.7 Attributes are considered to be intrinsic characteristics of actors like people, objects, or events. These persons, objects or events may at the same time be involved in some relationships, but these relationships are not an intrinsic characteristic of either party involved in isolation, nor an emergent property of the connection or linkage between units of observation. Where attributes of actors may persist over a long time, relations are context specific and alter or disappear upon an actor's removal from a network structure.

In addition to this, Jansen (2003, pp. 60–67) identified five levels according to the unit of analysis: (1) dyads, (2) triads, (3) actors and their personal networks, (4) subgroups and (5) complete networks.

For her, a dyad is the smallest unit of analysis in a network structure and the dyad itself can be regarded as a network consisting of two points of actors and their relation to each other. The next higher level of analysis is a network structure with three points or actors called triad and their relations to each other. Actors and their personal networks are a distinct type of networks called ego-centric networks, which are according to Shulman (1976, p. 310)

"[...] the set of persons connected to a particular individual (ego) who represents the focal point or anchorage of the network. The total [personal] network is not focused in an individual's connections but on all existing connections [sic] among all persons in some defined set or domain."

The fourth level is a subgroup of actors in a social network structure formed by several dyads, triads or even higher combinations of actors with special relations to each other. Finally the focus of analysis can be the complete network (sometimes called socio-centric network), including all actors and all relations in a network structure.

Burt (1980a, p. 80) provided a more structured view of measurement issues in network structures as shown in Table 5.1. According to him, social network

⁶Cf. Hummon and Carley (1993). Notably, only a few contributions about methodological issues in social network analysis are being published in German, e.g. Kahler (1975), Schenk (1984), Pappi ¨ (1987) and Jansen (2003).

⁷See Jansen (2003, pp. 53–58) with a similar definition.

	Actor aggregation in a unit of analysis
Analytical approaches	Actor (micro perspective)	Multiple actors as a network subgroup (meso perspective)	Multiple actors and/or subgroups as a structured system (macro perspective)
Relational approach	Personal network as extensive, dense and/or multiplex	Primary group as a network clique: a set of actors connected by cohesive relations	System structure as dense and/or transitive
Positional approach	Occupant of a network position as central and/or prestigious	Status or role-sets as a network position: a set of structurally equivalent actors	System structure as a stratification of status/role-sets

Table 5.1 Concepts of social network structure and level of analysis (Source: Burt (1980a, p. 80) with some extensions)

structures can be analyzed at three levels of aggregation. At the highest level (macro perspective), all actors in a social network structure and their relations to each other are treated as a single unit of analysis. At the lowest level, each actor in a social network structure and his relations to other actors can be subject to investigation (micro perspective). Between these extremes, actors in a social network structure can be aggregated to a network subgroup or clique, so that these subgroups or cliques within are compared as units of analysis (meso perspective).

In addition to this, Burt (1980a, p. 80) distinguished between two fundamentally different analytical approaches within these three levels of aggregation, namely the relational and the positional approach. In the relational approach, the focus is set on relationship patterns between pairs of actors and these are treated in isolation to others in the whole network structure. In the positional approach, an actor is regarded as one of many actors in a network structure and patterns of all relevant relationships are used to define his network position. Both the relational and the positional approaches describe relationships in network structures, but they do it from very different intellectual traditions. On one hand, the relational approach develops a social-psychological concept of differentiation. There, network structure is primarily described in terms of typical relationships in which individual actors are involved. On the other hand, the positional approach stands for more sociological and anthropological concepts of differentiation. Network structure is then regarded as consisting of interlocked, differentially prestigious status or role-set in which actors are stratified.

5.3 Analytical Instruments of Contemporary Social Network Analysis

In the following, some important analytical instruments of SNA with a focus on the above mentioned micro and macro perspective being relevant in the context of this

	Actor aggregation in a unit of analysis
Analytical approaches	Actor (micro perspective)	Multiple actors as a network subgroup (meso perspective)	Multiple actors and/or subgroups as a structured system (macro perspective)
Relational approach	Degree, multiplexity, heterogeneity brokerage and/or constraint	Clique analysis, triad zensus, brokerage	Size, density, cohesion, multiplexitiy, heterogeneity, brokerage
Positional approach	Centrality and prestige	Structural equivalence, blockmodelling and cluster analysis	Centralization and hierarchization

Table 5.2 Analytical instruments for social network structure per level of analysis

work are discussed (see Table 5.2 for a general overview).8 First, basic measures on the individual, relational and network level are described before more complex measurement concepts of the positional approach (centrality, centralization, prestige and hierarchization) and the relational approach (network constraint and brokerage capacity) are presented.

5.3.1 Actor's Degree, Density, Multiplexity, Heterogeneity and Connectedness

As already noted above, basic elements of all kinds of networks are a set of points or actors and a set of relations (or synonymously edges, links, or ties). In order to analyse social networks structures, these two basic elements have to be measured in some way.

The first and most natural measure to characterize a point or an actor in a social network structure is, apart from attributional or sociodemographic data, the amount of links or relations he or she has.⁹ Or in other words, it is simply the number of points or actors to which a given point or actor is adjacent to. This measure is usually called the actor's degree and can be expressed mathematically in the following way:

$$d(pi) = \sum{j=1|j\neq i}^{N} a_{ij} \ \forall \ i$$ (5.1)

⁸Of course, there are more than the following measurement concepts existing as tools for social network analysis like triad analysis, structural equivalence, blockmodelling but these measurement concepts are (1) not suitable in the research context and (2) therefore beyond the scope of this piece of work. For further reference of these measurement concepts see e.g. Burt (1980a, pp. 96–130), Burt (1982, pp. 37–87), Knoke and Kuklinski (1982, pp. 56–86), Schenk (1984, pp. 83–108), Scott (1991, pp. 103–148), Wasserman and Faust (1994, pp. 227–721), Jansen (2003, pp. 163–220) or Knoke and Yang (2008, pp. 72–117).

⁹Cf. Burt (1983, p. 177), Scott (1991, pp. 71–72) or Jansen (2003, pp. 103–104).

In addition to this, for directed networks an actor's outdegree d0.pi/ and an actor's indegree d1.pi / can be calculated accordingly by

$$d_0(pi) = \sum{j=1|j\neq i}^{N} a_{ij} \ \forall \ i$$ (5.2)

and

$$d_1(pi) = \sum{j=1|j\neq i}^{N} a_{ji} \ \forall \ i$$ (5.3)

where d1.pi / (d0.pi/) counts directed relations to (from) the point or actor pi from (to) other points or actors adjacent in a social network structure. Unfortunately, d.pi/, d1.pi/ and d0.pi/ are biased in N, but all three measures can be standardized by division with .N -1/ if ai i ^D ⁰ for all ⁱ ^D 1; : : : ; N holds.¹⁰ The last expression means, that reflexive relations (a relation from an actor to himself or herself) are not allowed by definition.

Furthermore, an actor's density is the number of actual existing relations divided by the number of possible relations or in case of a directed graph

$$AD(pi) = \frac{\sum{i=1}^{N} \sum{j=1|j\neq i}^{N} a{ij}}{N(N-1)} \,\,\forall \,\, i$$ (5.4)

with N.N - 1/ the maximum number or relations for a single point or actor in a network with N points or actors.¹¹ For some authors like Weimann (1989, p. 190) AD.pi/ stands for the degree of integration in a personal network.

Multiplexity is a special feature of social network structures, if there exist more than one distinct kind of relations between a pair of points or actors.¹² The network structure can then be regarded as a ^M-dimensional construct so that each *^z*ij .m/ ^D z.pi ; pj ; m/ is defined as the strength of the mth link between pi and pj . Based on this definition, a measure of multiplexity of actor's contacts can be calculated by

$$AM(pi(m)) = \frac{\sum{j=1|j\neq i}^{N} z_{ij}(m)}{(N-1)} \,\,\forall \,\, i$$ (5.5)

with *^z*ij ^D ⁰ if pi has only one kind of relation to pj and ¹ else.

¹⁰Cf. Jansen (2003, p. 104).

¹¹This relationship was first formulated by Kephart (1950). The same formulae can also be used for undirected graphs if all mutual relations are counted twice.

¹²Cf. Barnes (1979, pp. 411–412), Burt (1980a, p. 90), Burt (1982, p. 32) or Diaz-Bone (1997, p. 59).

Another special feature in social networks structures is actor's heterogeneity, where actors with different attributes form subgroups as a part of the whole network structure. A measurement of heterogeneity has to capture that an actor may have contacts especially to other actors in different subgroups. In order to operationalize this, actors with the same attribute(s) can be grouped together to form ^o distinct subgroups ^o D 1; : : : ; O. Then a ^N-dimensional construct with elements *^z*ij .o/ ^D ^z.pi ; pj ; o/ with ⁱ ¤ ^j and i;j ^D 1; : : : ; N and *^z*ij .o/ ^D ¹ if pi and pj belongs to the same distinct subgroup o and 0 else can map adequately memberships in different subgroups existing in a social network structure.¹³ Burt (1983, p. 178) proposed in the case of ego-centric networks to count the number of different subgroups of alters directly connected to ego's subgroup to get a measure of heterogeneity of actor's contacts¹⁴

$$AH(pi(o)) = \sum{j=1|j\neq i}^{N} z_{ij}(o) \ \forall \ i.$$ (5.6)

Last but not least there is a concept of connectedness, where effects of changes in the elements of a network are employed as a basis to describe the importance of a point or an actor in a given network structure.15 If the removal of a point or an actor including the relations associated with him or her results in a disconnected graph, this point or actor represents a cut point in a social network structure. The importance of these cut points or actors originates from their unique structural position and their ability to bring members of different subgroups together that are otherwise unrelated to each other. But this is considered to be closely related to the above already discussed measurement concepts of degree and density, so that no special measures of connectedness are proposed in social network analysis literature. The higher the average degree and density of actors in a social network structure, the more connected is a graph and therefore the less likely is the result of a disconnected graph, when a point or actor is removed.

5.3.2 Tie Strength, Durability, Directedness, Multiplexity and Connectedness

After discussing basic measurement concepts applicable to points or actors in a social network structure, now the focus is set on ways to characterize the relations between them. At the same time, they can be used to describe dyads accordingly,

¹³Cf. Wasserman and Faust (1994, pp. 145–117), Diaz-Bone (1997, p. 46).

¹⁴Some authors called this measure homogeneity or individual diversity, see e.g. Weimann (1989, p. 190).

¹⁵Cf. Knoke and Kuklinski (1982, p. 41), or Jansen (2003, pp. 97–98).

because they are defined as a set of relations between a pair of points or actors.16 As already defined in Sect. 4, every link between two points or actors has at least two basic attributes: tie strength and direction. Furthermore, duration of a relation can be assessed and a set of relations can be described along other dimensions like reciprocity and/or multiplexity.¹⁷

First, there are numerous ways to measure tie strength. In a lot of studies of social network structure, ties are just regarded to exist or not to exist, and then tie strength is a dichotomous parameter with *^z*ij ² ^f0^I ¹^g for all ⁱ ¤ ^j . In other cases, ties might have a different strength or magnitude (so-called valued relations) and/or they are quite opposite relations, indicated by a positive or negative sign. Quite often, tie strength is considered as a multidimensional construct, employing a set of measures like amount of time spend, frequency of contact, emotional intensity, intimacy and/or contractual basis only to mention a few possible dimensions.18 Employing these indicators, most studies of network analysis deal implicitly with strong ties, which is to some extent in contradiction to Granovetter (1973) "strength of weak ties," where weak ties are considered to be important due to their bridgefunction connecting members of different subgroups.

Another measure closely related to tie strength is the durability of a relation. Principally, there are two approaches to measure durability: (1) Each actor can be asked how long each particular relation exists if there is any or (2) he or she can be asked after some time again if a particular relation still exists in order to get longitudinal data about relations. But both approaches of measurement are quite problematic (1) with taking only a snapshot at a certain point of time and (2) only being applicable, if there is a perpetual access given to an actor's group in a network survey.¹⁹

Following graph theoretical definitions already presented in Sect. 4, three different types of linkages in terms of directedness between two points or actors can occur, namely asymmetric, mutual, and null ties. These three possible states lead to different types of graphs called directed or undirected graphs visualizing distinct forms of social network structures. Looking closer at measures of directedness on the link level, relations in undirected graphs are always considered to be either mutual or null and represent kinship, contractual relations, exchange of property right or resources, whereas relations in directed graphs can be subject to all different types of linkage and are often employed to show relations of asymmetric power distribution and/or differences in actor's status or prestige.

In order to get a more differentiated view of directedness, one can allow multiplexity of relations so that not only one single relation can exist between a

¹⁶Cf. Lincoln (1982).

¹⁷Cf. Lincoln (1982, pp. 4–6), Schenk (1984, pp. 65–74), Weimann (1989, pp. 190–192).

¹⁸See Mitchell (1969, pp. 20–27), Kahler (1975), Marsden and Campbell (1984, pp. 483–484) or ¨ Weimann (1989, p. 191) with further reference.

¹⁹Cf. Shulman (1976).

pair of points or actors. Strictly speaking, multiplexity occurs on a link level if a relationship between two points or actors (1) is supported by more than one undirected or directed relation with the same direction, (2) consists at least of more than one mutual relation or (3) there are several paths connecting them, including some intermediary points or actors in between. In addition to this, relations can be judged in terms of reciprocity as a more refined concept of directedness applicable to directed graphs, whereas in undirected graphs it has no meaning.²⁰ Reciprocity occurs if a relation in one direction is allowed to be reflected by another relation of the opposite direction. This reflection of relations, of course, can be instantaneous or occur with some time lag.

The concept of connectedness, as already mentioned before, builds on changes in the elements of a network and can be used again as a basis to describe the importance of a link in a given social network structure.²¹ If a graph becomes disconnected by removing a line as a single relation between two points or actors, this line represents a bridge between them if they are members of different subgroups otherwise unconnected to each other. In the case of directed graphs, connectedness can be further defined as a relation's attribute indicating how two points or actors are linked together by path consisting of directed lines. Then two points or actors are:

Null-connected, if there are no directed lines joining them in either direction.
One-connected if they are joined by relations disregarding their direction.
Two-connected, if they are joined by a path in one direction but not in another.
Three-connected, if they are joined by paths in both directions.

To conclude this variety of measurement issues on a link level, some comparisons have to be made in order to show that they are closely related to each other. Shulman (1976) asserts that networks in which most actors are directly linked to each other, offer a greater chance of developing shared views. Then durability of relations cause a higher density by providing greater opportunity for actors to meet each other, and members of high density networks are more likely to establish a shared view through mutual interaction with reciprocated relations. After re-analyzing three classical social network studies, Marsden and Campbell (1984) come to the conclusion that there must be a distinction made between indicators and predictors of tie strength. For them, indicators are actual components of tie strength, whereas predictors are aspects of actual relationships closely related to them.22 Furthermore, they found in their data that emotional intensity of a relationship seems to be the best indicator for tie strength, because other measures like duration, frequency of contact or time spend were too much correlated with other measures. Some authors like Weimann (1989, p. 191) or Scott (1991, p. 68) see multiplexity as an alternative measure of intensity or strength of relations.

²⁰Cf. Lincoln (1982, p. 5).

²¹Cf. Knoke and Kuklinski (1982, p. 41) or Jansen (2003, pp. 97–98).

²²Cf. Marsden and Campbell (1984, pp. 485–488).

5.3.3 Network Size, Density, Cohesion, Multiplexity and Connectedness

A basic indicator of a complete network structure is network size measured either in terms of points or actors or in terms of relations between them. So network size can be either measured simply by counting all points or actors being connected to each other in a social network structure or it can be calculated by summing up the number of all links between all pairs of points .pi ; pj / regardless of their strength or direction so that

$$NS = \sum{i=1}^{N} \sum{j=1}^{N} \sum{j\neq i}^{N} z{ij} \text{ with } z_{ij} \in {0, 1}$$ (5.7)

and *^z*ij ^D ¹ if there is a link between pi and pj and ⁰ otherwise. In the case of ego-centric networks, another measure of network size is proposed by Burt (1983): to him, it is just the number of alters named by ego because "[t]he number of actors directly connected to an individual is an index of the extent to which the individual is involved in many different relationships" (Burt (1983, p. 177)) so that

$$NS{ego} = \sum{j=1|j\neq i}^{N} a_{ij} = d(p_i) \ \forall \ i.$$ (5.8)

Second, network density is a good measure at the complete network level indicating a general level of linkage among the points forming a social network structure. It seems to be one of the most widely used concept of measurement in social networks.23 A maximum network density is reached in a complete graph where all points or actors are adjacent to each other. In case of a directed graph, with ^N points the maximum number of edges is then N.N - 1/ and for an undirected graph N.N - 1/=2, respectively.²⁴ Therefore, a measure for network density e.g. in a directed binary graph is

$$ND = \frac{\sum{i=1}^{N} \sum{j=1}^{N} |j \neq i| z_{ij}}{N^2 - N}$$ (5.9)

and ND valued in a range between 0 and 1. ²⁵ Again, in ego-centric networks, ego as the focal actor and his or her direct links should not be necessarily taken into

²³Cf. Niemeijer (1973), Mitchell (1974, p. 288), Shulman (1976), Barnes (1979, pp. 406-408), Scott (1991, pp. 72–84), Wasserman and Faust (1994, p. 181) or Jansen (2003, p. 108).

²⁴See Kephart (1950) for a formal derivation of these figures.

²⁵ND is, virtually by definition, a relative measure of network density. An absolute measure of network density, of course, was developed by Scott (1991, p. 98), but discussing this measure in

account, because these relations exist almost by definition.26 An alternative measure for network density in a directed ego-centric network is therefore excluding ego and calculating the alters' network density or

$$ND{alter} = \frac{\sum{i=2}^{N} \sum{j=2|j\neq i}^{N} z{ij}}{N^2 - 3N + 2}$$ (5.10)

and i=j ¤ ¹ if ego is indexed by ⁱ D ¹. ²⁷ A quite fundamental problem common to both network density measures is that they are biased in N due to the fact that with an increasing number of points the maximum number of possible relations between them increases exponentially.²⁸ But considering that each actor in a network structure can only handle a small portion of all possible relations, a standardized measure of network density based on a mean actor's degree can be calculated by

$$ND' = \frac{N\frac{1}{N}\sum_{i=1}^{N}d(p_i)}{N^2 - N}$$ (5.11)

in the case of an undirected binary graph.²⁹

Another measure is network cohesion, which expresses the number of mutual links in a directed network structure:

$$NC = \frac{2\sum{i=1}^{N} \sum{j=1|j\neq i}^{N} (z{ij} + z{ji})}{N^2 - N}$$ (5.12)

with iC *^z*j i/ ^D ¹ if relations in both directions are existing and ⁰ otherwise.³⁰ Values of NC range between 0 an 1 with larger values for networks with a greater proportion of reciprocated relations. Usually in empirical studies of network structure, there seems to be not much difference between network density and cohesion.31

Looking at the overlap of several kinds of relations on the network-level, a network is considered to be multiplex to the extent that there are multiple types

depth is beyond the scope of this work. In case of an undirected graph, the same formulae can be employed with mutual relations counted twice.

²⁶Cf. Burt (1980a), Scott (1991, p. 75), Jansen (2003, p. 108) or McCarty and Wutich (2005).

²⁷See e.g. Jansen (2003, p. 109). Again, this formulae can be applied to an undirected graph with mutual relations counted twice.

²⁸This is a well known attribute common to all network structures, usually called the "network effect."

²⁹Cf. Scott (1991, pp. 77-78).

³⁰Cf. Knoke and Kuklinski (1982, p. 50), Witt (1996, p. 35) or Jansen (2003, p. 111).

³¹Cf. Knoke and Kuklinski (1982, p. 50) or Witt (1996, p. 35).

of relations between points or actors.³² This network multiplexity can be expressed by the amount of multiplex relations divided by the amount of possible relations or

$$NM = \frac{\sum{i=1}^{N} \sum{j=1}^{N} |j \neq i| z_{ij}}{N^2 - N}$$ (5.13)

with *^z*ij ^D ⁰ if pi has only one kind of relationship to pj and ¹ else.³³ Furthermore, under the notion that a network with multiple types of relations is more dense or cohesive, NM can be at the same time regarded as an alternative measure of density or internal cohesion.

The concept of connectedness on the actors level can be expanded on the network level, too.³⁴ There, a graph as a visual representation of a given social network structure is:³⁵

Strongly connected if every pair of points or actors is three-connected.
Unilaterally connected, if every pair of points or actors is two-connected.
Weakly connected, if all pairs of points or actors are one-connected.
Disconnected, if at least one point or actor is null-connected towards the others.

Having a measure of heterogeneity of actor's contacts, a cautious reader might miss now some measure of heterogeneity on the network level. Indeed, in the literature of SNA, there exists no specific measure of heterogeneity on the network level. But the measurement concept of prominence, especially the measures of hierarchization which will be discussed in the following section, capture the fact of heterogeneity.

5.3.4 Prominence: Centrality, Centralization, Prestige and Hierarchization

According to Wasserman and Faust (1994, p. 169), "one of the primary uses of graph theory in SNA is the identification of the 'most important' actors in a social network." In order to describe the location or position of an actor or highlight differences between more or less important actors in a social network structure, two distinct concepts of importance, or synonymously, prominence, can be employed: centrality and prestige.³⁶ In both concepts, actors being in a relative central location

³²Cf. Burt (1980a).

³³Cf. Knoke and Kuklinski (1982, p. 51) or Jansen (2003, p. 111).

³⁴Cf. Knoke and Kuklinski (1982, p. 41) or Jansen (2003, pp. 97–98).

³⁵According to Barnes (1969, p. 239), these definitions are originally made by Harary et al. (1965), but several other classifications of connectedness on the network level are existing, too.

³⁶See Knoke and Burt (1983, pp. 198–200), Wasserman and Faust (1994, pp. 169–178) or Kosch¨utzki et al. (2005) for a more thoroughly discussion of these two concepts.

or represent a focal point within a social network structure are considered usually to be the most important or the most prominent.³⁷

Measurement concepts of centrality stick to the fact, that an important actor in a social network structure has a lot of ties to other actors and therefore he is simply more involved in comparison to more "dispersed" located actors with less ties and therefore less involvement. They are originally designed for nondirected binary graphs, so that in this section the overall focus is laid on them with additions to refinements for directed and/or valued graphs, if necessary.

The idea of centrality in social network structure was first introduced by Bavelas (1948), studying communication in small groups and hypothesizing a relationship between structural centrality and influence in group processes. In the following, many other authors developed three distinct centrality concepts based on degree (e.g. Shaw, 1954, Rogers, 1974 and Nieminen, 1974), betweenness (e.g. Shaw, 1954 and Freeman, 1977) and closeness (e.g. Bavelas, 1950; Beauchamp, 1965; Sabidussi, 1966 and Rogers, 1974). Freeman (1979) summarized these early developments and offered a whole family of measures for each of these three concepts, one absolute and one relative measure of the centrality positions in a network called point or actor centrality and one reflecting the degree of overall centralization of an entire network called graph centrality. In addition to this, Bonacich (1972a,b) developed a rank- or status-based measure of centrality and generalized it to a whole family of local and global measures in Bonacich (1987). All four common forms of centrality measures will be discussed in the next section.

In order to quantify prestige, usually digraphs are employed to distinguish choices sent to from choices received by actors. The most prestigious actors are then those who receive the highest amount of choices from others. Knoke and Burt (1983, p. 199) commented this by:

"The fact of being involved in a relation is less important than the distinction between being the source versus the object of the relation. Control over valued resources, and the authority and deference produced by inequalities in control, is a substantive area requiring this emphasis on the asymmetry of relations, with the source of a relation being qualitatively distinct from the object of the relation in the sense of the distinction between subordination and superordination."

Therefore, actors with high prestige can be seen as leaders who are object to extensive relations from followers, while the latter are only object to few relations. Both, leaders and followers, are involved in relations, but these relations are different in terms of amount and direction which is highlighted by the following measurement concepts of prestige. According to Knoke and Burt (1983, p. 175), first studies

³⁷Scott (1991, pp. 95–96) questions this implicit assumption that a network structure has always an easily identifyable focal point being quite in the centre, and refers to Christofides (1975) concept of eccentricity, which builds on a distance matrix containing shortest path distances (or geodesics) between each pair of point. Then the eccentricity of a point is defined as the length of the longest geodesic in the distance matrix and the point with the lowest eccentricity is the absolute centre of a given graph. In some graphs, two or more points have equally low eccentricity. Then an imaginary point lying on a path between them is considered to be the absolute centre.

of prestige date back e.g. to Moreno (1934, pp. 98–103), Zeleny (1940a,b, 1941), Proctor and Loomis (1951) and Katz (1953), which used the term of status instead of prestige in their works. Three measurement concepts, namely prestige based on indegree of an actor, his or her proximity to other actors and status or rank of other actors and their extension to measures of hierarchization in a given social network structure will be discussed later on.

5.3.4.1 Measures of Centrality

Thinking of network structure looking like a star or a wheel, a central point seems to have the most central position possible. In order to measure this obviously unique position in such a network, we can assert that this central point (1) has the maximum possible degree of all points, (2) falls on the geodesics between the largest possible numbers of other points and (3) is maximally close to them because it is located at the minimum distance to all other points.³⁸

A first family of measures of point centrality can be regarded as some functions of the degree of a point. Shaw (1954) first introduced the idea of using degree as a measure of point centrality and, together with other authors like e.g. Rogers (1974) and Nieminen (1974), he regarded centrality as being the same as degree without further discussion.³⁹ Following Nieminen (1974), a degree-based absolute measure of point centrality can be calculated simply by adding up the number of adjacencies aik ^D a.pi ; pk/ with ⁱ ¤ ^k for a point pk or more formally:

$$C_D(pk) = \sum{i=1|i \neq k}^{N} a_{ik} \ \forall \ k$$ (5.14)

where aik ^D ¹ if and only if pi and pk are adjacent otherwise aik ^D ⁰ . 40 CD.pk/ is therefore large, if pk is adjacent to a larger number of other points and small if pk has only a few direct contacts. At the minimum, the point pk is totally unconnected and so minCD.pk/ D ⁰. Conversely, at the maximum all other points are adjacent to pk and so maxCB .pk/ ^D ^N -1. Unfortunately, CD.pk/ as being an absolute measure of centrality strongly depends on the total size of the network for which it is calculated and so a comparison between network structures of different size cannot be made seriously. In order to get a degree-based relative measure of

³⁸Cf. Freeman (1979), Schenk (1984, pp. 51–56), Witt (1992) or Wasserman and Faust (1994, pp. 178–192).

³⁹See Freeman (1979), Wasserman and Faust (1994, p. 178) or Witt (1996, pp. 19–28) for a more rigorous discussion.

⁴⁰It is worth mentioning, that the magnitude of CD.pk / is virtually identical to the actors degree d.pi/, because both are defined as a sum of adjancies, see Burt (1983, p. 190) for further discussion. In the case of digraphs, Wasserman and Faust (1994, p. 199) proposed to count the outdegrees of an actor.

point centrality, the effect of network size must be removed. According to Freeman (1979), it can be simply done by dividing CD.pk/ by ^N -1, so that

$$C'_D(pk) = \frac{1}{N-1} \sum{i=1 | i \neq k}^{N} a_{ik} \ \forall \ k$$ (5.15)

is the proportion of other points that are adjacent to pk and C⁰ ^D.pk/ varies between 0 and 1.

A second family of measurement concepts can be constructed on the structural attribute of betweenness of a point pk considering all the geodesics gij ^D g.pi ; pj / with ⁱ ¤ ^j that contain pk as a point between them. Both Bavelas (1948) and Shaw (1954) suggested that when a person is strategically located on the communication paths linking pairs of others, that person can be considered to have a central position because this person can influence the others by withholding or distorting information in transmission.41 On this theoretical base Freeman (1977) developed a betweenness-based measure defining Gij ^D g.pi ; pj / with ⁱ ¤ ^j as the number of all geodesics linking pi and pj ,and Gij .pk/ with ^k ¤ ⁱ ¤ ^j as the number of geodesics linking pi and pj that contain pk. Then, a betweenness-based absolute measure of point centrality for a point pk can be constructed by summing up the proportion of geodesics linking pi and pj that contain pk:

$$C_B(pk) = \sum{i=1|i \neq j \neq k}^{N} \sum{j=1|j \neq i \neq k}^{N} \frac{G{ij}(pk)}{G{ij}} \,\forall \, k.$$ (5.16)

This provides a measure of the overall centrality of a point pk in a given network structure.⁴² Furthermore, Freeman (1977) showed that for any undirected graph containing ^N points maxCB .pk/ ^D ..N - 1/.N - 2//=2 for all k. Again, this absolute measure of point centrality depends on the total size of the network for which it is calculated and so it should be normalized to get a betweenness-based relative measure of point centrality by calculating

$$C_B'(pk) = \frac{2}{N^2 - 3N + 2} \sum{i=1}^{N} \sum{j=1 | j \neq k}^{N} \frac{G{ij}(pk)}{G{ij}} \,\forall \, k \text{ and } i < i$$ (5.17)

so that the magnitude of C⁰ ^B .pk/ falls again in the range between zero and one.43

⁴¹See Freeman (1979), Burt (1980a, pp. 91-93), Burt (1982, p. 33–35), Wasserman and Faust (1994, pp. 188–190) or Kosch¨utzki et al. (2005, pp. 28–32) with a similar discussion.

⁴²See Anthonisse (1971) or Jansen (2003, pp. 148–149) for the digraph case.

⁴³Following Wasserman and Faust (1994, p. 200), C⁰ ^B .pk / must be multiplied by 2 for digraphs because maxC⁰ ^B .pk / D .N - 1/.N -2/.

The third family of measures is based on closeness or the degree to which a point is close to other points in a network. But looking at communication or efficiency issues, there are some other meaningful interpretations of closeness possible, as shown in e.g. Freeman (1979) or Wasserman and Faust (1994, pp. 183–184). Bavelas (1948) suggested that a message originating in the most central position in a network would spread out in minimum time. Beauchamp (1965) considered closeness as being a basic principle to build up organizations of optimum efficiency in communication. Sabidussi (1966) defined the most central point in a network as that with the minimum costs or time for communication with all other points. For Freeman (1979), a point is central to the extent he can avoid the control potential of other points in the network. Closeness-based measures of centrality have been developed e.g. by Bavelas (1950), Beauchamp (1965), Sabidussi (1966) and Rogers (1974), but for Freeman (1979), the measurement of decentrality by Sabidussi (1966) seems to be the simplest and most natural of these measures. Following him, a closeness-based absolute measure of point centrality can be constructed by summing up all distances dik ^D d.pi ; pk/ with ⁱ ¤ ^k on the geodesics linking a point pk to all other points in the network:44

$$C_C(pk)^{-1} = \sum{i=1|i\neq k}^{N} d_{ik} \ \forall \ k.$$ (5.18)

Being an inverse measure of point centrality for pk , the magnitude of CC .pk/-1 is bigger (smaller) with increasing (decreasing) distance between pk and other points and it is furthermore only meaningful for a connected network, since an unconnected point is equal to an infinite distance dik D1 to all the other points. Again, this absolute measure of point centrality is dependent upon the number of points from which it is calculated. But Beauchamp (1965) solved this problem of standardizing CC .pk/-¹ by dividing it through ^N - 1 as being the minimum sum of distances for pk being adjacent to all other points in a network. A closenessbased relative measure of point centrality can therefore be defined as

$$C'{C}(p{k}) = \left[\frac{1}{N-1} \sum{i=1 | i \neq k}^{N} d{ik}\right]^{-1} = N-1 \left[\sum{i=1 | i \neq k}^{N} d{ik}\right]^{-1} \forall k$$ (5.19)

with minC⁰ ^C .pk/ ^D ⁰ and maxC⁰ ^C .pk/ ^D ¹. ^C⁰ ^C .pk/ can then be viewed as the inverse average distance between pk and all the other actors and equals 1 when pk is adjacent or maximally close to all other actors, like in a star.⁴⁵

Central to the idea of a rank-based measure of centrality (sometimes also called eigenvector centrality) is that the centrality of an actor should be proportional to the

⁴⁴Following Kosch¨utzki et al. (2005) this resembles to much earlier formulations by Wiener (1947) or Shimbel (1953). Further, Wasserman and Faust (1994, pp. 200–201) for the digraph case.

⁴⁵Cf. Wasserman and Faust (1994, p. 185).

strength of the actor's ties to other network members and their centrality. Originally, Bonacich (1972a,b) motivated this centrality measure as an indicator of popularity, related to the measures of rank or status prestige discussed further later on.

Following Bonacich (1972a,b) or Mizruchi et al. (1986), a rank-based centrality measure for the kth actor denoted by CR.pk/, can be expressed as a linear combination of the same centrality measures of other actors in a network, multiplied by the tie strength between them:

$$C_R(pk) = z{1k} C_R(p1) + z{2k} C_R(p2) + z{3k} C_R(p3) + \dots + z{Nk} C_R(p_N) \forall k.$$ (5.20)

This leads to a system of N interdependent linear equations with N unknown rank-based measures of centrality, which can be expressed alternatively in matrix notation by

$$C = Z'C (5.21)$$

with a vector ^C ^D .CR.p1/; CR.p2/; CR.p3/; : : : ; CR.pN //⁰ of rank-based centrality measures for each actor and a ^Z⁰ as a matrix with elements*z*ij ^D ^z.pi ; pj / for all ⁱ ¤ ^j and i;j D 1; : : : ; N representing the relations between all actors. Unfortunately, this system of interdependent linear equations has no trivial nonzero solution. However, Bonacich (1972b) showed in the case of Z being a symmetric matrix of relationships that a solution can be ensured just by multiplying CR.pk/ with a constant equal to the largest positive eigenvalue of Z. Doing this, a similar linear combination of rank-based centralization measures of the form

$$\lambda C_R(pk) = z{1k} C_R(p1) + z{2k} C_R(p2) + z{3k} C_R(p3) + \dots + z{Nk} C_R(p_N) \ \forall \ k$$ (5.22)

occurs, and the resulting system of interdependent linear equations can be expressed in matrix notation by

$$\lambda C = Z'C. \tag{5.23}$$

This is similar to the problem of finding eigenvalues and corresponding eigenvectors in an eigensystem with being an eigenvalue and C an eigenvector. Bonacich (1972b) showed for any symmetric matrix Z that there is always an eigenvalue - = 0 such that all elements of its eigenvector C are greater than or equal to zero. Knowing this, the rank-based measures of centrality from CR.p1/ up to CR.pN / in a given social network structure can be obtained quite easily by (1) extracting the eigenvalues of Z (2) defining as the largest eigenvalue of Z and (3) extracting the corresponding eigenvectorC from Z, whose elements are essentially the rank-based measures of centrality from CR.p1/ up to CR.pN /. 46

⁴⁶Of course, there are a huge amount of other methods to solve this problem of obtaining eigenvalues and eigenvectors in a eigensystem in the field of network analysis proposed by other authors like Katz (1953), Hubbel (1965), Taylor (1969), Bonacich (1972b) and briefly discussed

Although some authors like Mizruchi et al. (1986), Bolland (1988), Bonacich et al. (1998) or Poulin et al. (2000) proposed various methods to standardize these rank-based measures of centrality, a one-for-all solution of standardization for rank-based measures of centrality does not exist up to now, as shown by Ruhnau (2000). The very problem is that either (1) CR.p/ D max.CR.p1/; CR.p2/; CR.p3/; : : : ; CR.pN /⁰ has to be assigned equal to 1 for each network structure separately, because the maximum possible value of CR.p/ is virtually unknown for most N until now or (2) by employing other standardization methods, a maximum value of CR.p/ is not equal to 1 by definition.

5.3.4.2 Measures of Centralization

In addition to all these above mentioned measures of point centrality, corresponding measures reflecting the degree of centralization of an entire network can be calculated. According to Freeman (1979), they all have certain features in common: (1) They should index the degree to which the centrality of the most central point or actor exceeds the centrality of all other points or actors, and (2) they should each be expressed as a ratio of that excess to its maximum possible value for a graph containing the observed number of points or actors. If we define CX.pk/ as one of the three point centrality measures based on degree, closeness and betweenness already defined above and CX .p/ as the maximum value of CX .pk/ for all ^k D 1; : : : ; N in the network then

$$CX = \frac{\sum{k=1}^{N} [C_X(p^) - C_X(pk)]}{\max \sum{k=1}^{N} [C_X(p^) - C_X(p_k)]}$$ (5.24)

is an acceptable measure of centralization of an entire network. It determines the degree to which CX .p/ exceeds the centrality of all other points or actors and, since CX is a ratio of a sum of differences to its maximum value, it will vary between 0 and 1. CX equals 0, when all points or actors in a given social network structure have exactly the same centrality index and it equals 1, if one point or actor is in the centre of a star or wheel and therefore dominating the other points or actors.47

Using the above mentioned expression, a degree-based measure of graph centrality is defined as

$$CD = \frac{\sum{k=1}^{N} \left[ C_D(p^*) - C_D(p_k) \right]}{N^2 - 3N + 2}$$ (5.25)

in Wasserman and Faust (1994, pp. 206–209) or more recent contributions by Richards and Seary (2000), Poulin et al. (2000), Bonacich and Lloyd (2001) or Kosch¨utzki et al. (2005, pp. 46–51).

⁴⁷Cf. Wasserman and Faust (1994, p. 177).

because .N -1/.N -2/ is the largest sum of differences in terms of degree. It can be regarded as a general formula for determining the centrality of a network in terms of degree. Furthermore, Freeman (1977) defined a betweenness-based measure of graph centrality as the average difference between the relative centrality of the most central point C⁰ ^B .p/ and that of all other points. Freeman (1979) shows, that

$$CB = \frac{\sum{k=1}^{N} \left[ C_B'(p^) - C_B'(pk) \right]}{N-1} = \frac{\sum{k=1}^{N} \left[ C_B(p^) - C_B(p_k) \right]}{N^3 - 4N^2 + 5N - 2}$$ (5.26)

and that this measure takes its maximum value for network in form of a star or a wheel. Thus, CB provides a general measure of graph centrality based on betweenness. Finally, a closeness-based measure of graph centrality can be defined as

$$CC = \frac{\sum{k=1}^{N} \left[ C_C'(p^*) - C_C'(p_k) \right]}{(N^2 - 3N + 2)/(2N - 3)}$$ (5.27)

due to the fact that the maximum possible difference in terms of closeness is .N - 2/.N - 1/=.2N - 3/. ⁴⁸ In all of these three measures, a network in form of a star or a wheel yields the maximum sum of differences by considering the alternatives. Conversely, the lowest value of graph centrality is assigned to a complete graph or a social network structure where all points or actors are connected to each other.

In the case of rank-based measures of centralization, this proposed method of calculating a graph centrality cannot be seriously employed for most networks, because the maximum of CR.p/ is virtually unknown there. But Wasserman and Faust (1994) proposed another family of measures expressing centralization of a graph or social network structure by simply calculating overall means and variances, which at the same time can be considered as a measure for heterogeneity of points or actors in a graph.

Following them, degree-based point centrality can be summarized on the overall graph level by

$$\overline{C}D = \frac{1}{N} \sum{k=1}^{N} C_D(p_k) \text{ and } SD^2 = \frac{1}{N} \sum{k=1}^{N} [C_D(p_k) - \overline{C}_D]^2$$ (5.28)

where C ^D can be regarded as an average of actor degrees and S² ^D expresses the variance of actor degrees. Unfortunately, the maximum values of C ^D and S² D depend on N and the entire set of degrees.49 In order to obtain a standardized measure of ^C ^D, it can be simply divided by .N -1/ to get a measure called average

⁴⁸Cf. Freeman (1979) or Wasserman and Faust (1994, p. 186).

⁴⁹See Snijders (1981) or Wasserman and Faust (1994, p. 181) for this and the following discussion.

standardized degree

$$\overline{C}'{D} = \frac{1}{N(N-1)} \sum{k=1}^{N} C{D}(p{k}) = \frac{1}{N} \sum{k=1}^{N} C'{D}(p_{k})$$ (5.29)

which is simply an average of degree-based measures of point centrality. To standardize S² ^D, a complicated procedure calculating a maximum possible variance given the set of degrees actually observed to obtain a dimensionless index is proposed by Snijders (1981).

In analogy to the degree-based measures, closeness-based measure of point centrality can be summarized on the overall graph level simply by calculating

$$\overline{C}'C = \frac{1}{N} \sum{k=1}^N C'_C(p_k) \text{ and } SC^2 = \frac{1}{N} \sum{k=1}^N [C'_C(p_k) - \overline{C}_C]^2 \text{ with } \overline{C}C = \frac{1}{N} \sum{k=1}^N C_C(p_k),$$ (5.30)

where C⁰ ^C can be called average standardized closeness and S² ^C variance of actor closeness, respectively.⁵⁰ S² ^C especially can be regarded as an index of heterogeneity, because with an increasing (decreasing) S² ^C , a network becomes less (more) homogeneous with respect to distance and thus more (less) centralized. The same calculus can be conducted for betweenness-based and rank-based measures of point centrality, obtaining

$$\overline{C}'{B} = \frac{1}{N} \sum{k=1}^{N} C'{B}(p{k}) \text{ and } S{C}^{2} = \frac{1}{N} \sum{k=1}^{N} [C'{B}(p{k}) - \overline{C}{B}]^{2} \text{ with } \overline{C}{B} = \frac{1}{N} \sum{k=1}^{N} C{B}(p_{k})$$ (5.31)

and

$$\overline{C}R = \frac{1}{N} \sum{k=1}^{N} C_R(p_k) \text{ and } SR^2 = \frac{1}{N} \sum{k=1}^{N} [C_R(p_k) - \overline{C}_R]^2$$ (5.32)

as betweenness-based and rank-based measures of centralization.

5.3.4.3 Measures of Prestige and Hierarchization

As discussed already before, prestige can be measured by looking at asymmetric relations between actors in social network structures with actors being prestigious

⁵⁰Cf. Wasserman and Faust (1994, p. 187). Further, Wasserman and Faust (1994, pp. 200–201) for the digraph case.

to the extent they receive choices from others. Therefore, the simplest measure of prestige on actor-level is the actor's degree prestige, which is just the indegree of an actor.51 The idea is that actors who are prestigious tend to receive many nominations or choices, so that

$$P_D(p_k) = d_1(pk) = \sum{i=1|i\neq k}^{N} z_{ik} \ \forall k.$$ (5.33)

Again, the value of PD.pk/ depends on ^N, so that it should divided by ^N - 1 to get

$$P_D'(p_k) = \frac{d_I(p_k)}{N-1} \,\forall \, k$$ (5.34)

as a standardized degree prestige measure for the kth actor.⁵² Then, a maximum of actor's prestige based on degree occurs for P⁰ ^D.pk/ ^D ¹, when pk is chosen by all other actors and a minimum of P⁰ ^D.pk/ ^D ⁰ if pk does not get choices from other actors.

Using the indegree as an indicator of prestige, only those actors who are adjacent are taken into consideration. But other actors outside this first-order zone of ego can make some contribution to the prestige of an actor, too. Therefore, the concept of degree prestige should be generalized by defining an influence domain of pk as the set of actors who are either directly or indirectly connected to him (or her).⁵³ Such actors are reachable to him (or her), or alternatively, are those, from which he (or she) is reachable. A simple measure for the magnitude to this influence domain of pk is just the number of actors in this influence domain, denoted by Ik. Furthermore, it has to be taken into account that these actors in the influence domain of pk are more or less distant located to him (or her) so that at least an average distance of the other actors to pk has to be included. Then, an actors's proximity prestige can be calculated as a proportion of actors who can reach pk to the average distance these actors are from pk or expressed mathematically

$$P_P(p_k) = \frac{Ik/(N-1)}{\sum{i=1|i\neq k}^{N} d_{ik}/I_k} \,\forall \, k.$$ (5.35)

⁵¹Cf. Knoke and Burt (1983, pp. 200–202), Wasserman and Faust (1994, p. 202) or Jansen (2003, pp. 142–145). According to Knoke and Burt (1983, p. 201), Moreno (1934, pp. 98–103) was the first author who introduced such a degree-based measure of prestige.

⁵²This sometimes called a relative indegree, according to Wasserman and Faust (1994, p. 203).

⁵³Cf. Burt (1980a, pp. 93–95), Burt (1982, pp. 33–35), Knoke and Burt (1983, pp. 203–204), Wasserman and Faust (1994, pp. 203–204) or Jansen (2003, pp. 145-147). According to Wasserman and Faust (1994, p. 203), Lin (1976, pp. 340–349) was the first author who defined such an influence domain in order to construct a measure for proximity prestige.

Again, the maximum of PP .pk/ ^D ¹ where all actors are adjacent to him, so that Ik ^D ^N -¹ and dik ^D ¹. The minimum of PP .pk/ ^D ⁰ is reached, if pk is

unreachable, so that Ik ^D ⁰. In addition to prestige based on indegree or proximity of actors to pk, a measurement concept based on the prominence of actors can be defined, which is called status or rank prestige of pk. ⁵⁴ There, pk is considered to have a high status or prestige, if the influence domain of pk is full of actors, who have a high status or prestige, too. Unfortunately, to calculate this measure again some sophisticated mathematics are needed, because an actor's status or rank depends on the status or rank of the others connected to him or her and the ranks or status of those again depends on the status or rank of the others connected to them, and so ad finitum. Seeley (1949) first proposed a solution to this infinite chain of dependencies and this solution was subject to further discussion e.g. by Katz (1953), Hubbel (1965), Taylor (1969), Bonacich (1972a,b, 1987) or Mizruchi et al. (1986) to name only a few authors.⁵⁵

The simplest way to express this problem of infinite chain of interdependencies is e.g. to define a rank prestige measure for the kth actor, denoted by PR.pk/, as a linear combination of rank prestige measures of the other actors in a network:

$$P_R(pk) = z{1k} P_R(p1) + z{2k} P_R(p2) + z{3k} P_R(p3) + \dots + z{Nk} P_R(p_N) \forall k.$$ (5.36)

This leads to a system of N interdependent linear equations with N unknown variables, which can be expressed in matrix notation by

$$P = Z'P (5.37)$$

with a vector ^P ^D .PR.p1/; PR.p2/; PR.p3/; : : : ; PR.pN //⁰ of rank prestige measures for each actor and ^Z⁰ as a matrix with elements *^z*ij ^D ^z.pi ; pj /´ for ⁱ ¤ ^j and i;j D 1; : : : ; N representing the relations between all actors. By rearrangement

$$(\Phi - Z')P = 0 \tag{5.38}$$

is obtained, which is identical to a characteristic equation usually used to extract eigenvalues and eigenvectors in an eigensystem, where P is an eigenvector of Z⁰ corresponding to an eigenvalue of 1 and ˚ standing for an identity matrix. One of various methods to tackle this problem of eigendecomposition is recommended by Katz (1953). Following him, this problem of eigendecomposition can easily be solved by (1) standardizing Z to have columns of unity, (2) extracting the

⁵⁴Cf. Knoke and Burt (1983, pp. 206–207), Wasserman and Faust (1994, pp. 207–209) or Jansen (2003, pp. 149–153).

⁵⁵See Wasserman and Faust (1994, pp. 207–210) or Kosch¨utzki et al. (2005, pp. 46–51) for a brief review of this literature.

eigenvalues of Z⁰ getting at least one largest eigenvalue with a value of 1 and (3) extracting the corresponding eigenvector P from Z⁰ , whose elements are essentially the rank prestige measures from PR.p1/ up to PR.pN /. 56

Last, but not least, some might think that a betweenness-based measure of prestige can also be defined, but this kind of measurement concept is virtually nonreflected in literature up to now.⁵⁷ There is a strong reason for this: All measurement concepts of prestige discussed above are based on choices received from other actors, but a betweenness-based measure has to build on the notion that an actor is between other actors and therefore he receives choices in order to give choices to other actors. Therefore, betweenness-based measures of centrality CB .pk/ and C⁰ ^B .pk/ applied to directed graphs might be more useful.

In analogy with the measures of centralization, measures of hierarchization on the network-level can be calculated employing actor's prestige measures presented above. If we define PY .pk/ standing for one of the three actors prestige measures defined above and PY .p/ as the maximum value of PY .pk/ for any point in the network then

$$PY = \frac{\sum{k=1}^{N} [P_Y(p^) - P_Y(pk)]}{\max \sum{k=1}^{N} [P_Y(p^) - P_Y(p_k)]}$$ (5.39)

is a measure of network hierarchization. ⁵⁸ Accordingly, mean and variance of prestige can be calculated in order to get

$$\overline{P}Y = \frac{1}{N} \sum{k=1}^{N} P_Y(p_k)$$ (5.40)

and

$$S_{PY}^2 = \frac{1}{N} \sum{k=1}^{N} [P_Y(p_k) - \overline{P}_Y]^2$$ (5.41)

as measures of prestige or hierarchization on the network-level.⁵⁹ PY will then be between 0 for an empty directed graph and 1 for a complete directed graph. S² PY will be positive and measures how much heterogeneity is present with respect to indegree, proximity, rank or status.

⁵⁶See e.g. Burt (1980a, pp. 93–95), Burt (1982, pp. 33–35), Scott (1991, p. 55), Wasserman and Faust (1994, pp. 206-209), Richards and Seary (2000) and Poulin et al. (2000) for a detailed discussion of this and other more refined eigendecomposition methods.

⁵⁷Cf. Knoke and Burt (1983, pp. 204–206) or Jansen (2003, pp. 149–149).

⁵⁸Cf. Jansen (2003, p. 144).

⁵⁹Cf. Wasserman and Faust (1994, p. 204) or Jansen (2003, pp. 142–143).

5.3.4.4 Refinements and Comparisons

Up to this point, four general motivations for centrality were exploited: (1) Actors are central if they are active in the network (degree-based centrality), (2) actors are central if they can contact others through efficient (short) paths (closenessbased centrality), (3) actors are central if they have the potential to mediate and/or control flows of resources or information between other actors (betweenness-based centrality) or (4) actors are central if they have ties to other actors that are themselves central (rank-based centrality).⁶⁰ But as these measures of centrality are not suitable for every instance, following refinements were proposed in the literature:

Bonacich (1987) refined his rank-based measure of centrality introducing a beta factor which allowed him to define power-based measures of centrality with varying degree to which status is transmitted from one actor to another. Moreover, this beta factor permits the assessment of power in negatively connected networks.61
Thinking of communication not always occurring on the shortest possible path along the geodesics like in the case of closeness-based centrality, Stephenson and Zelen (1989) constructed information-based measures of centrality by taking all possible paths (including the geodesics) and their information content into consideration. It focuses on the information contained in all paths originating from a specific actor, the information of an actor is then considered averaging the information in these paths, which, in turn, is inversely related to the variance in the transmission of a signal from one actor to another.
Freeman et al. (1991) introduced flow-based measures of centrality based on a model of network flows.⁶² Comparing it with the original betweenness-based measure of centrality, they found that they do not yield to the same results except the special case where the set of paths linking each pair of points or actors pi and pj is equal to the set of geodesic linking those points or actors.
Friedkin (1991) derived three complementary measures called (1) total effects centrality (TEC), (2) immediate effects centrality (IEC) and (3) mediative effects centrality (MEC) out of a network effect model, which are by definition rather congruent with the rank-based, closeness-based and betweenness-based measures of point centrality already defined above.
Furthermore, measures of centrality based on random-walks were defined e.g. by Noh and Rieger (2004) or Newman (2005) with the first closely related to closeness-based and the second to betweenness-based centrality so that it can be called random-walk closeness and random-walk betweenness, respectively. Following Noh and Rieger (2004), random-walk closeness is a measure of

⁶⁰See e.g. Bolland (1988).

⁶¹Cf. Bonacich and Lloyd (2004) or Bonacich (2007).

⁶²These centrality measures are often called flow betweenness, see e.g. Newman (2005) or Borgatti and Everett (2006).

relative speed with which randomly walking messages reach a player from elsewhere in the network, whereas Newman (2005) random-walk betweenness is appropriate to networks in which information wanders about essentially random until it finds its target and it includes contributions from many paths that are not optimal in any sense, although shorter paths still tend to count more than longer ones since it is unlikely that a random walk becomes very long without finding the target.

While describing procedures for finding key players in a social network structure, Borgatti (2006) constructed two new measures: (1) degree of fragmentation defined as the average reciprocal distance among actors after removal of a given actor and (2) degree of reach as the weighted proportion of all actors reached by a clique, where actors are weighted by their minimum distance from the clique and only actors at distance 1 are given full weight. The first catches the amount of reduction on cohesiveness of a network structure that would occur if a key player would not be present, whereas the second describes the extent to which key players are connected to and embedded in the network around them.
Last but not least, Ballester et al. (2006) introduced an intercentrality measure which takes into account both an actor's centrality and his contribution to the centrality of the others. In contrary to rank-based centrality, it does not derive from individual considerations, but from collective concerns in order to identify key players whose removal would result in the maximal decrease in overall activity.

Apart from these refinements, some authors like Freeman (1979), Bolland (1988), Nakao (1990), Wasserman and Faust (1994, pp. 198–202), Marsden (2002), Everett and Borgatti (2005) or Borgatti et al. (2006) conducted studies concerning reliability and/or applicability of centrality measures for different sorts of network structure:

Freeman (1979) himself underwent an extensive study to show the usability of his measures of centrality. According to him, betweenness-based centrality seems to be most suitable to indicate importance of an actor in social network structure. Other important findings were that betweenness-based measures of centrality are subject to the highest and degree-based measures of centrality to the lowest variation in values.
Bolland (1988) surveyed Freeman's measures of centrality based on degree, closeness and betweenness as well as the rank-based centrality measure by Bonacich (1972a,b), which the author named "continuous flow model" combining degree-based and closeness-based centrality. He conducted a series of analysis in order to test all four centrality measures for robustness and sensitivity under conditions of random and systematic variations in a community influence network. All four measures of centrality performed well, although all three Freeman's measures of centrality are quite sensitive to random and systematic errors. Furthermore, degree-based, closeness-based and continuous-flow-based measures of centrality were subject to high intercorrelations, suggesting quite a similarity in the measurement concept. Summing up his findings, Bolland
recommended to use betweenness-based and rank-based measures of centrality, because they are complements to each other with one based on the ability to facilitate network flow while the other is based on the ability to hinder flow.
Nakao (1990) compared the three Freeman's measures of graph centrality CB , CC and CD. For him, there is an extremely high linear association between CC and CD with all three closely interrelated but subject to different order of magnitude: in chain-like or decentralized networks divided into sub-clusters which are connected to each other by a line or via a focal player with a small amount of relations close to ^N - 1 the order tend to be CB > CC > CD whereas in one-cluster network settings coming along with many relations it changes to CC > CD > CB .
Wasserman and Faust (1994, pp. 198–202) showed that the centrality measures based on degree, betweenness, closeness, information and network flow as well as the alternative centrality measures by Friedkin (1991) can be applied to digraphs. But only degree-based and closeness-based measures of centrality are considered to be suitable for directed relations and therefore are recommended by them, in addition to all measures of prestige discussed in the last section.
Marsden (2002) examined ego-centric versions of Freeman's centrality measures, focusing on edges inside the first-order zone around an ego point pk and compared it to sociometric centrality measures calculated out of complete (socio-centric) network data. He came to the conclusion that in the case of degree-based measures of centrality both versions yield identical information, because (by definition) an ego-centric network in the first-order zone around the ego point pk includes all other points pi (his alter) adjacent to him. Furthermore, a betweenness-based measure of centrality in an ego-centric network will correspond imperfectly to socio-centric version, being smaller in magnitude. A simple reason to this is that for ego-centric network data, the maximum geodesic distance between two points pi and pj is 2, since by definition all pairs of points within the first-order zone of pk are connected either directly or indirectly via pk (as well as other possible connections between pi and pj within the ego's first-order zone). So the betweenness measure of centrality will not cover indirect linkages with a geodesic distance of length 3 and more in ego-centric networks. But comparing betweenness-based measures calculated both in the ego-centric and socio-centric versions for 17 classical network structure studies of varying network size, they find a quite high correlation between them, being usually above 0:9. In some special kinds of network structure, both measures of betweenness are identical. For example, in a complete (fully connected) graph, all points have a betweenness of 0, since there are no geodesics longer than 1 in such a network. Likewise, in a network of maximum centralization in form of a star, both centrality measures are again identical, because all peripheral points (having a betweenness of 0) are directly connected to the central ego point pk. At last, closeness-based measures of centrality are found to be uninformative in the case of ego-centric networks since all geodesic distances from the ego point pk to other points in the first-order zone are virtually 1 by definition.
Everett and Borgatti (2005) again recommended betweenness-based centrality as being useful in ego-centric networks and showed a high correlation association to the socio-metric counterpart by simulating networks of different size. Furthermore, they found degree-based centrality to be virtually the same as the degree of an actor, whereas closeness-based and rank-based measures of centrality are simply not applicable because all relations outside of the first-order zone around an ego point pk are left out.
Finally, Borgatti et al. (2006) analyzed the robustness of measures of centrality in the face of random error in the network data. One of their key findings was that the four centrality measures based on degree, betweenness, closeness and rank are surprisingly similar with respect to deletions or additions of edges and nodes of randomly generated network data of different size. Generally, dense networks were the most robust to all error types with the notable exception of edge deletion where sparse networks were more accurately measured.

To sum up this review of measurement approaches to importance or prominence, classifications by Knoke and Burt (1983) and Borgatti and Everett (2006) are shown in order to clarify their similarities and differences.

It were Knoke and Burt (1983) who presented a first coherent methodological framework including most of these measures of centrality, centralization, prestige and hierarchization well defined before. According to them, measures of centrality show prominence reflected by an actor's aggregate involvement in relations, his access to and control of information in a social network structure of symmetric relations, whereas prestige measures reflect the extent to which an actor is the object of relations and show a grade of difference accorded to him by others in the same networks with asymmetric relations. Furthermore, Knoke and Burt (1983) distinguished at least eight different measurement concepts as shown in Table 5.3. 63
Borgatti and Everett (2006) followed a graph-theoretic approach and reviewed centrality measures according to the way they are calculated in the case of undirected graphs. There, they distinguished four basic dimensions of which two are of special interest in this context: the properties of walks measured (or walk property) and the type of nodal involvement (or walk position).64 The walk property dimension distinguishes between volume measures that evaluate the number of walks a node is involved in and length measures that evaluate the length of those walks. The walk position dimension distinguishes between radial measures that evaluate walks emanating from a node and medial measures that

⁶³Originally, Knoke and Burt (1983) named their measurement concepts VAR1 to VAR5, with VAR1 equivalent to degree-based measures, VAR3 to closeness- or proximity-based measures, VAR4 to betweenness-based measures and VAR5 to rank- or status-based measures. VAR2 are measures based on connectivity or reachability, a measurement concept quite similar to degree, which was excluded from investigation.

⁶⁴The other two are the types of walks considered (such as only geodesics, only true paths, limited length walks, and so on) and the choice of summary statistic (e.g. sum or average).

| Table 5.3 | Measurement concepts for prominence (Source: adapted from Knoke and Burt (1983, | | |-----------|---------------------------------------------------------------------------------|--| | p. 200)) | | |

	Centrality and Centralization: Prominence in a Network of Symmetric Relations	Prestige and Hierarchization: Prominence in a Network of Asymmetric Relations
Volume of relations:
Many ties: Degree or Indegree	Many actors directly and/or indirectly connected with ego	Many actors directly and/or indirectly connected to ego
Many close ties: Closeness or Proximity	Strong relations with other actors	Strong relations from other actors
Quality of relations:
Kinds of ties: Betweeness	Involvement in all connections between actors	Involvement in all asymmetric connections between actors
Kinds of actors: Rank or Status	Strong relations with central actors	Strong relations from prestigous actors

Table 5.4 Cross-classification of centrality measures (Source: adapted from Borgatti and Everett, 2006)

	Radial (Focus on Group Membership)	Medial (Focus on Brokerage)
Volume (number of walks envolved)	Degree by Freeman (1979), TEC by Friedkin (1991)	Betweenness by Freeman (1979), MEC by Friedkin (1991), Freeman et al. (1991), Anthonisse (1971), Newman (2005)
	Bonacich (1972a, 1972b, 1987),
	Katz (1953), Hubbel (1965)
Length
(length of walks envolved)	Closeness by Freeman (1979), Stephenson and Zelen (1989), IEC by Friedkin (1991)	Degree of Fragmentation by Borgatti (2006)

evaluate walks passing through a node.65 Table 5.4 shows a cross-classification of walk property (volume vs. length) and walk position (radial vs. medial) for a number of centrality measures discussed before. Centrality measures within the same quadrant can be thought of as competitive, because they are constructed likely the same way and so substitutable. In contrary, measures from different quadrants differ in fundamental ways and should be regarded at best as complementaries with the differences in walk position being more important than the distinction of counting number of walk or evaluating their length.

⁶⁵Or in the words of Borgatti and Everett (2006, p. 481): "Whereas radial measures assess group membership, medial measures assess bridging, reminiscent of the distinction in the social capital literature of bonding social capital and bridging social capital, or closed versus open ego networks."

5.3.5 Network Constraint and Brokerage Concepts

After discussing basic measures on the individual, relational, and network level as well as measurement concepts of prominence, now a focus is set on two relational approaches of measurement. So in the following, first Burt (1992, 2000) network constraint concept in networks with structural holes is presented before a more focused measurement concept aiming at the capacity of brokerage of actors in networks with non-overlapping subgroups by Gould (1989) or Gould and Fernandez (1989) is shown. Following Marsden (2002, pp. 417–418) or Borgatti and Everett (2006, p. 474), both measurement concepts can be considered as being closely related to the already above discussed measures of betweenness-based centrality applied to ego-centric networks.

5.3.5.1 Network Constraint Concept

According to Burt (2000, pp. 346–355), a variety of network mechanisms make contradictory predictions about social capital. First, network models of prominence (as shown in the last section) argue that well connected actors perceive a high amount of social capital from their position, because they get all information available most directly. Second, if the risks associated with incomplete information provision are high, close networks, where everyone is connected with each other are generating a trustworthy environment, which creates social capital. By opposing this idea of the network closure argument, Burt (1992) described social capital as a function of brokerage opportunities in networks with structural holes separating nonredundant sources of information. Therefore, actors spanning structural holes are better off, being able to monitor and process information under their control as a source of social capital.

In order to quantify his theory of social capital in networks with structural holes, Burt (1992, pp. 51–56, 1998:Appendix, 2000:373-375 and Appendix) developed (1) a measure of effective network size NS^k and (2) a network constraint index NC^k to describe the extent to which a social network structure around k is constrained by a multitude of redundant contacts with other actors.

Following Burt (1992, pp. 51–54), effective network size NS^k for an actor k can be defined as

$$NSk = \sum{i=1|i \neq j \neq k}^{N} \left( 1 - \sum{j=1|j \neq i \neq k}^{N} p{kj} v_{ij} \right) \forall k$$ (5.42)

with

$$p{kj} = \frac{(z{kj} + z{jk})}{\sum{i=1|i \neq j \neq k}^{N} (z{ki} + z{ik})} \,\forall \, k, j$$ (5.43)

as the proportion of k's network time and energy invested in the relationship with another actor j to whom actor i has strong ties and

$$v{ij} = \frac{\left(z{ij} + z{ji}\right)}{\max\left(z{il} + z_{li}\right)} \quad \forall \ j \neq i \neq l$$ (5.44)

as the marginal strength of ⁱ's relations with contact ^j where max .*z*i l ^C *^z*l i/ is the largest of i's relations with anyone else denoted by l. Actually, NS^k measures the amount of non-redundant contacts, as the bracketed term equals 1 if i is completely disconnected from all other primary contacts in the network. If relations between i and other contacts strengthen, the bracketed term approaches pki , indicating that i is completely redundant with other contact in k's network. Therefore, NS^k varies between 1, indicating the network only provides a single contact, and N, indicating that every contact in the network is non-redundant.

Thinking again of k engaged in a relationship with another actor i, his contactspecific constraint can be expressed as

$$c{ki} = \left(p{ki} + \sum{j=1|j\neq i\neq k}^{N} p{kj} \, p_{ji}\right)^2 \, \forall \, k, i$$ (5.45)

where

$$p{ki} = \frac{z{ki}}{\sum{j=1|j\neq i\neq k}^{N} z{kj}} \,\,\forall\,\, k, i$$ (5.46)

is the proportion of k's time and energy invested in a direct relationship with i and the other fraction in parentheses is an expression for the proportion of k's relationships in redundant contacts j , because k invested in other direct relationships with third actors j at the same time, who in turn invested in a relationship with i, too.⁶⁶ Being a squared sum of these valuations of direct and indirect relationships between k and i, cki varies from a minimum constraint of p² ki (i is disconnected from all other contacts of k) up to maximum constraint equal to 1 (if i is the only contact). Summing up all these contact-specific constraints of k leads to the network constraint index NC^k with

$$NCk = \sum{i=1|i\neq k}^{N} c{ki} = \sum{i=1|i\neq k}^{N} \left( p{ki} + \sum{j=1|j\neq i\neq k}^{N} p{kj} p{ji} \right)^2 \forall k.$$ (5.47)

⁶⁶See Burt (1992, pp. 54–55), Burt (1998, Appendix) or Burt (2000, Appendix) for further details.

Following Burt (2000, p. 373), NC^k is high if actors in a social network structure are directly connected to each other (like in a dense network) or indirectly connected via a central actor (like in a hierarchical network) or in his words:

"More constrained networks span fewer structural holes, which means less social capital according to the hole argument. If networks that span structural holes are the source of social capital, then performance should have a negative association with network constraint. More constraint means more network closure, and so more social capital according to the closure argument. If network closure is the source of social capital, then performance should have a positive association with constraint."

Furthermore, it can be shown that NC^k varies in three other measures of social network structure, namely (1) network size, (2) network density and (3) network hierarchy. By binomial expansion, cki and NC^k can be restated as

$$c{ki} = p{ki}^2 + 2p{ki} \sum{j=1|j\neq i\neq k}^{N} p{kj} p{ji} + \left(\sum{j=1|j\neq i\neq k}^{N} p{kj} p_{ji}\right)^2 \forall k, i$$ (5.48)

and

$$NC{k} = \sum{i=1|i\neq k}^{N} p{ki}^{2} + 2\sum{i=1|i\neq k}^{N} p{ki} \sum{j=1|j\neq i\neq k}^{N} p{kj} p{ji} + \sum{i=1|i\neq k}^{N} \left(\sum{j=1|j\neq i\neq k}^{N} p{kj} p{ji}\right)^{2} \forall k.$$ (5.49)

Then, the last expression can be regarded as a sum of three components called NC-size, NC-density and NC-hierarchy. ⁶⁷ All these three components reach a high value for an actor k being constrained and can be interpreted as follows:

First, constraint is more severe in smaller networks, because if there are only few members in a network structure they have few alternative contacts to provide information and opportunistic play against the others. Furthermore, pki decreases on average as the total number of relationships increases. The NC-size component captures this fact as a Herfindal index measuring the extent to which k's relationships are concentrated in a single contact.
Second, constraint is more severe in more dense network structure, because this contains, in effect, fewer alternative contacts. Furthermore, strongly intertwined contracts are more likely to provide identical information, and it is then more difficult for each actor to play opportunistically against another. In a more dense network, there are more indirect relationships, which can be quantified either by ND or by the NC-density component measuring the extent to which k's strongest relationships are strongly tied to his other contacts.

⁶⁷Cf. Burt (1998, Appendix, 2000, Appendix).

• Third, constraint is more severe in more hierarchical network structures, because they again contain, in effect, fewer alternative contacts. As network structure is more hierarchical to the extent that it is organized around a single contact, k is more constrained when this contact has exclusive relations with k's other relationships. Therefore, the NC-hierarchy component measures the extent to which k's contacts concentrate their relations on one central contact in the network.⁶⁸

5.3.5.2 Brokerage in Networks with Non-overlapping Subgroups

Gould and Fernandez (1989, pp. 92–102) (1) proposed a formal definition of brokerage in networks with non-overlapping subgroups, (2) showed that these structures correspond to intuitive and meaningful brokerage roles and (3) developed a set of quantitative measures based on the concept of betweenness.

In first instance, any brokered relationship between two actors (one called sender or first-party denoted as pi and one receiver or second-party denoted as pj ) can be seen as a set of relations forming a triad where the third actor (denoted as pk) is involved in their relationship as a third-party in-between called intermediary, agent or broker.⁶⁹ Assuming that there is no other direct relationship between pi and pj , a two-step path with .pi ; pk/ and .pk; pj / is a set of brokerage relations forming the triad. Summing up these two-step paths for e.g. a particular pk leads to individual raw brokerage scores standing for total capacity for brokerage. Following Freeman (1977) concept of betweenness, a partial brokerage score can also be defined to reflect the extent to which pk is in control of the path .pi ; pk/ and .pk; pj / given any pi and pj . But two possible situations have to be taken into account: (1) pk is the only intermediary between pi and pj with maximum control and partial brokerage score of 1 or (2) there are k intermediaries between pi and pj , each with a partial brokerage score of 1=k. ⁷⁰ Which form of brokerage measure is employed in an analysis of brokerage roles depends on the focus of an analysis (Gould and Fernandez (1989, p. 98)):

"If the researcher is interested in the number of brokerage relations an actor is capable of mediating, the appropriate measure is the absolute number of paths on which the actor lies; this corresponds to the individual's total capacity for brokerage. If, on the other hand, the central issue is the degree to which the actor actually controls brokerage relations in the network, then the appropriate measure is the partial [brokerage] score."

⁶⁸Burt (1998, Appendix, 2000, Appendix) proposed two alternative measures of network hierarchy: (1) the Coleman-Theil disorder index applied to contact-specific constraint scores as developed first in Burt (1992, pp. 70–71) and (2) the betweeness-based measure of point centrality Cb.pk / by Freeman (1977) as already defined in the last section measuring the extent to which one contact stands in between all others.

⁶⁹Strictly speaking, the terms sender and receiver imply to some extent directed relations from a sender to a receiver. So in the following, the terms first- and second-party instead of sender and receiver are used in order to indicate that these relations may be mutual.

⁷⁰Cf. Gould (1989, p. 537) or Gould and Fernandez (1989, p. 98).

In addition to this, the three actors forming a triad can be members of distinct non-overlapping subgroups, so that exactly five structurally different types of brokerage relations can occur as shown in Fig. 5.2: ⁷¹

All actors may belong to the same subgroup, so that the brokered relationship is completely internal to the group, and the intermediary therein involved can be regarded as a local broker or coordinator.
If both pi and pj belong to the same subgroup while pk is staying outside as a member of another subgroup, then the intermediary can be called an itinerant broker or cosmopolitan.
If the pj and pk form a subgroup and pi is in another, the intermediary is considered as a gatekeeper, acting as a second-party's agent vis-`a-vis the first-party.

Fig. 5.2 Graphic representation of five different types of brokerage. (Source: adapted from Gould and Fernandez (1989, p. 93))

⁷¹Notably, this classification of brokerage roles draws heavily reference to individual communication roles in organizations as defined by Rogers and Agarwala-Rogers (1976, pp. 132–140) proposing a four-fold typology with (1) gatekeepers, (2) liaison, (3) opinion leaders and (4) cosmopolites or boundary spanners. Whereas the gatekeeper and liaison role is defined exactly in the same way as in Rogers and Agarwala-Rogers (1976) and the opinion leader corresponds to the local broker or coordinator, they defined the cosmopolite or boundary spanner as an individual who has a relatively high degree of communication with the outside environment of an organization so that it corresponds at best with the representative role as defined by Gould and Fernandez (1989). Further, as Rogers and Agarwala-Rogers (1976) set a focus on communication networks in organizations, they did not define a counterpart to Gould and Fernandez (1989) itinerant broker or cosmopolitan.

Conversely, if pi and pk form a subgroup excluding pj , the intermediary is called a representative, because he is the agent of the first-party representing him vis-`avis the second-party.
Last but not least, all three actors can be members of different subgroups, so that the intermediary's brokerage role is simply to act as a third-party to link two members of distinct subgroups in a liaison.

These five types of brokerage represent specific structural positions as concrete brokerage roles in triads that actors can occupy forming a social network structure. But it is worth to mention that while any given brokerage relationship falls into one of these five categories, an individual actor having several relationships with other actors can serve of course different brokerage roles simultaneously.

In order to construct now measures for brokerage roles, Gould and Fernandez (1989) defined an actor's total raw brokerage score for the kth intermediary

$$Tk = w{1k} + w{0k} + b{01k} + b{10k} + b{0k} \ \forall \ k$$ (5.50)

as the sum of five individual raw brokerage scores for the kth intermediary

$$w{1k} = \sum{i=1|i \neq k}^{N} \sum_{j=1|j \neq i \neq k}^{N} w_1(ij) \ o(ij)$$

with o.ij / ^D ¹, if *^z*ij .o/ ^D *^z*ik.o/ ^D *^z*kj .o/ ^D ¹ and *^w*1.ij / ^D ¹ if a set of relations is between pi, pk and pj , and else 0

$$w{0k} = \sum{i=1|i \neq k}^{N} \sum_{j=1|j \neq i \neq k}^{N} w_0(ij) \ o(ij)$$

with o.ij / ^D ¹, if *^z*ij .o/ ^D ¹ but *^z*ik.o/ ^D *^z*kj .o/ ^D ⁰ and *^w*0.ij / ^D ¹ if a set of relations is between pi , pk and pj , and else 0

$$b{01k} = \sum{i=1|i\neq k}^{N} \sum{j=1|j\neq i\neq k}^{N} b{01}(ij) \ o(ij)$$

with o.ij / ^D ¹, if *^z*kj .o/ ^D ¹ but *^z*ij .o/ ^D *^z*ik.o/ ^D ⁰ and b01.ij / ^D ¹, if a set of relations is between pi , pk and pj , and else 0

$$b{10k} = \sum{i=1|i\neq k}^{N} \sum{j=1|j\neq i\neq k}^{N} b{10}(ij) \ o(ij)$$

with o.ij / D ¹, if *^z*ik.o/ ^D ¹ but *^z*ij .o/ ^D *^z*kj .o/ ^D ⁰ and b10.ij / ^D ¹, if a set of relations is between pi , pk and pj , and else 0

$$b{0k} = \sum{i=1|i\neq k}^{N} \sum_{j=1|j\neq i\neq k}^{N} b_0(ij) \ o(ij)$$

with o.ij / ^D ¹, if *^z*ij .o/ ^D *^z*ik.o/ ^D *^z*kj .o/ ^D ⁰ and bO.ij / ^D ¹, if a set of relations is between pi , pk and pj , and else 0

(5.51)

for all k corresponding to each different type of brokerage defined above where *^z*ij .o/ ^D ¹ stands for pi and pj being in the same subgroup and *^z*ij .o/ ^D ⁰ for pi and pj being in different subgroups, respectively.

Accordingly, an actors's total partial brokerage score for the kth intermediary can be defined as

$$Tk^* = w{1k}^* + w{0k}^* + b{01k}^* + b{10k}^* + b{0k}^* \forall k$$ (5.52)

being again a sum of five individual partial brokerage scores for the kth intermediary

$$w{1k}^* = \sum{i=1|i \neq k}^{N} \sum_{j=1|j \neq i \neq k}^{N} \frac{w1(ij)}{p{ij}(pk)} o(ij)$$ with $o(ij) = 1$ if $z{ij}(o) = z_{ik}(o) = 1$ and $w_1(ij) = 1$ if a set of relations is between $p_i$ , $p_k$ and $p_j$ , and else $0$

$$w{0k}^* = \sum{i=1|i\neq k}^{N} \sum_{j=1|j\neq i\neq k}^{N} \frac{w0(ij)}{p{ij}(pk)} o(ij) \quad \text{but } z{ik}(e)$$ \nif a set constant and $p_i$ .

with $$o(ij) = 1$$ , if $z{ij}(o) = 1$
but $z{ik}(o) = z_{kj}(o) = 0$ and $w_0(ij) = 1$ ,
if a set of relations is between $p_i$ , $p_k$
and $p_j$ , and else 0

$$b{01k}^* = \sum{i=1|i\neq k}^{N} \sum{j=1|j\neq i\neq k}^{N} \frac{b{01}(ij)}{p_{ij}(p_k)} o(ij)$$

with $$o(ij) = 1$$ , if $z{kj}(o) = 1$
but $z{ij}(o) = z{ik}(o) = 0$ and $b{01}(ij) = 1$ ,
if a set of relations is between $p_i$ , $p_k$
and $p_j$ , and else 0

$$b{10k}^* = \sum{i=1|i\neq k}^{N} \sum{j=1|j\neq i\neq k}^{N} \frac{b{10}(ij)}{p_{ij}(p_k)} o(ij)$$

with $$o(ij) = 1$$ , if $z{ik}(o) = 1$
but $z{ij}(o) = z{kj}(o) = 0$ and $b{10}(ij) = 1$ ,
if a set of relations is between $p_i$ , $p_k$
and $p_j$ , and else 0

$$b{0k}^* = \sum{i=1|i\neq k}^{N} \sum_{j=1|j\neq i\neq k}^{N} \frac{b0(ij)}{p{ij}(p_k)} o(ij)$$

with $$o(ij) = 1$$ , if $z{ij}(o) = z{ik}(o) = z_{kj}(o) = 0$ and $b_0(ij) = 1$ , if a set of relations is between $p_i$ , $p_k$ and $p_j$ , and else 0

(5.53)

for all ^k and pij .pk/ ¤ ⁰ corresponding to each different type of brokerage defined above, where pij .pk/ is the number of two-step paths with pk between pi and pj , *^z*ij .o/ ^D ¹ again stands for pi and pj being in the same subgroup and *^z*ij .o/ ^D ⁰ for pi and pj being in different subgroups.⁷²

⁷²This mathematical expression of the kth intermediary's raw brokerage scores is adapted from Gould (1989), which is a refinement of the original formulation by Gould and Fernandez (1989). Further, it is worth to mention, that *w*1k and *w*0k are not expressed in the most parsimonious way, but this is done to show a consistent set of equations.

To show that these formal expressions make sense, let us look a little bit closer at b 01k as one out of five individual partial brokerage scores of the kth intermediary. There, three actors, namely a first-party (denoted as pi), a second-party (denoted as pj ) and the intermediary (denoted as pk) form a triad. The brokerage role of the kth intermediary is a between-group brokerage of the gatekeeper-type (indicated by b standing for between-group brokerage) with the intermediary and the second-party belonging to the same subgroup (*z*kj .o/ ^D ¹, indicated by the subscript ¹), whereas the first-party is a member in another distinct subgroup (*z*ij .o/ ^D *^z*ik.o/ ^D ⁰, indicated by the subscript 0).

Next, it is easy to see that each of these five measures can be split up in two different subgroups: (1) *within-group brokerage scores w*I k and *w*Ok (or w I k and w Ok) on one hand and (2) between-group brokerage scores bOI k, bIOk and bOk (or b OI k, b IOk and b Ok) on the other which can be summarized by

$$wk = w{1k} + w_{0k} \text{ or } wk^* = w{1k}^* + w_{0k}^* \ \forall \ k$$ (5.54)

and

$$bk = b{01k} + b{10k} + b{0k} \text{ or } bk^* = b{01k}^* + b{10k}^* + b{0k}^* \ \forall \ k$$ (5.55)

for each actor with respect to both his brokerage roles and his subgroup memberships. In addition to this, each individual raw brokerage score can be summed up to get global raw brokerage scores as well as global partial brokerage scores on the complete network level:

$$W{1} = \sum{k=1}^{N} w{1k} \qquad W{1}^{} = \sum{k=1}^{N} w{1k}^{}$$

$$W{0} = \sum{k=1}^{N} w{0k} \qquad W{0}^{} = \sum{k=1}^{N} w{0k}^{}$$

$$B{01} = \sum{k=1}^{N} b{01/k} \qquad B{01}^{} = \sum{k=1}^{N} b{01k}^{}$$

$$B{10} = \sum{k=1}^{N} b{10k} \qquad B{10}^{} = \sum{k=1}^{N} b{10k}^{}$$

$$B{0} = \sum{k=1}^{N} b{0k} \qquad B{0}^{} = \sum{k=1}^{N} b{0k}^{}$$

$$W = W{1} + W{0} \qquad W^{} = W_{1}^{} + W_{0}^{*}$$

$$B = B{01} + B{10} + B{0} \quad B^{*} = B{01}^{} + B_{10}^{} + B_{0}^{*}.$$

$$(5.56)$$

Unfortunately, all these measures of brokerage roles depend heavily on network size, type of brokerage and size of subgroups under observation, but Gould and Fernandez (1989, pp. 102–115) developed a test statistic which is assumed to follow sufficiently a standard normal distribution for large networks of about ^N D ¹⁵ for global scores and ^N D ³⁰ for individual scores.

5.4 Summary and Concluding Remarks

In this part of the book, a closer look on contemporary social network analysis was offered. After reviewing shortly the various streams of research that led to contemporary social network analysis, the level of analysis and the graph-theoretic expression of measures to assess social network structures were presented with a focus laid on (1) actors and their personal network around them (micro- or ego-centric perspective) as well as (2) network structures as a whole (macro- or socio-centric perspective). For both levels of analysis, several distinct measurement concepts were outlined and discussed.

All in all, this wide field of contemporary social network analysis can look back to a tremendous development since initial treatments by Simmel (1908) with a highly accelerating speed after the mid 1970s, as e.g. Borgatti and Foster (2003) showed by counting publications indexed by Sociological Abstracts containing 'social network' in the abstract or title. What is left to do is to indicate, how these measures can be best applied to empirical data of social network structures. This seems to be a rather neglected field of research only with a few notes e.g. about applicability of centrality measures by Borgatti (1995, 2005), a discussion of personal networks by McCarty and Wutich (2005) or measurement issues of social capital by Borgatti et al. (1998) and Taube (2002, pp. 134–201, 2004) which will be ¨ outlined now.

First, Borgatti (1995, 2005) attempted to interpret centrality measures by their underlying network flow in order to show which measure is fully applicable to which specific network flow. Doing this, Borgatti (2005) distinguished between two dimensions of variation, namely (1) the kinds of trajectories that traffic may follow (geodesics, paths, trails or walks) and (2) the method of spread (serial or parallel duplication and transfer) which collapses to a matrix in the context of this work as shown in Table 5.5. ⁷³ There, four representative network flow processes are mentioned:

• Betweenness- as well as closeness-based measures of centrality by Freeman (1979) imply a package delivery process with fixed destination or target. Whereas closeness centrality can be interpreted as an index of expected time-until-arrival of something flowing through a network structure, betweenness centrality stands for frequency of arrival: It provides expected values for how often something

⁷³The category of serial duplication refers to network flows of miotic reproduction, viral infection or gossip and is therefore not of interest in the context of this work.

	Parallel duplication (Broadcast)	Transfer (Movement)
Geodesics	–	Package Delivery: Betweenness by Freeman (1979), Closeness by Freeman (1979)
Paths or trails	Information diffusion: Closeness by Freeman (1979), Degree by Freeman (1979)	–
Walks	Attitude influence: Closeness by Freeman (1979), Bonacich (1972a, 1972b, 1987) IEC by Friedkin (1991), Katz (1953), Hubbel (1965), Taylor (1969), Stephenson and Zelen (1989)	Money exchange: Newman (2005), Noh and Rieger (2004)

Table 5.5 Network flow typology of centrality measures (Source: adapted from Borgatti (2005))

passes a given point (Borgatti, 1995). The package itself can be regarded as a solid, indivisible object that can only be once in one place at a time like a parcel or documentary paperwork, which moves (hopefully) on the shortest possible path through a network.

As the image of the money exchange process is a solid, indivisible object like a banknote following a random walk process, the refinements of random-walk closeness (Noh and Rieger, 2004) or random-walk betweenness (Newman, 2005) fit best to this kind of network flow.
In an information diffusion process, a message spreads by replication (serial as well as parallel duplication) rather than by transference and normally does not pass the same link twice but can pass the same point multiple times.74Therefore, degree centrality as well as closeness centrality by Freeman (1979) are recommenced with degree centrality as a measure of immediate influence or the probability of receiving something that is randomly distributed in a network structure which is entirely a function of the number of relations that a given actor has.
If parallel duplication is assumed to take place via unrestricted walks rather than being constrained by trails, paths or geodesics, it can be regarded as an attitude influence process: The idea is that even if an actor influences just one other actor, who influences many others who themselves influence still more others (and so on), then the first actor is highly influential. Again closeness centrality by Freeman (1979), together with its refinements by Stephenson and Zelen (1989) or Friedkin (1991), is proposed. Furthermore, rank-based centrality measures by Bonacich (1972a,b, 1987) as well as other measures of prestige by Katz (1953), Hubbel (1965) or Taylor (1969) based on eigenvector constructs can be employed, which catch the essence of this influence process.

⁷⁴More definite, trails are paths, where a point may be reached more than once in a sequence of points but not adjacent ones, see e.g. Borgatti (1995).

Table 5.6 Measures for social capital (Source: adapted from Borgatti et al. (1998))

Measure	Graph-theoretic description	Relation to social capital
NSego	The number of alters that an ego is directly connected to	Positive: The more people you have relationships with, the greater the chance that one of them has the resource you need
ND	The proportion of pairs of alters that are connected	Negative: If all your alters are tied to each other, they are redundant. Given limits on relational energy, need to put eggs in more than one basket
NSk	The number of alters, weighted by strength of tie that an ego is directly connected to	Positive: The more different regions of the network an actor has ties with, the greater the potential information and control benefits
NCk	The extent to which all of ego's relational investments directly or indirectly involve a single alter	Negative: The more constrained the actor, the fewer opportunities for action
CB(pk)	The number of times that ego falls along the shortest path between two other actors	Positive: Actors with high betweenness link together actors who are otherwise unconnected, creating opportunities for exploitation of information and control benefits
CC(pk)	The total graph theoretic distance from ego to all others in network	Negative: The greater the distance to other nodes, the less the chance of receiving information in a timely way
CR(pk)	The extent to which ego is connected to nodes who are themselves high in eigenvector centrality	Positive: An actor has high eigenvector scores when they are connected to well-connected others

Second, McCarty and Wutich (2005) discussed arguments for including or excluding an ego from structural analyses of personal networks. For them, there are three possibilities present: (1) leaving ego out, (2) including ego with forced ties to all alters and (3) including but allowing null ties between ego and some subset of alters. Which one is right, depends again on the type of network flow under observation. As in the first option ego is excluded, his role in such a network structure can be regarded as a totally passive one – having no control over network flow and therefore being just a receiver of information and/or resources. In contrary to this, if ego is connected to all alters, he (or she) will dominate the network structure and therefore this approach should only be followed if his (or her) overwhelming influence is a valid question. As (1) and (2) imply special cases of network flow, the third option seems to be the most appropriate approach to personal network structure as it allows ego to have no direct relationship with some alters.

The third topic is about approaches to measure social capital. Unlike the relational and structural aspects of embeddedness which are directly covered by various measures from this section, the identification and measurement of social capital is not as clear. However, Borgatti et al. (1998) suggested several measures already presented that might be used to formalize the notion of social capital.75 Assuming social capital to be a network structure effect as the value of an individual's social relationships (Burt, 1992), they proposed NSego and ND to describe ego's personal network, NS^k and NC^k to cope with the impact of structural holes and CB .pk/, CC .pk/ as well as CR.pk/ to explore ego's network position as shown in Table 5.6, together with their relationship to social capital according to conventional wisdom. Furthermore, Taube (2002, pp. 134–201, 2004) showed, how social capital can be ¨ measured employing the brokerage roles concept by Gould (1989) or Gould and Fernandez (1989). Dividing the characteristics of social capital in (1) support capital of strong ties for which locals are particularly important and (2) leverage capital of weak ties as well as structural holes which is the domain of cosmopolitants, he asserted locals to be of a coordinator or representative type with cosmopolitans to serve liaison, gatekeeper and itinerant brokerage roles. By making use of Hummel and Sodeur (1987) census of triadic role patterns, he demonstrated how these brokerage roles can help to identify locals (coordinators or representatives) and cosmopolitans (itinerants or gatekeepers) in social network structures. Alternatively, CC .pk/ is used for the general distinction between brokers and non-brokers on one hand and for the differentiation between locals and cosmopolitans on the other with a rank ordering of CC .locals/>CC .cosmopolitans/>CC .non-broker/. Furthermore, CB .pk/ allows to distinguish between itinerants and gatekeeper types of cosmopolitans as CB .itinerant/>CB .gatekeeper/>CB .coordinator/ > CB .liaison and others/.

⁷⁵More precisely, they proposed measures for three different conceptions of social capital of which only one is of interest in this context.

Chapter 6 Freight Forwarder's Personal Network in Multimodal Transport Chains: An Empirical Investigation

So far in this piece of work, the topic of freight forwarding in multimodal transport chains with a leg by sea or air was described and theoretically reflected from different perspectives. After reviewing the forwarding business from a legal and functional point of view, contemporary multimodal chains with a leg by sea or air were described based on desk research in Part 2 as a status quo for further theoretical treatments. In Chapter 3 freight forwarder's intermediary role as a strange hybrid providing all kinds of transport and logistics services as well as financial intermediation on behalf and account of his clients was clarified well grounded in NIE, MMT and ITF. Seeing that such multimodal transport chains cannot be considered as a conglomerate of isolated bilateral or trilateral relationships, a network perspective of freight forwarding in multimodal transport chains with a leg by sea or air was developed in Chapter 4. There, the social network approach with SNA as a measurement concept showed the highest usability in the context of this work and therefore, relevant measurement concepts of e.g. network size and density, centrality or brokerage capacity and their connection to the notion of social capital were worked out in Chapter 5. Now it is time to show, whether actual business practice supports the assertions made and conclusions drawn before. This is done by analyzing empirical social network structure data from freight forwarders engaged in multimodal transport chains including a leg by sea or air in Germany and Austria.

6.1 Survey Design

In the following, a generic multimodal transport chain model is derived from insights generated during the predecing research process of finding a suitable theoretical framework in Chapters 3 and 4 of this work. Then, levels of analysis are determined and appropriate analytical instruments are discussed to unveil the social network structure as well as the amount of social capital employed. Furthermore, as freight forwarding in such multimodal transport chains takes place in an overall complex international to global environment, environmental control parameters are introduced which help to approximate the heterogeneity of the international to global environment where these freight forwarders are operating.

6.1.1 A Generic Multimodal Transport Chain Model

Referring to the inherent network structure of multimodal transport chains including a leg by sea and air, some efforts can already be remarked:

In Sect. 2.2.1, three distinct general approaches to analyze transport chains according to Beplat (1970:17–25) were described (namely functional, institutional and a combination of both).
In Sect. 2.2.4, the inter-organizational interaction and cargo movement at sea and airports were shown, following the approach of Martin and Thomas (2001).
In Sect. 4.4.2, a stylized model of hierarchical network structures in multimodal transport chains based on a system of schematic diagrams by Casson (1993, 1997a, 1997c) was established.

Now, for the following empirical investigation, a generic multimodal transport chain model is needed which allows to assess the social network structure around a freight forwarder. It follows the combined approach by Beplat (1970:17–25)

Fig. 6.1 Freight forwarder's personal network in export operations

Fig. 6.2 Freight forwarder's personal network in import operations

as depicted in Figs. 6.1 and 6.2. Here again, nodes are standing for persons or institutions being part of a multimodal transport chain which were connected by two types of lines: (1) big arrows as directed transport links indicating physical movement of shipments by some carrier from one place to another without further determining the mode of transport employed and (2) bidirectional lines representing all kinds of business relationships apart from simple exchange of cargo and their accompanying documentary paperwork. At the bottom, the subsequent steps of physical goods movement in a multimodal transport chain originating from a shipper (exporter) to a consignee (importer) are shown with carriers and transhipment terminals in between. Above this, the freight forwarder at home as "The Architect of Transport" is considered to act upon request as a full service intermediary between his principal (usually the shipper or consignor) and the other participants involved in actual goods movement, organization of shipment flow and/or handling documentary paperwork needed.

Being quite puzzling in first instance, this stylized model framework of a multimodal transport chain as shown in Figs. 6.1 and 6.2 can be used for both multimodal transport chains under consideration in this contribution, because apart from different naming of the participants, unit loads used, mode specific legal regimes and special business usances (see Table 6.1), both show principally the same social network structure. This is especially true, if we compare full-container-loads (FCL) and air freight consignments as multimodal door-to-door shipments from a shipper towards a consignee without break-bulk cargo or stuffing and stripping seagoing containers or unit load devices in transit.

Of course, much more other possible multimodal transport chain participants can be taken into consideration, but these depicted here seem to be the most crucial

Stylized model framework	SEA	AIR
Transport mode on the main haul	Main haul by sea	Main haul by air
Carrier on the main haul	Shipping company	Airline
organizational forms on pre and onward carriage legs	Carrier haulage or merchant haulage	Airline trucking, shipper or agent trucking
Points of transhipment	Container terminal at seaport	Cargo terminal at airport
Main actor at transhipment	Terminal operator	Ground handler
Types of transport brokers	Loading broker, discharging broker or non-vessel operator	Air freight broker or non-vessel operator
Types of transport agents	Shipping or liner agent	IATA sales agent
Unit loads used	Seagoing containers	Unit load devices
Legal regime	Hague, Visby or Hamburg Rules	Warsaw or Montreal Convention
Accompanying transport document	Bill of Lading (B/L) or sea waybill (SWB)	Air waybill (AWB)
Applicable trade terms according to INCOTERMS 2000	EXW, FCA, FAS, FOB, CFR, CIF, CIP, CPT, DES, DEQ, DAF, DDU, DDP	EXW, FCA, CPT, CIP, DAF, DDU, DDP
Applicable trade terms according to INCOTERMS 2010	EXW, FCA, FAS, FOB, CFR, CIF, CIP, CPT, DAP, DAT, DDU, DDP	EXW, FCA, CPT, CIP, DAP, DAT, DDP

Table 6.1 Comparison of multimodal transport chains with a leg by sea or air

ones, as shown by Cisic et al. (2000a, 2000b), Pocuca et al. (2000), Kylaheiko et al. (2000), Neiberger (2007, 2008) or UN/CEFACT (2000)¹

By analyzing goods and documentary paper flows in a sea transport chain well documented in Cisic et al. (2000a, 2000b), Pocuca et al. (2000) and Kylaheiko et al. (2000) a similar network structure evolved with buyer, seller, shipper, forwarder, shipping agent, customs broker, shipping company and cargo handling company being the most active and central participants therein.
In a survey about freight forwarding companies engaged in air cargo operations at Frankfurt-on-Main and Schiphol, Neiberger (2007, 2008) ended in an analogue network structure of goods and communication flows with the air freight forwarder as the central figure in between the exporter, consignee, overland transportation carriers for pre- and onward carriage, warehousing service providers for temporary storage, ground handling agents and airline actual transporting cargo on the main haul.

¹Notably, Stefansson (2004, 2006) developed a collaborative logistics management model to describe logistics outsourcing setups likely in the same way, where three groups of third-party service providers (namely carriers, logistics services providers and logistics service intermediaries) offer their services to the clients denoted by shipper and receiver. But apart from analyzing material and information links between actors, he conducted a series of case studies and focused on the interfaces of these links and their attributes.

• Last but not least, the Business Process Analysis Working Group (BPAWG) of the United Nations Centre for the Facilitation of Procedures and Practices for Administration, Commerce and Transport (UN/CEFACT) worked out their BPAWG Model for international supply chains based on UN/CEFACT Unified Modelling Methodology.² It covers "all stages following the recognition of need by a customer for a product or service to the fulfilment of an order by a supplier and the resulting financial settlement [and] it incorporates any necessary activities carried out by intermediaries and authorities." (UN/CEFACT 2000). While modelling about 31 different actors (and defining their roles) potentially engaged in such international supply chains between buyer and seller and their goods movements or data flows in between, the BPAWG Model shows again the freight forwarder as one of the key intermediaries in international goods movements.³ Other actors engaged in goods movements are carriers (and their agents) on the main haul by road, rail, sea, air or postal service, transhipment points like sea or airports (denoted by receiving authorities in export and import) as well as customs and health authorities in the country of the buyer and seller.4

A quite natural approach of modelling such network structures – apart from just visualizing them in form of a graph – is to quantify these points and their connections in order to describe them mathematically, which will be assessed in the next sections.

6.1.2 Levels of Analysis and Analytical Instruments

As the business relationships between the actors on this generic multimodal transport chains model as shown in Figs. 6.1 and 6.2 sum up to an undirected graph, several analytical instruments from contemporary social network analysis on the actor level as well as the network level as outlined in Sect. 5.3 can be employed, which will be described in the next two sections. After that, four environmental

²This is a further development of the former UN/CEFACT International Trade Transaction Model (ITT), see UN/CEFACT (2000). Further, for more about the underlying UN/CEFACT Unified Modelling Methodology see http://www.untmg.org/specifications/.

³According to UN/CEFACT (2000), a freight forwarder is "[an] intermediary employed by buyer or seller (depending on the terms of trade [sic!]) who may carry out a variety of tasks concerned with the movement of goods. These can include collection and transport of goods and the completion of an export declaration on the exporter's behalf."

⁴According to UN/CEFACT (2000), a receiving authority in export is "[a] body (normally a Port or Airport Authority) who has responsibility for receiving the goods, for their safe storage and the loading of the goods onto the transport," whereas in import it is "[a] body within the maritime regime (normally the Port Authority) who has responsibility for the safe unloading of the goods from the vessel." Designating them as "receiving authorities" may point to the fact, that sea- or airports can be regarded as private intermediaries and public authorities at the same time, see e.g. Martin and Thomas (2001) or Sect. 2.2.4.2.

control parameters are derived to assess possible impacts of different countries of destination or departure on the sample.

6.1.2.1 Personal Network Structure and Position

Whereas the unit of analysis in contemporary social network analysis can vary between micro, meso and macro perspective, analytical instruments can be divided in a relational and a positional approach (see Tables 5.1 and 5.2). Following this taxonomy, ego's personal network structure of a freight forwarder can be described by measures of network size and density (NSego and ND) as well as by measures of prominence (CD.pk/, CB .pk/, CC .pk/ and CR.pk/) that give information about his position therein. Furthermore, Table 6.1 offers a network flow typology of measures which help to interpret values of CD.pk/, CB.pk/, CC .pk/ and CR.pk/ in a transport chain context. Moreover, most of these measures can be at the same time easily interpreted as an expression of the underlying social capital as already shown in Table 6.2. This offers a way to develop testable propositions about social capital issues in the context of freight forwarding in multimodal transport chains with a leg by sea and air as follows:

As NSego is an expression of the number of alters (transport chain members) that an ego (freight forwarder) is directly connected to, high values of NSego give an indication that he possesses a high amount of social capital as he is well embedded in the underlying social network structure.
Conversely, as ND stands for the proportion of alters (transport chain members) connected to each other, high values of ND indicate that freight forwarders as egos in social network structures possess a lot of redundant contacts or stem simply from the fact that he might not be the central actor in the transport chain at all. Both assertions lead to consider ego to have a low amount of social capital if ND shows high values.
Being defined as a function of the number of direct relations ego has, a high CD.pk/ indicates strong immediate influence on alters directly connected to him. If we think of an information diffusion process, high CD.pk/ means that ego serves as a well connected hub and direct source of information for his direct neighbourhood which indicates again a high amount of social capital for him.
CR.pk/ then goes a step further as it is defined as the extent to which ego is connected to others who are themselves well connected. This can be interpreted as an influence process, where ego exerts influence not only on alters directly connected to him but on all other transport chain members, too. Therefore, it can be asserted that a high amount of social capital is present if CR.pk/ is high as ego is able to influence other members in a transport chain to a high extent.
CB .pk/ is the number of times ego falls along the shortest path between two alters. If the network flow is regarded to be a package delivery process or documentary paper flow moving along the transport chain, a high CB .pk/ indicates a high amount of social capital as ego serves a highly frequented intermediary

6.1 Survey Design 257

position connecting transport chain members. Following Granovetter's (1973, 1982) and Burt's (1992, 1997, 1998) weak tie argument, this coincides with the structural view of social capital where high social capital comes along with ability of ego to bridge and exploit structural holes in a social network structure (see Sect. 4.6.1.2).

• Last but not least, CC .pk/ can be interpreted in different ways. When looking on the expected time until arrival of something in documentary paper flows, information diffusion or parcel delivery processes, CC .pk/ increases in the total graph theoretic distance from ego to all others in the network. Therefore, a high (low) CC .pk/ indicates low (high) amount of social capital which is fully in line with Coleman's (1988, 1990) and Krackhardt's (1992) strong tie argument of relational social capital which is especially high in a tight, cohesive group with extensive personal bonds among its members (see Sect. 4.6.1.2).

To sum up these propositions, high values of NSego, CD.pk/, CR.pk/ and CB .pk/ as well as low values of ND and CC .pk/ indicate a high amount of social capital for ego which is the freight forwarder in a multimodal transport chain in the context of this work. The same is valid for C⁰ ^D.pk/, C⁰ ^B .pk/ and C⁰ ^C .pk/, too, as these prominence measures are just normalized versions of CD.pk/, CB .pk/ and CC .pk/ with values varying between 0 and 1.

6.1.2.2 Intermediary Roles

Regarding intermediary roles of a freight forwarder, the brokerage roles concept by Gould (1989) or Gould and Fernandez (1989) can be employed to identify different intermediary roles of (1) being a local broker (or coordinator) linking two parties within one organization, (2) being an intinerant broker bringing two parties of one organization together as an outside third-party intermediary, (3) serving as a gatekeeper for the second-party vis-a-vis* the first-party, (4) being a representative for the first-party *vis-a-vis the second-party or (5) being an intermediary in a liaison relationship triad, where all parties come from different organizations. To clarify the usefulness of this classification, one may think of an indirect conclusion of affreightment as shown in Fig. 2.1: There the international freight forwarder is an intermediary (third-party) between a shipper (first-party) and a carrier (secondparty) actually performing the resulting goods transport. If all three parties (shipper, freight forwarder and carrier) belong to different organizations, the brokerage role of the freight forwarder is of the liaison type. However, if the freight forwarder conducts the resulting goods transport as a performing carrier, too, he is then a gatekeeper as third-party and second-party belong to same organization and so on.

Furthermore, Taube (2002:134–201, 2004) showed how social capital can be ¨ measured by dividing the characteristics of social capital in (1) support capital of strong ties and (2) leverage capital of weak ties as well as structural holes (see Sect. 5.4). In the case of support capital, locals of the coordinator or representative type are important, whereas cosmopolitans of the liaison, gatekeeper or itinerant type stand for leverage capital. The distinction between these two forms of social capital within a given social network structure can be made via CC .pk/ with a rank ordering of CC .locals/>CC .cosmopolitans/>CC .non-broker/. Moreover, CB .pk/ allows to distinguish between itinerants and gatekeeper types of cosmopolitans as CB .itinerant/>CB .gatekeeper/>CB .coordinator/>CB .liaison and others/:

Keeping this in mind, the differences between freight forwarding and integrator activities can directly be compared. The integrator stands for a fully integrated transport chain of door-to-door shipment services as usually offered by integrators like DHL, FedEx, UPS or TNT. Investing heavily in own transport capacity as well as transhipment hub facilities, they mainly conduct all tasks necessary to move their client's shipments on own account while minimizing the employment of other independently operating transport chain members. Ideally, the resulting network structure is then concentrated in one central intermediary standing between the shipper and the consignee which does everything on own account and therefore does not employ any other independently operating transport chain members. This leads in the context of this work to C⁰ ^D.pk/ D ^C⁰ ^B.pk/ D ^C⁰ ^C .pk/ ^D ¹ and NDalter ^D ⁰ for both export and import operations.

When looking now on dominant intermediary roles in the context of this work, the integrator should be a gatekeeper in export operations and a representative in import operations, whereas freight forwarders should be mostly intermediaries of liaison type if they only organize multimodal transport operations but not act on own account. Therefore, C⁰ ^C .integrator/>C⁰ ^C .freight forwarder/ and C⁰ ^B .integrator/ > C⁰ ^B .freight forwarder/ should be the case if we normalize CC .pk/ and CB .pk/ to adjust for differences in network size as the social network data under observation are drawn from different samples.

6.1.2.3 Environmental Control Parameters

Taking modern information processing technologies like the Internet into consideration, there is always seen a 'death of distance' for communication and information transmission. But regarding distribution of physical goods as the main purpose of multimodal transport chains, geographic distance still matters. Moreover, distance has also cultural, administrative (or political) and economic dimensions, as Ghemawat (2001) proposed. To operationalize this, he put them in a conceptual framework called CAGE in order to identify and access the impact of distance on various industries along these dimensions.⁵ In the following, these different concepts of distance are further developed with administrative (or political) and economic distance put together as one complex. The aim of this task is to construct environmental control parameters, which quantify possible impacts on network

⁵CAGE is simply an abbreviation for cultural, administrative, geographical and economic distances.

structure stemming from different countries of destination (departure) in export (import) operations in the sample.

Geographic Distance

Looking at macro-economic trade surveys, increasing geographic distance always lead to lower level of exchange of goods within and across countries.⁶ According Ghemawat (2001), geographic distance affects the costs of transport as well as communication and it is of particular importance for companies that deal with heavy or bulky goods or whose operations require a high degree of coordination between highly dispersed people and/or activities.⁷ Furthermore, physical attributes of goods like perishability or special handling requirements add complexity to cross-border transport chains.⁸ Moreover, geographic distance is not simply the transport distance between a source and a sink denoted in miles or kilometres, because the physical size of a country, average within-country distances to borders, access to waterways and the ocean, topography of the landscape and even the level of transportation and communication infrastructure might have to be taken into account.9 Another aspect is capital bound of goods in transit, which is more and more an important decision factor in transport mode selection.¹⁰ Therefore, transport time in days or even hours instead of transport distance seems to be an alternative measure for geographic distance.

For the purpose of the survey, both measures of geographic distance were taken into account and so two Geographic Distance Indexes, namely GDhf standing for transport distance and GD⁰ hf for transport time with h the home country, f the country abroad and (certainly) ^h ¤ ^f were raised. First, transport distances on the main haul between country, sea- or airport of departure and destination given in the questionnaire were calculated, employing data sources freely available on the Internet.11 Due to sea or airports abroad not directly asked for in the questionnaire, the most important ones in terms of cargo throughput (denoted in TEUs for sea or tons for air cargo shipments) for the country of departure or destination were used

⁶See Frankel and Romer (1999), Rose (2000), Irwin and Tervio (2000), Frankel and Rose (2000, ¨ 2002) only to name the most recent ones.

⁷Cf. Ghematwat (2001:139–140).

⁸See Kummer et al. (2010:34–41) with further references.

⁹Cf. Ghemawat (2001:145). Following Rose (2000) or Frankel and Rose (2000, 2002), common land borders raise bilateral trade by about 80% and landlocked countries have about 40% less trade in comparison to other countries.

¹⁰See Kummer et al. (2010:41–43) with further references.

¹¹The sources for distance measures were databases at http://www.distances.com for sea and http://www.webflyer.com for air transport operations. These distances, originally provided in nautical miles (1 nm = 1.853167447 km) or statute miles (1 sm = 1.612377598 km), were transformed into kilometers as a common measurement unit.

as a proxy.¹² To calculate transport time on the main haul, all transport distances were multiplied by a fixed factor depending on the mode of transport. Doing this, an average speed of 20 kn for sea and 500 mph for air cargo transport operations was assumed, leading to quite reasonable average transport times on the main haul.13 Of course, actual transport time of a shipment by sea or air is heavily influenced by routing decisions, frequency of service, ports of call or transhipment points on the way, time needed at customs for release of cargo as well as distances and mode of transport on the pre- and onward-carriage legs, but assessing all these factors is beyond the scope of this work.¹⁴ To conclude this section, it can be assumed that with increasing GDhf or GD⁰ hf , geographic distance between two countries is getting bigger leading to diminishing cross-border trades and/or add complexity to transport operations.

Cultural Distance

To Ghemawat (2001:140–141), cultural distance between two countries determines heavily how people interact with each other and with companies and institutions. Some straight forward examples to this are common ethnic heritage, language, social norms and/or religious beliefs. At least three different streams of contemporary cultural research activity in the field of economics and business can be distinguished: (1) a rather qualitative, descriptive anthropologist-like approach, (2) a merely subjective psychologist-like approach centering around individual perceptions on perceived cultural differences and (3) a more objective sociologistlike approach focusing on people's values from different cultures.

¹²This seems to be a viable approach, because in each country (with the United States as one notable exception) being observed, there is always one most important sea- or airport for international shipments, serving a hub function. In the case of the United States, there are of course several sea- or airports possible, depending on the original source or sink of a shipment observed. Having no better information available, the Port of New York – New Jersey and the John F. Kennedy International Airport were chosen as the most important sea- and airports on the east coast range, which lead at least to some underestimation of geographic distance for the United States.

¹³Strictly speaking, transport distances by sea were divided by 20 and air by 100, taking into account that an air cargo shipment stays on average about 80% of total transport time at airport premises (see e.g. Reinheimer and Bodendorf (1999) or Vahrenkamp (2002:17–18)). In the first instance, this seems to be a very crude approach, but the figures obtained by this calculations were checked against average travel times of container ships and non-stop flights operated by Lufthansa. 14Constructing a similar measure of geographic distance, Linnemann (1966) took the shortest navigable between the main ports of respective countries plus the overland distance form the ports to the economic centres of gravity for each pair of countries concerned. But he noted that "a somewhat subjective element has slipped into the analysis in determining the economic centre of gravity of the various countries, and the corresponding hinterland distance of the ports. General geographical knowledge and encyclopædia information resulted in the selection of ports and the estimates of hinterland [...]" (Linnemann 1966:71).

In the first approach, culture is regarded as being some external factor and people going abroad or getting into contact with people from other cultures have to adjust themselves. Having this picture of culture in mind, authors like Hall (1960, 1976), Ricks (1993) or Weiss (1994a, 1994b) provided a lot of anecdotes to their readers and gave hints for best practice in inter-cultural business situations but never quantified their findings as a starting point for further cross-cultural comparisons.

The second approach is centred around the concept of "psychic distance" which was heavily used by researchers of the Uppsala School around Jan Johanson, Jan-Erik Vahlne and Finn Wiedersheim-Paul describing internationalization processes of Swedish firms.¹⁵ Johanson and Vahlne (1977:24) defined it "...as the sum of factors preventing the flow of information from and to the market." Later on, O'Grady and Lane (1996:330) expanded this concept defining it as ..." a firm's degree of uncertainty about a foreign market resulting from cultural differences and other business difficulties that present barriers to learn about the market and operating there." But all these authors have in common that, using "psychic distance" merely as a metaphor and employing merely weak indicators, they failed to operationalize their constructs.¹⁶ Therefore, M¨uller and Koglmayr ¨ (1986) and M¨uller (1991) proposed a first dedicated measurement approach by constructing cognitive maps in order to visualize subjectively perceived distance between countries.¹⁷ But with their empirical research being restricted only to a sample of a few typical countries, this measure of "psychic distance" was not quite useful for the purpose of this work.

In the third approach, culture is regarded to be an endogenous factor, a "software of the mind"18, which can be metered by statistical evaluation of people's values coming from different countries. It is closely related to an extensive research agenda by Geert Hofstede, who participated in a series of empirical surveys at IBM (and validated them later on at the IMD Lausanne) in the 1960s and 1970s on work-related values.19 Based on his findings, Hofstede (1980) formulated four cultural dimensions, namely "power distance", "individualism", "masculinity" and "uncertainty avoidance", which were joined by a fifth dimension called "long term orientation", originally based on the Chinese Values Survey (see

¹⁵See Johanson and Wiedersheim-Paul (1975), Johanson and Vahlne (1977), Andersen (1993), O'Grady and Lane (1996) or Barkema et al. (1996) with further references. But following these authors, the term "psychic distance" was already used in prior research like Beckermann (1956) or Linnemann (1966) with the last calling it "economic horizon" (see Linnemann (1966:27–28)).

¹⁶Cf. M¨uller and Koglmayr (1986), St ¨ ottinger and Schlegelmilch (1998), Dow (2000), Child et al. ¨ (2002), M¨uller (2002) or M¨uller and Kornmeier (2002:522–523).

¹⁷See also Stottinger and Schlegelmilch (1998) or M¨ ¨ uller and Kornmeier (2002:542–552).

¹⁸See Hofstede (1991:4) referring to the subtitle of this book. Further, he distinguished between the above already described anthropologist conception and his view of culture as "the collective programming of the mind which distinguishes the members of one group or category of people from one another," see Hofstede (1991:5).

¹⁹See Hofstede (1980, 1991, 2001) for an exhaustible description of methodology and data sampling issues.

The Chinese Culture Connection (1987), Hofstede and Bond (1988) or Hofstede (1991:159–174)). Following Hofstede (2001:29), these five cultural dimensions can be described shortly as follows:

"1. Power distance [PDI], which is related to the different solutions to the basic problem of human inequality.
- 2. Uncertainty avoidance [UAI], which is related to the level of stress in society in face of an unknown future.
3. Individualism versus collectivism [IDV], which is related to the integration of individuals into primary groups.
4. Masculinity versus femininity [MAS], which is related to the division of emotional roles between men and women.
5. Long-term versus short-term orientation [LTO], which is related to the choice of focus for people's efforts: the future or the present."

Of course, Geert Hofstede is not the only one who conducted such inter-cultural research in this way²⁰ and there is an ongoing dispute about the validity of his approach to cultural differences.²¹ But his work seems to have the strongest impact on the development of cross-cultural studies in economics, sociology, psychology as well as other fields of social sciences.²²

According to Kogut and Singh (1988), deviations in the four cultural dimensions from Geert Hofstede adjusted for differences in variances can be averaged to a Cultural Distance Index CDhf with

$$CD{hf} = \frac{1}{4} \sum{c=1}^{4} \frac{(CI{ch} - CI{cf})^2}{CV_c} \,\forall \, h \neq f$$ (6.1)

where CIch and CIcf stand for values of Hofstede's first four cultural dimensions (as given in Hofstede (2001:500–502)) and C Vc for their overall variance in the sample indexed by h for the home country, f for each country abroad and c for one

²⁰Other large-scale value surveys of this kind are e.g. the Schwartz Value Survey (Schwartz (1992, 1994, 1999) or Schwartz and Sagiv (1995)), the World Values Surveys (Inglehart et al. (1999), Inglehart (2000) or Inglehart and Baker (2000)), the Trompenaars' Database (Trompenaars (1993) or Trompenaars and Hampden-Turner (2002)) and many other authors cited in Hofstede (2001:29–34).

²¹See Yeh and Lawrence (1995), Shenkar (2001), Søderberg and Holden (2002), McSweeney (2002a) and the following by Smith (2002), Hofstede (2002), McSweeney (2002b) and Williamson (2002) with further references – only to name a few.

²²See Søndergaard (1994) or Hofstede (2001:461–466, 2002) with an overview and some comments on contributions refering to Hofstede (1980). Actually, this first edition and Hofstede (1991) together were subject to many replications or extensions to other countries not part of the original sample as well as numerous cross-cultural studies matching his findings with other data, so that in the second revised edition, the author could list about 280 surveys with correlations to the five cultural dimensions (Hofstede (2001:503–520)), provide data for additional 16 countries (Hofstede (2001:502)) and give about 877 references of contributions for the time after 1980 (Hofstede (2001:525)).

of the four cultural dimensions under observation.23 In conformity with Euclidean distance as the square-root of the squared deviation of each country abroad with the home country, an alternative Cultural Distance Index CD⁰ hf with

$$CD'{hf} = \sum{c=1}^{4} \sqrt{\frac{(CI{ch} - CI{cf})^2}{CV_c}} \,\,\forall \, h \neq f$$ (6.2)

can be calculated.24 Although both distance measures rely on different concepts, they follow the same direction: A low score of CDhf or CD⁰ hf indicates a low cultural distance between two countries, which can be interpreted as cultural similarity, making things easier in cross-border trade as well as transport operations.

Administrative, Political and Economic Distance

As they seem to be closely related to each other, administrative or political differences and economic differentials between countries can be interpreted together as a third dimension of distance. Recently, Rose (2000) or Frankel and Rose (2000, 2002) explored empirically in a sample of over 200 countries or dependencies, that historical and/or political ties between countries greatly affect trade.25 Especially, friendly ties remaining from colonialism, preferential trading arrangements, common currency, or political unions boost cross-border activities between countries.²⁶ Conversely, tariff and non-tariff measures raised on cross-border traded goods as well as capital flows impose considerable barriers of trade, leading to less cross-border economic activity between countries.²⁷ Furthermore, weak legal infrastructure and political instability may depress cross-border trade and/or direct investment in a country because high risks shy away outside countries doing business there. On the opposite, countries can get very attractive for foreign companies, if they open their economy and facilitate cross-border economic activities. Another important aspect is corruption in its many facets, which seems to be

²³Although this construct is a rather simplistic aggregate based on Hofstede's cultural dimensions, it got wide-spread application. Further see Shenkar (2001) or M¨uller and Kornmeier (2002:529–530) for critical comments on Kogut and Singh (1988). Further, M¨uller (2002) or M¨uller and Kornmeier (2002:523-529) discussed other constructs of cultural distance.

²⁴Cf. Barkema and Vermeulen (1997).

²⁵See also Ghemawat (2001:138) citing Frankel and Rose (2000).

²⁶According to their estimates, former colony-colonizer relationships increase bilateral trade by 700%–900%, a common colonizer by 150%–190%, common regional trading blocks by 290%– 350%, common currency by 300%–390% and common polity by 260%–300% in comparison to other countries, see the tabulated values in Rose (2000:16), Ghematwat (2001:138) or Frankel and Rose (2000:28, 2002:441).

²⁷Cf. Ghemawat (2001:142-144).

more likely in countries with a weak legal infrastructure in connection with low income-levels.28

Referring explicitly to economic distance, wealth or income of consumers seems to be the most important economic attribute that creates distance between two countries affecting both direction and amount of trade.29 Rich countries, e.g. with a high income level denoted by Gross National Product (GNP) per capita, engage more in cross-border economic activity than poorer ones for many reasons.30 Following Krugman (1995:332–337), at least four new aspects of modern world trade can be identified, which have much in common with these wealth differentials. First, the exploitation of comparative advantages of cost and/or price differentials between countries as theoretically well reflected in mathematical models of the neoclassical trade theory create huge trade flows originating from poorer, newindustrializing countries. Second, there is an ongoing trend in manufacturing to slice up the value chain by producing goods in a number of stages in a number of locations, adding a little bit of value at each stage which leads in sum to a tremendous increase in cross-border trade. Third, a high proportion of world trade today is intra-industry and occurs between rich countries or trade blocks like the European Union, the North American Free Trade Area as well as Japan and South Korea, well known under the term "the triad".31 And last but not least, some small countries in the world like Hong Kong, Singapore, Malaysia, Belgium, Ireland or the Netherlands accumulate wealth as being "supertrading economies" with an overwhelming share of exports in comparison to their Gross Domestic Product (GDP).³²

One short-cut approach to quantify these various aspects of administrative, political and economic distance is the employment of already existing country indexes. Of course, such country indexes are by far exhaustible in describing the administrative, political and economic situation of a particular country, but they seem to be a good indicator for this complex of influences on cross-border trade

²⁸See e.g. Shleifer and Vishny (1993), Bardhan (1997), M¨uller (2001) or the annually published Global Corruption Report by Transparency International (http://www.transparency.org) for further discussion about different forms of corruption and their impact on economy.

²⁹Cf. Ghemawat (2001:145–147). Further, Beckermann (1956) e.g. surveyed intra-European trade in terms of economic distance with the mark-up between average FOB-value of exports of specific goods and their corresponding CIF-value in the country of destination as a proxy. Doing this, he came to the conclusion, that high mark-ups indicating greater distance lead to lower trade volumes and vice versa.

³⁰See Frankel and Romer (1999) or Irwin and Tervio (2000) for some recent empirical evidence. ¨

³¹See Kummer et al. (2010:71–72) for this contemporary interpretation of the term originally coined by Ohmae (1985, 1989).

³²According to Krugman (1995:334–335), their share of export exceeded GDP by more than 50% with Singapore having an export share of 174% of GDP, Hong Kong 144%, Malaysia 78%, Belgium 70%, Ireland 64% and the Netherlands 52% in 1990.

and/or investment. For the purpose of this study, four country indexes with figures as of 2003 were chosen:³³

1. The Index of Economic Freedom (IEF) from the Heritage Foundation provides an examination of ten factors (namely trade policy, fiscal burden of government, governmental interventions in the economy, monetary policy, capital flows and foreign investment, banking and finance, wages and prices, property rights, regulations and black market activity) based on 50 independent economic objectives contributing to economic freedom and prosperity in 161 countries.³⁴
2. The Economic Freedom of the World Index (EFW) from the Fraser Institute measures the degree of economic freedom in five major areas (size of government, legal structure and property rights, access to sound money, freedom to exchange with foreigners and regulation of credit, labour and business) for 123 countries, compiling 38 objectives from third-party statistical data.35
3. The Corruption Perception Index (CPI) from Transparency International is a composite index derived from surveys of business people and assessment by country analysts from 13 independent institutions, focusing on extensive perceptions of corruption within 133 countries under observation.36
4. The Globalization Index (GI) from A.T. Kearney and Foreign Policy Magazine, which makes use of 13 indicators spanning information technology, finance, trade, personal communication, politics and travel to determine a country ranking in terms of global integration among 62 countries.³⁷

To sum up these four economic risk indexes, basically the same procedure as in the case of Hofstede's cultural dimensions was applied to obtain an Economic Distance Index EDhf with

$$ED{hf} = \frac{1}{4} \sum{e=1}^{4} \frac{(EI{eh} - EI{ef})^2}{EV_e} \,\forall \, h \neq f$$ (6.3)

³³Of course, many other similar country indexes are provided through the World Competitive Yearbook of the IMD Lausanne (http://www.imd.ch/wcy), the Global Competitive Report of the World Economic Forum (http://www.weforum.org), the International Country Risk Guide Rating System of the Political Risk Service Group (http://www.icrgonline.com) or the Business Risk Service from the Business Environment Risk Intelligence (http://www.beri.com) or the Riskwire Service from the Economist Intelligence Unit (http://riskwire.eiu.com). But common to all these country indexes is, that they (1) are not publicly available, (2) cover less countries than needed and/or (3) are used as sources in the country indexes chosen.

³⁴Each of the ten factors is scored within a range of one to five, and these scores are averaged to obtain a final IEF score for each country with a value around one for the most freest countries and five for the most economically repressed ones, see Beach and O'Driscoll (2003), Eiras and Schaefer (2003) or http://www.heritage.org for further methodological issues.

³⁵See http://www.freetheworld.com/ for more information about the EFW.

³⁶See http://www.transparency.org/policy research/surveys indices/ for more information about the CPI.

³⁷See Foreign Policy (2003), http://www.foreignpolicy.com or http://atkearney.com for further information about sampling and methodological issues.

where EIeh and EIef stand for one of the four economic indexes as described above and EVe for their overall variance in the sample, again indexed by h for the home country, f for the country abroad and e for one of the four economic indexes under observation. Again, an alternative Economic Distance Index ED⁰ hf based on the Euclidean concept can be calculated:

$$ED'{hf} = \sum{e=1}^{4} \sqrt{\frac{(EI{eh} - EI{ef})^2}{EV_e}} \ \forall \ h \neq f.$$ (6.4)

Before applying these formulae, the rankings of IEF and GI were standardized into scores from one to ten and the scores of EFW and CPI had to be reversed in order to get a common direction of impact. Doing this, low (high) scores of EDhf or ED⁰ hf are standing for low (high) economic distance between two countries, indicating an economic environment supporting (depressing) cross-border trade and transport operations.³⁸

6.2 Sample Design and Preliminary Analysis

During 2003, several surveys among freight forwarding companies in Austria (about 600 full members of the freight forwaders'specialist association at the Austrian Chamber of Commerce called Wirtschaftskammer Osterreich (WKO), Fachverband ¨ Spedition und Logistik), Hamburg (342 members of the regional freight forwarders' association Verein Hamburger Spediteure e.V. (VHSp)), Bremen-Bremerhaven (165 members of the regional freight forwarders' association Verein Bremer Spediteure e.V. (VBSp)) were conducted to obtain social network data as shown in Table 6.2.

In the first and the second survey (FCL-A and FCL-HH), the already above mentioned freight forwarder associations distributed a cover letter and a fivepage, double column questionnaire from the author attached to their monthly email-newsletter directed to the head of each member firm. In the case of Bremen-Bremerhaven (FCL-HB), an announcement appeared in the VBSp monthly circular letter, followed by a cover letter and a questionnaire sent directly to the freight forwarding companies by mail. All freight forwarders were asked to fill in and send the questionnaire back by facsimile or mail – alternatively, they could fill in and submit the questionnaire online as it was hosted on a web-server at Dresden University of Technology, too. A reminder was send about three weeks later by

³⁸Notably, there were three interesting country indexes, namely (1) the Access Index by SRI International and FedEx (http://about.fedex.designcdt.com/our company/publications and reports) from 2006 and 2008, (2) the Logistics Performance Index (LPI) by the World Bank (http://www.worldbank.org/lpi) as of 2007 and 2010 and (3) the Enabling Trade Index (ETI) (http://www.weforum.org/en/initiatives/gcp/GlobalEnablingTradeReport/index.htm) annually published since 2008. But as the survey was already in 2003, they could not be employed, although they address directly main environmental issues in international transport chains.

Sample	Sample region	Main haul	Sampling method
FCL-A	Austria	By sea	With WKO-newsletter
FCL-HH	Hamburg (D)	By sea	With VHSp-newsletter
FCL-HB	Bremen (D)	By sea	Questionnaire by mail
FCL-D	FCL-HH and FCL-HB	By sea	Various, see above
SEA	All regions	By sea	Various, see above
AIR	Vienna (A)	By air	Structured interviews

Table 6.2 Sampling frame

facsimile, containing again the cover letter with a link to the web-server, where the original questionnaire and the corresponding online version were provided. Last but not least, a sample of 24 Austrian air freight forwarders (AIR) was raised by conducting structured interviews by telephone among 130 members of the Wirtschaftskammer Osterreich (WKO), Fachverband Spedition und Logistik ¨ considered to be engaged in air freight forwarding with a special focus on companies engaged at the Vienna International Airport.

In all above described surveys, the questionnaire was split into four sections:

In an introductory section, several company-specific questions were raised like fields of activity, regional focus, status of the company site and number of subsidiaries inland and abroad.
In the next section, the respondents were asked to think of a typical export shipment and describe this in the following. First, some key information about the shipment itself (e.g. point of exportation, country of destination, trade terms used, type of transport documentation and value-added services offered) were asked for, before the social network of their contacts in this case from their client's order (usually by the exporter or shipper) to the carrier on the main haul or even the freight forwarder receiving the shipment abroad was mapped along several dimensions like duration, intensity or mode of communication.
The third section was about a typical import shipment the respondents should think of and after a similar set of general questions about the shipment itself, the task was again to map the accompanying import operations from a client's order to their subsidiary in Germany or Austria to the point of destination analog to the preceding section.
In the final section, some personal information about the respondents was demanded: position served, number of years working in the company and in the forwarding business.

Unfortunately, the response rates of the first two surveys were very low with 1.0% for FCL-A, 2.0% for FCL-HH, so that the amount of useful observations obtained in comparison to the freight forwarders asked for seems to be negligible. Only the FCL-HB survey with a response rate 6.7% and the AIR survey with 18.5% were within the range of survey response rates of contributions in the Transportation Journal, Journal of Business Logistics and International Journal of Physical Distribution and Logistics Management as reported by Griffis et al. (2003),

Sample	No. of Resp.	No. in export	No. in import
FCL-A	6	4	4
FCL-HH	6	5	4
FCL-HB	11	9	10
FCL-D	18	14	14
SEA	23	18	18
AIR	24	23	22

Table 6.3 Survey respondents

Larson and Poist (2004) or Larson (2005).39 Testing for non-response bias between early and late respondents as proposed by Armstrong and Overton (1977) was rather inconclusive with the actual sample sizes being indeed very small and the within-sample variations reflecting merely heterogeneity of the respondents and their operations rather than indicating a systematic non-response bias.

All in all, a total of 48 responses could be sampled, with 24 for each main haul mode. Because of missing values in some questionnaires, not all of them could be included in the following analysis. At the end, there were 18 to 23 observations of transport chains including a leg by sea or air available for further assessment which can be further split in export and import operations (see Table 6.3). Some of several reasons for this poor outcome might be:

First, there might exist an overall resistance of Austrian and German freight forwarding companies to response to such surveys – some of them were so kind to communicate to the author that they are fed up with work and so they are not able to follow such a proposal.
Second, the research design requested the respondents to fill in a five-page, double column questionnaire in order to obtain a minimum of information about their social network structure which in turn might have been too long and boring.
Furthermore, the distribution of questionnaires by email-newsletters as a low-cost survey method for data collection for the first two samples (FCL-A and FCL-HH) turned out to be a bad alternative in comparison to more conventional sampling methods like mail survey approach with pre-notification in the FCL-HB survey and structured interviews by telephone in the case of the AIR survey.
Another viable reason for this poor outcome for the first three samples might be that in Austria and even at the two big German seaports, there are not so many freight forwarding companies engaged in door-to-door FCL-shipments. Moreover, it seems to be quite common for large-scale shippers, exporters and/or importers to contact directly the sales representatives of carriers on the main haul in order to save costs and they are at the same time willing to organize all tasks needed to end in a door-to-door shipment service.

³⁹Kaltenbrunner (2005:59) reported that only 46 air cargo freight forwarding companies are active at the Austrian market and even four of them are double countings as they are wholly owned subsidiaries of other freight forwarding companies aready included in that figure. Therefore, the response rate in the AIR sample is an underestimation of actual population coverage.

6.3 Presentation of Empirical Results

In this section, first sample characteristics are thorougly discussed, before personal network structure and position as well as intermediary roles of the freight forwarders in the sample are assessed. In order not to overload this section with tables and graphs, detailed data as well as results of statistical tests are tabulated in the appendix.

6.3.1 Sample Characteristics

In the following, first, the respondents and their companies are described, before a general description of typical shipments and their organization is given. Whenever available, a comparison to other surveys is offered. The section closes with further comparisons based on statistical tests.

6.3.1.1 Respondents and Their Companies

Taking the whole sample, most of the survey respondents (53.2%) stated that their position is either being company owner, chief executive officer (Geschaftsf¨ ¨ uhrer in German) or general manager (Prokurist in German).⁴⁰ Furthermore they showed considerable work experience with an average of 21:80 years in forwarding business and 12:20 years in their company with a maximum of 50 years (!) for both figures. Moreover, the respondents of the SEA sample could be regarded to be slightly more experienced than in the AIR Sample with 23:26 versus 19:71 years working in the forwarding business and 14:07 versus 10:71 years in their company.

Taking the number of offices as a proxy for company size, a freight forwarding company in the AIR sample has on average 3:50 sites in Austria and 49:71 subsidiaries abroad.⁴¹ In the SEA sample, it has on average 7:74 sites in Austria and Germany respectively and 80:04 subsidiaries abroad (see Appendix to Question 2). But looking more closely on the distribution of these figures, the majority of freight forwarder firms are tiny with 10 single site companies and 21 companies without any subsidiary abroad – only 16 companies in the total sample (and 10 out of 14 or 71.4% in FCL-D) had more than 10 subsidiaries in total. This is in line with findings of e.g. DSLV (2005), where 85% of all freight forwarders in Germany with more than one single site have less than 10 subsidiaries and even only 4%

⁴⁰See Appendix to Question 8 for this and the following.

⁴¹By counting employees, Kaltenbrunner (2005:60–63) can partly confirm this with 82.7% of his respondents having less than 50 employees in Austria with 81.8% less than 50 of them engaged in air cargo operations. In contrary to this, 50.0% have more than 1; 000 employees and even 31.8% more than 5; 000 employees working for them abroad.

of them have more than 20. ⁴² Furthermore, the respondents defined their office in 60.4% of all cases as a single company, with the other being either a main office (Hauptniederlassung), branch office (Zweigniederlassung) or trade office (Handelsb¨uro).

Referring to the respondents' company fields of activity, they seem to be very specialized with a combination of sea, air and/or international freight forwarding as well as customs brokerage and warehousing usually offered by 66.7% or more on average in the total sample (cf. Appendix to Question 1 for further reference):

In the SEA sample, the companies are mainly focused on sea cargo forwarding (79%) as well as international freight forwarding (67%). In addition to this, other services usually offered are customs brokerage (83%), followed by forwarding of project cargo (71%), air cargo forwarding (67%), warehousing (67%) and groupage service (54%). Furthermore, 46% act as a NVO, 42% offer packaging services and 38% stuffing and stripping of containers.43
In the AIR sample, the companies are focused on air cargo forwarding (67%), followed by international freight forwarding (58%), customs brokerage (54%) and sea cargo forwarding (50%), too. Apart from this, they usually offer warehousing (67%), road transport on own account (79%), groupage service (75%) and are engaged in parcel or express services (46%), affreightment of hauliers (46%), act as an IATA cargo sales agent (46%) or removal agent (46%) as part of their business.
Comparing the AIR and FCL-A sample with WIFI (1993), the respondent companies can be regarded as freight forwarding specialists, as they usually offer on average more international freight forwarding, customs brokerage and warehousing than Austrian freight forwarding companies according to WIFI (1993). Moreover, their activity in groupage service as well as road transport on own account is higher than in WIFI (1993).
Comparing the FCL-D sample with DSLV (2005), the respondents from Hamburg, Bremen and Bremerhaven show a real specialization on a combination of sea, air and/or international freight forwarding in connection with customs brokerage, warehousing and project cargo forwarding usually offered, whereas groupage service, road transport on own account and affreightment of road hauliers are of minor interest. In contrast to this, at least some of them (35%) offer more affreightment of railway operators and forwarding in inland

⁴²Markides and Holweg (2006) obtained similar results in their survey of international freight forwarding companies in the UK with an average of 3:2 offices and 1:1 warehouses per firm.

⁴³Moreover, Murphy and Daley (2001) in the USA as well as Markides and Holweg (2006) in the UK give further support for the main fields of activity in the SEA sample. Following Murphy and Daley (2001), 81.0% of the international freight forwarding companies in their sample offer customs brokerage and 57.0% NVOCC service, whereas 48.0% of the companies in Markides and Holweg (2006) offer customs brokerage, 58.4% NVOCC, 48.1% warehousing and 42.9% packaging services.

navigation than the average German freight forwarding company according to DSLV (2005).

• A comparison between FCL-HH, FCL-HB and DSLV (2005) adds nothing special than that the respondent companies from Hamburg more often offer warehousing, groupage service and affreightment of hauliers than their colleagues from Bremen and Bremerhaven coming along with more offers in packaging services and project cargo forwarding.

Last but not least, the share of turnover generated by sea, air and international freight forwarding was asked for. Most respondents (79:1% in the AIR sample and 81:8% in the SEA sample) stated that more than 50% of the turnover of their business operations fall in the fields of international freight forwarding, and freight forwarding in shipping and/or air freight operations. Furthermore, 50:0% of all companies in the AIR sample and 36:3% in the SEA sample can be regarded as highly specialized as they generate more than 75% of their turnover in these fields of operation.

6.3.1.2 General Description of Typical Shipments

As already indicated in Sect. 6.2, the respondents were asked for a typical shipment. The resulting export shipment can be characterized as follows (see Appendix to Question 4 for further reference):

USA was the most mentioned country of destination for export shipments, with a share of 52:2% in the AIR and 38:9% in the SEA sample, followed by various destinations in the Far East like Japan, South Korea, People's Republic of China, Taiwan or Indonesia with 43:5% (AIR) and 38:9% (SEA).
The majority of air freight shipments in the AIR sample departed at Vienna International Airport (56:5%), followed by Frankfurt-on-Main (21:7%) and Linz (13:0%). In the SEA sample, main sea ports of departure for export shipments were Hamburg (55:5%) followed by Bremerhaven (27:8%).
Referring to INCOTERMS used, 43:5% of all export shipments in the AIR sample were EXW (ex works) and 34:8% FCA (Free Carrier) with the rest being either DDU (duty delivered unpaid) or DDP (duty delivered paid). As the consignor was the freight forwarder's client for all export shipments in AIR, it can be concluded that most of the exporting shippers were not interested too much in being involved in the arrival and final delivery of their export shipments in the country of destination.⁴⁴ In case of the SEA sample, various INCOTERMS from EXW to DDP were used with no clear favourite.
According to transport documents used, all AIR shipments in export were accompanied by an AWB (air waybill) and 61:1% of the SEA shipments by a B/L

⁴⁴See Neiberger (2007, 2008) with a similar discussion about the growing trend towards EXW and FCA in air cargo transport operations.

(Bill of Lading) which are at the same time the dominant types of accompanying transport documents on air and sea transport operations. But in a considerable number of cases, other transport documents were employed for export shipments with the main haul by sea, too, like CBL (Combined Bill of Lading) or FBL (FIATA Bill of Lading), indicating the onward carriage leg to the final destination to be a rather complex undertaking organized by the respondents.

• Other services like coverage of cargo insurance, export packaging and documentary collection were mentioned in the case of export shipments with cargo insurance mostly offered to their clients (91:3% (AIR) and 55:6% (SEA)). Interestingly, export packaging as well as documentary collection was more often offered by the companies of the AIR sample with 52:2% and 43:5% of all export shipments than in the case of SEA with each mentioned only in 22:2% of all cases.

Last but not least the respondents stated that their typical export shipment as described occurs 2; 662 times per year in AIR and 383 times per year in SEA on average, which gives some evidence that they should be very typical ones. But a considerable spread in the answers cannot be negated with a standard deviation of 5; 013 in the case of AIR and 834 for SEA.

Accordingly, the import shipments can be characterized as follows (cf. Appendix to Question 6 for further reference):

Referring to the country of origin, USA was again most mentioned with 27:3% of all responses in AIR and 44:4% in SEA. But the overall majority of import shipments came from the Far East (e.g. Japan, South Korea, People's Republic of China or Hong Kong) with 63:6% in AIR and 27:8% in SEA.
In the AIR sample Vienna International Airport (68:2%) was again in lead as the airport of arrival, followed by Frankfurt-on-Main (9:1%) and Linz (9:1%). Furthermore, Hamburg (61:1%) followed by Bremerhaven (16:7%) were again mentioned as the main sea ports of arrival in the SEA sample.
In the AIR sample, EXW (54:6%) and FCA (22:7%) were again the INCOTERMS mostly mentioned. The fact that the consignee was the freight forwarder's client for 95:6% of the import shipments in AIR reinforces the initial guess from above that export shippers generally are not too much interested being involved in arrival and final delivery of their shipments. In the case of SEA, EXW (33:3%) is in the lead, closely followed by FOB (free on board) with 27:8% of all responses. Furthermore, for 66:7% of the import shipments, arrangements for onward carriage had to be made on behalf of the importer which again indicates that the majority of these import shipments were just dispatched at the sea port of departure and send to them without further involvement by the exporter abroad.
All import shipments in the case of AIR were again accompanied by an AWB. In the SEA sample, the majority came in with a B/L (61:1%), but also SWB (sea waybill), CBL as well as FBL were used as an accompanying transport document.

• Concerning other services offered, coverage of cargo insurance was again mostly mentioned with 90:9% in AIR and 33:3% in SEA. Furthermore, documentary collection was arranged for 36:4% of the incoming air cargo and 16:7% of the sea cargo shipments and in 22:2% of all SEA cases, the incoming container was stripped.

As the respondents stated that their import shipment as described occurs on average 2; 362 times per year in AIR and 148 times per year in SEA, again it can be regarded as typical, although the standard deviation of 3; 715 (AIR) and 305 (SEA) is again rather high.

6.3.1.3 Organization of Typical Shipments

The arrangements made in case of a typical export shipment are as follows (see Appendix to Question 5 for further reference):45

In the case of all export shipments in AIR and 77:7% in SEA, the freight forwarder was assigned by the shipper or exporter, mainly based on written contracts. The amount of contacts per year were 361:6 in AIR and 252:2 in SEA on average, which seems to be rather high and dense with at least one contact per working day.⁴⁶
The pre-carriage arrangement in AIR was either transport on own account (30:4%) or by some other haulier (60:9%) than the freight forwarding company with another 8:7% of the export shipments being transported even by the airline itself called airline trucking. Furthermore, the majority of these arrangements was closed on the basis of a written contract relationship. In the case of SEA, again various alternatives were employed like carrier haulage (37:5%), merchant haulage (25:0%) as well as other sorts of hauliers (31:3%).47 With 367:9 (AIR) and 224:8 (SEA) on average, the amount of contacts per year is again rather high with at least one contact per working day.
Customs procedures were mainly completed manually in AIR (60:9%), where as in the case of SEA, the majority of the freight forwarding companies employed electronic procedures (61:1%). This outcome stems from the fact that at Hamburg as well as Bremen/Bremerhaven, freight forwarding companies have direct

⁴⁵In the SEA sample, for each relationship coming along with organizing an export or import shipment additional statements had do be judged with the help of a 6-point Likert scale. The results thereof were rather inconclusive and are therefore only recorded in Appendix to Question 5 without further efforts of interpretation.

⁴⁶Assuming that in 50 weeks each year, a freight forwarding company may open its office Monday to Friday, there are about 250 working days per year. This rough estimation is also in line with official labour regulations in Germany as well as Austria where office clerks may work 5 days with 8 h per day.

⁴⁷Other hauliers might be there some container trucking companies or intermodal railway transport operators.

access to the cargo community systems DAKOSY and DBH which eases the handling of paperwork for export shipments to a high extent. Furthermore, they have about two times per working day contact to customs officials or customs agent employed, which count up to 584:5 (AIR) and 666:7 (SEA) contacts per year on average. Moreover, the contractual basis of their relationships is mostly personal and/or occasional in AIR and oral and/or occasional in SEA in comparison to other stages of an export shipment.⁴⁸
Referring to the terminal at the sea or air port of departure, the contacts per year are much higher in AIR (741:9) than in SEA (204:1) on average, which again gives some evidence that e.g. the cargo community systems of DAKOSY and DBH are working very efficiently in automatizing information processing at both sea ports. In the case of export shipments by air, there seems to be much more need in direct communication – about three times per working day on average. In addition to this, the majority of freight forwarding companies in the SEA sample declared that their relationship to the terminal at the sea port of departure is occasional whereas in the AIR sample, the relationship to the terminal at the airport of departure is more often based on a written or oral contract which fits well to the above mentioned assertions.
In the AIR sample, the main haul is mainly arranged either via the carrier agent like an IATA cargo sales agent (47:8%) or direct booking at the airline (39:1%), whereas in case of SEA, direct booking at the shipping company is done in 38:9% of all cases, with employing transport intermediaries like carrier agent, own NVO, other NVO or even others being a more common way to close the arrangement with the shipping companies for the main haul. This part of the organization of an export shipment seems to be very personal driven, as the contacts per year with the airline, shipping company or transport intermediaries in between are on average 523:5 (AIR) and 508:5 (SEA). Furthermore, the majority of all main haul arrangements are based on written contracts which show the importance of holding good, long lasting relationships with the carrier on the main haul or the transport intermediary in between.
Regarding arrangements for onward-carriage for export shipments, the majority of freight forwarding companies in the AIR sample was not involved (52:2%), which is in line with an already above observed potential resistance of shippers to be involved in transport arrangements of their shipments from the airport of destination to the consignee. In the SEA sample, correspondence relationships (44:4%) might be a dominant way to arrange onward-carriage of export shipments, but a closer look on the data reveals that this outcome stems from the fact that 66:7% of respondents from the FCL-HB sample alone employed correspondence relationships where the other two (FCL-A and FCL-HH) were

⁴⁸In this context, occasional means that the respondents have no special relationship with customs authorities or their customs agents with written contract, oral contract or personal contact determined as more tighter forms of relationship.

more often not involved, counting up to 27:8% in the whole sample.49 Asking about the contacts per year with the partners abroad (if there were any), the respondents declared that they have 145:6 (AIR) and 299:1 (SEA) contacts on average, which can be regarded as rather low in comparison to the other figures mentioned.

• Concerning the modes of communication, traditional ones like telephone and fax are heavily employed by the respondents in both samples but email is gaining ground being second after telephone. Other modes of communication mentioned are postal correspondence and EDI – especially in the case of the SEA sample. This can be explained by the need for traditional paperwork e.g. in the case of B/Ls as negotiable documents of evidence on one hand and well established cargo community systems like DAKOSY and DBH on the other.50 In the AIR sample, EDI was not employed at all and telephone was the outmost important mode of communication followed by email and fax despite there is no need for physical presence of AWBs as accompanying transport documents and TRAXON (among other cargo community systems in the airline industry) can be used for all kinds of electronic exchange.⁵¹

In case of typical import shipments the results are as follows (see Appendix to Question 7 for further reference):

The freight forwarding companies' client was mainly the consignee or importer of that import shipment with 95:5% (AIR) and 66:7% (SEA) of all cases and this relationship was mainly based on a written contract. The contacts to their clients seemed to be on average more intensive in the AIR sample with 589:4 times per year than in the case of SEA with 267:7 contacts per year.
The same holds true regarding the amount of contacts per year on average with the sea- or airport of arrival: 601:9 (AIR) versus 268:4 (SEA). The basis of relationship was defined to be either occasional, based on oral or written contract without indicating a clear trend.
Customs procedures in the AIR sample were mainly arranged manually (81:8%) and in the case of SEA, the majority of the freight forwarding companies employed electronic procedures (50:5%) or authorized intermediaries like own declarants or customs brokers to deal with customs authorities on their behalf. Furthermore, the relationship to them was very often regarded as occasional. Moreover, the contacts with customs authorities or agents employed were very

⁴⁹In addition to this, in both samples (SEA and AIR) two times joint ventures were mentioned as a partner to arrange onward-carriage of export shipments. Moreover, these correspondence as well as joint venture relationships are either based on a written or oral contract.

⁵⁰This is in line with findings of e.g. DSLV (2005:45) that 43% of all freight forwarding companies in Germany use cargo community systems like DAKOSY, DBH or TRAXON and hold contact to customs authorities via ATLAS. Further, the traditional paperwork in maritime shipping is burdensome and steps to reduce them significantly are demanded, see e.g. Frankel (2002) or Stopford (2002).

⁵¹See Kaltenbrunner (2005) or Gillis (2008) for more detailed empirical evidence about this.

intensive, especially in the case of SEA with 812:6 times per year (or more than three times a working day) on average with freight forwarding companies in the AIR sample having only 340:4 times per year contact to them on average.

Concerning onward-carriage, 36:4% of all import shipments were transported to the consignee on own account and 50:0% by other hauliers in AIR based on a written contract, whereas in the SEA sample, the results are mixed with carrier haulage (27:8%), merchant haulage (27:8%) or other haulier (38:9%) mostly mentioned.⁵² Accordingly, the contacts per year to the carrier of the onward-carriage are more intensive in AIR (471:0) than in SEA (166:9) on average.
Assessing the modes of communication, the findings from the export shipments are merely reinforced: Again telephone, fax as well as email are heavily employed in both samples (AIR and SEA). But in the case of the SEA sample other modes of communication seem to be important at least for some respondents. Whereas postal correspondence occurs in all stages, telex (as the predecessor of fax) was mentioned as a mode of communication vis-`a-vis the terminal at the sea port of arrival, and EDI was used if electronic customs procedures like ATLAS were employed in the case of FCL-HH and FCL-HB.

All in all, these descriptions of typical export and import shipments from AIR and SEA mirror the general statements about contemporary sea and air transport operations made in Sect. 2.2.4 given that at time of the surveys in 2003, usage of electronic customs procedures (although applicable, see Kummer et al. (2010:222 –226)) were non-obligatory and cargo community systems like TRAXON in air cargo transport operations may get more attention in the meantime.

6.3.1.4 Further Comparisons

As the size of the social networks under observation was considerably below ^N D ¹⁵ and an overwhelming majority of parameters turned out to show a non-normal standard distribution after testing it with the One-Sample Kolmogoroff– Smirnov Test (Kolmogoroff (1933), Smirnov (1939a, 1939b)) with a correction factor by Lilliefors (1967) as well as the Shapiro–Wilk Test (Shapiro and Wilk 1965) at the 5% level, only simple nonparametric test methods for independent samples (namely Mann-Whitney-U Test (Wilcoxon 1945; Mann and Whitney 1947), Two-Sample Kolmogoroff–Smirnov Test (Kolmogoroff 1941), Several-Sample Kruskal– Wallis Test (Kruskal 1952; Kruskal and Wallis 1952) and Several-Sample Median Test (e.g. Westenberg 1948 or Siegel (1956:179–184)) well implemented in SPSS

⁵²Looking more closely to the figures, FCL-HH seemed to favour carrier haulage, FCL-HB merchant haulage and FCL-A other hauliers like container trucking companies or intermodal railway transport operators. But regarding sample size (N D 14 in total but only N D 4 in the case of FCL-HH and FCL-A), this findings cannot be generalized.

could be applied to the samples.⁵³ The findings from intensive non-parametric testing of all samples were as follows (all differences shown turned out to be statistically significant at the 5% level, see Appendix of Statistical Tests for further reference):

Testing AIR vs. SEA, it can be asserted that outgoing air cargo shipments generally move to more distant destinations than shipments by sea in every aspect (geographic, cultural as well as economic distance), whereas incoming air cargo shipments come only from more geographic distant origins. In contrary to exporters of seagoing cargo, exporters of air cargo shipments want to be involved primarily in despatching. Therefore, they prefer EXW and FCA as an INCOTERM. Moreover, the freight forwarders engaged in air cargo shipments account for significantly more shipments per year of the same kind and hold more formal relationships with their customers in both directions (export and import). Focusing on the amount of contacts per year, the air cargo freight forwarders have on average a significantly higher amount of contacts per year with the cargo terminals at the airport in export and import and with their clients in import, but a significantly less amount of contacts per year with customs offices or agents coping with them. This reflects actual business practice where the air cargo freight forwarder always needs to supervise and direct air cargo shipments whereas seagoing shipments are processed with the help of CCSs like DAKOSY and DBH. Last but not least, the status of the respondent persons' position in the AIR sample is higher on average than that of their counterparts engaged in freight forwarding operations including a leg by sea on the main haul.
Comparing FCL-D vs. FCL-A, there seem to be no statistically significant differences between German and Austrian freight forwarding operations including a leg by sea on the main haul except for EDhf in export. But considering the sample size, no general conclusions can be drawn from this.
Looking more closely on the data and comparing FCL-A vs. FCL-HH vs. FCL-HB, some statistically significant differences between them can be argued, like the formalism of relationships with the terminals in export and import as well as the relationships with their clients in import. Moreover, the respondents' number of years employed with their company differs significantly.

⁵³All parameters tested are at least ordinal scaled with "Status of Responding Company Site" showing a meaningful descending rank ordering of status with values from 1 (single company) to 2 (main office), 3 (branch office) and 4 (trade office), "INCOTERMs used..." as a trade term can be rank ordered with designating values from 1 (EXW) to 13 (DDP) indicating an increasing involvement of an exporter in the shipment, "Basis of Relationship to ..." can be ordered in a decending order of formalism with values from 1 (written contract), 2 (oral contract), 3 (personal contact) to 4 (occasional), and 'Position of Respondent' again can be interpreted as decending order of status with values from 1 (company owner), 2 (chief executive officer), 3 (general manager), 4 (division manager) to 5 (other position).

• Finally, testing FCL-HH vs. FCL-HB, freight forwarders from Hamburg seem to have less contacts per year with the local customs office (or agents acting on behalf of them) in the case of export shipments and hold less formal relationships to the terminals in export as well as their clients in import than their counterpart from Bremen and Bremerhaven. But the overall statistical support for this is rather weak. However, the respondents from Hamburg are significantly less years employed with their company.

All in all, not too many statistically significant differences between the samples could be identified, which may mainly stem from the fact that the overall sample size is rather small and the group of respondents is very heterogeneous referring e.g. to company size and a huge variety of typical shipments in export and import were assessed.

6.3.2 Personal Network Structure and Position

Based on Question 5 and 7, several measures of network structure (namely NS, NSego, ND, ND⁰ and NDalter ) and position (CD.pk/, CB .pk/, CC .pk/-¹, CR.pk/, C⁰ ^D.pk/, C⁰ ^B .pk/ and C⁰ ^C .pk/) can be calculated and compared with the benchmark of a fully integrated transport chain of an integrator (see Sect. 6.1.2).

First, ego's personal network structure can be described by measures of network size and density (e.g. NSego and ND) as shown in Tables 6.4 and 6.5. A first look at Table 6.4 reveals significantly higher NS and lower ND⁰ in export operations for FCL-HB in comparison to FCL-HH, which indicates that freight forwarders from Bremen or Bremerhaven may have a higher amount of social capital than their counterparts from Hamburg , although NSego and ND turned out to be not

Table 6.4	Network size and density in export
Sample	NS	NSego	ND	ND0	NDalter
FCL-A	7.00	5.00	0.28	0.28	0.08
	(1.00)	(0.82)	(0.04)	(0.04)	(0.02)
FCL-HH	6.80	5.20	0.30	0.30	0.05
	(0.45)	(0.84)	(0.02)	(0.02)	(0.04)
FCL-HB	7.78	5.22	0.28	0.26	0.09
	(0.46)	(0.83)	(0.03)	(0.02)	(0.03)
FCL-D	7.43	5.21	0.28	0.27	0.08
	(0.65)	(0.80)	(0.03)	(0.02)	(0.04)
SEA	7.35	5.17	0.28	0.28	0.08
	(0.70)	(0.79)	(0.03)	(0.03)	(0.04)
AIR	7.22	5.39	0.28	0.28	0.04
	(0.67)	(0.58)	(0.03)	(0.03)	(0.04)
Integrator	7.00	6.00	0.29	0.29	0.00

Table 6.4 Network size and density in export

Sample	NS	NSego	ND	ND0	NDalter
FCL-A	5.50	4.00	0.37	0.37	0.05
	(0.58)	(0.00)	(0.04)	(0.04)	(0.06)
FCL-HH	4.80	3.60	0.43	0.43	0.02
	(0.58)	(0.58)	(0.06)	(0.06)	(0.00)
FCL-HB	5.00	3.78	0.40	0.40	0.04
	(0.33)	(0.44)	(0.02)	(0.02)	(0.07)
FCL-D	4.93	3.71	0.41	0.41	0.03
	(0.47)	(0.47)	(0.04)	(0.04)	(0.06)
SEA	5.06	3.78	0.40	0.40	0.04
	(0.54)	(0.43)	(0.04)	(0.04)	(0.06)
AIR	5.00	3.74	0.40	0.40	0.01
	(0.44)	(0.83)	(0.04)	(0.04)	(0.03)
Integrator	5.00	4.00	0.40	0.40	0.00

Table 6.5 Network size and density in import

significantly different in magnitude.54 The same is true if we look at Table 6.5, where for import operations NS is significantly higher and ND⁰ as well as ND is significantly lower for FCL-HB in comparison to FCL-HH.55 This can be interpreted again as a higher amount of social capital for freight forwarders at Bremen or Bremerhaven than for their counterparts at Hamburg. Furthermore, FCL-D offers a slightly higher NS and lower ND as well as ND⁰ in comparison with FCL-A for import operations, so that German freight forwarders tend to show a higher amount of social capital there than the Austrian ones.56 Unfortunately, these results for measures of network size and density are not really convincing, as differences in NSego are non-significant for all samples. Again, this seems to be a matter of an overall high heterogeneity of responses paired with low sample sizes.

Second, measures of prominence (e.g. CD.pk/, CB .pk/, CC .pk/ and CR.pk/) can give information about ego's network position as shown in Tables 6.6 and 6.7. Looking again first at export operations, Table 6.6 shows that freight forwarders engaged in air transport have a higher amount of social capital as their CC .pk/ -1, CR.pk/, C⁰ ^B .pk/ and C⁰ ^C .pk/ are significantly higher than that of their counterparts engaged in maritime shipping.⁵⁷ Furthermore, freight forwarders at Bremen or Bremerhaven tend to have a higher social capital as CC .pk/ -¹ and C⁰ ^C .pk/ is higher

⁵⁴Statistically significant at the 5% level for NS according to Mann-Whitney-U (z D -2:758; - D 0:006) as well as Kolmogoroff–Smirnov (z D 1:394; - ^D 0:041) and for ND⁰ according to Mann-Whitney-U (z D -2:460; -D 0:014), see Appendix Statistical Tests.

⁵⁵Statistically significant at the 5% level for NS, ND and ND⁰ according to Mann-Whitney-U (z D -2:169; -D 0:030), see Appendix Statistical Tests.

⁵⁶Statistically significant at the 10% level for NS, ND and ND⁰ according to Mann-Whitney-U (z D -1:889; -D 0:059), see Appendix Statistical Tests.

⁵⁷Statistically significant at the 5% level for CC .pk /-¹ according to e.g. Mann-Whitney-U (z D -2:425; - D 0:015), CR.pk/ according to e.g. Mann-Whitney-U (z D -2:580; - D 0:010) C0 ^B .pk / according to e.g. Mann-Whitney-U (z D -2:109; - ^D 0:035) and ^C⁰ ^C .pk/ according to

Sample	CD.pk/	CB .pk /	1 CC .pk /	CR.pk /	C0 D .pk /	C0 B .pk/	C0 C .pk/
FCL-A	5.00	14.50	7.50	2.51	0.88	0.95	0.92
	(0.82)	(4.20)	(1.29)	(0.36)	(0.10)	(0.05)	(0.07)
FCL-HH	5.20	13.40	6.40	2.50	0.89	0.95	0.84
	(0.84)	(2.51)	(0.55)	(0.31)	(0.10)	(0.04)	(0.08)
FCL-HB	5.44	17.28	8.89	2.4	0.80	0.88	0.77
	(0.92)	(3.57)	(0.76)	(0.07)	(0.11)	(0.12)	(0.09)
FCL-D	5.36	15.89	8.00	2.44	0.83	0.90	0.82
	(0.84)	(3.63)	(1.41)	(0.19)	(0.11)	(0.10)	(0.11)
SEA	5.28	15.58	7.89	2.46	0.83	0.90	0.82
	(0.83)	(3.7)	(1.37)	(0.23)	(0.10)	(0.09)	(0.10)
AIR	5.43	15.57	8.83	3.19	0.88	0.95	0.92
	(0.59)	(3.63)	(0.94)	(0.88)	(0.10)	(0.05)	(0.07)
Integrator	6.00	15.00	7.00	2.45	1.00	1.00	1.00

Table 6.6 Centrality measures for complete network in export

Table 6.7 Centrality measures for complete network in import

Sample	CD.pk/	CB .pk /	/ 1 CC .pk	CR.pk /	C0 D .pk /	C0 B .pk/	C0 C .pk/
FCL-A	4.00	7.50	5.00	2.65	0.90	0.95	0.92
	(0.00)	(1.73)	(1.15)	(0.75)	(0.12)	(0.06)	(0.10)
FCL-HH	3.50	4.50	3.50	1.87	1.00	1.00	1.00
	(0.58)	(1.73)	(0.58)	(0.15)	(0.00)	(0.00)	(0.00)
FCL-HB	3.80	6.10	4.40	2.21	0.93	0.96	0.94
	(0.44)	(1.17)	(0.73)	(0.44)	(0.12)	(0.07)	(0.09)
FCL-D	3.71	5.64	4.14	2.11	0.95	0.97	0.96
	(0.47)	(1.45)	(0.77)	(0.39)	(0.10)	(0.06)	(0.08)
SEA	3.78	6.06	4.33	2.23	0.94	0.96	0.95
SEA	(0.43)	(1.66)	(0.91)	(0.52)	(0.10)	(0.06)	(0.08)
AIR	3.91	6.00	4.09	2.09	0.98	0.99	0.98
	(0.29)	(1.31)	(0.68)	(0.40)	(0.06)	(0.03)	(0.05)
Integrator	4.00	6.00	4.00	2.00	1.00	1.00	1.00

than in the case of their counterparts from Hamburg.⁵⁸ Referring to Table 6.7, freight forwarders from Bremen or Bremerhaven again show some higher amount of social capital in their import operations as both CC .pk/-¹and CR.pk/ are significantly higher than in the case of freight forwarders from Hamburg.⁵⁹ Last but not least,

e.g. Mann-Whitney-U (z D -2:986; - D 0:003), see Appendix Statistical Tests. Further, it is notable that CC .pk / ¹ has a positive relation to social capital as it is the inverse of CC .pk /.

⁵⁸Statistically significant at the 5% level for CC .pk /-¹ according to e.g. Mann-Whitney-U (z D -3:068; - ^D 0:002), and statistically significant at the 10% level for ^C⁰ ^C .pk / according to Mann-Whitney-U (z D -1:818; -D 0:069), see Appendix Statistical Tests.

⁵⁹Statistically significant at the 5% level for CC .pk/-¹ as well as CR.pk / according to Mann-Whitney-U (z D -2:149; -D 0:032), see Appendix Statistical Tests.

FCL-A shows significant higher CR.pk/ but lower CB .pk/ in comparison to FCL-D which seems to be a rather inconclusive result at first sight. But taking into account that CB .pk/ indicates the degree of betweeness and CR.pk/ the influence of an actor in a social network structure, it can be presumed that their source of social capital seems to be different: Freight forwarders at Hamburg, Bremen and Bremerhaven are engaged in organizing local operations at seaports whereas Austrian freight forwarders exploit their position as a remote acting agent for their clients in the hinterland.⁶⁰ To conclude, freight forwarders engaged in air cargo operations show a higher amount of social capital than their counterparts engaged in door-todoor FCL-shipments. Furthermore, freight forwarders at Bremen and Bremerhaven possess an overall higher amount of social capital than their colleagues from Hamburg. Both outcomes may stem from the fact that these Austrian air cargo freight forwarders and freight forwarders from Bremen and Bremerhaven do not act very often on own account but employ other transport chain participants to organize their shipments.

6.3.3 Intermediary Roles

In Tables 6.8 and 6.9, averages of individual raw brokerage scores (b01k, b0k, b10k and *w*1k ) for each freight forwarder engaged in multimodal transport chains as an intermediary k between his client i and the other transport chain participants j adjusted by the sum of all global raw brokerage scores .B C W / are shown. The averages for itinerant brokerage scores *w*0k are omitted from these tables, because they were virtually 0 for all observations. In the last column, the averages of the degree of vertical integration V Ik are shown, again with the corresponding standard normal deviations in parentheses below.61

The same holds true for the integrator's import operations, where he despatches incoming cargo from an overseas subsidiary (his client) as a local broker (namely

⁶⁰Statistically significant at the 5% level for CR.pk / according to Mann-Whitney-U (z D -2:232; - D 0:026), and statistically significant at the 10% level for CB .pk/ according to Mann-Whitney-U (z D -1:940; -D 0:052), see Appendix Statistical Tests.

⁶¹Keeping this inherent network structure of multimodal transport chains in mind, the degree of vertical integration V Ik can be shortly defined as the movement of door-to-door shipments as well as handling of documentary paperwork as a logistics system run in a one-stop-shopping manner, so that most of the tasks usually done by different independently operating transport chain participants are concentrated in one central intermediary standing in between the shipper and the consignee. More specifically, the degree of vertical integration of the kth intermediary V Ik can be expressed formally as the ratio between the amount of tasks done or organized by him on own account and the amount of tasks done or organized at all along the multimodal transport chain under observation. Therefore V Ik will be high, if the central intermediary in question organizes many tasks on own account with the other transport chain participants serving only a minor role or being totally disintermediated. Furthermore, maximum V Ik D 1 with all tasks along the transport chain under observation are done or organized by one central intermediary on own account.

| Sample | Gatekeeper | Liaison | Representative | Local broker | V Ik | | |------------|-------------------|----------------------|-----------------------|----------------------|--------|--| | | b01k.B C W /
1 | .B C W /
1
b0k | .B C W /
1
b10k | .B C W /
1
w1k | | | | FCL-A | 0.20 | 0.57 | 0 | 0 | 0.36 | | | | (0.14) | (0.08) | – | – | (0.20) | | | FCL-HH | 0.08 | 0.79 | 0 | 0 | 0.52 | | | | (0.11) | (0.12) | – | – | (0.16) | | | FCL-HB | 0.04 | 0.69 | 0 | 0 | 0.31 | | | | (0.12) | (0.17) | – | – | (0.18) | | | FCL-D | 0.05 | 0.73 | 0 | 0 | 0.39 | | | | (0.11) | (0.14) | – | – | (0.19) | | | SEA | 0.08 | 0.69 | 0 | 0 | 0.38 | | | | (0.02) | (0.02) | – | – | (0.20) | | | AIR | 0.12 | 0.74 | 0 | 0 | 0.33 | | | | (0.12) | (0.12) | – | – | (0.16) | | | Integrator | 0.80 | 0.20 | 0 | 0 | 1.00 | |

Table 6.8 Brokerage scores and degree of vertical integration in export

Table 6.9 Brokerage scores and degree of vertical integration in import

Sample	Gatekeeper b01k.B C W / 1	Liaison .B C W / 1 b0k	Representative .B C W / 1 b10k	Local broker .B C W / 1 w1k	V Ik
FCL-A	0.06	0.81	0	0	0.25
	(0.10)	(0.17)	–	–	(0.19)
FCL-HH	0.07	0.88	0	0	0.40
	(0.06)	(0.17)	–	–	(0.14)
FCL-HB	0.00	0.93	0	0	0.26
	(0.08)	(0.00)	–	–	(0.17)
FCL-D	0.02	0.91	0	0	0.31
	(0.09)	(0.15)	–	–	(0.19)
SEA	0.03	0.89	0	0	0.30
	(0.10)	(0.17)	–	–	(0.19)
AIR	0.12	0.86	0	0	0.44
	(0.16)	(0.16)	–	–	(0.21)
Integrator	0	0	0.33	0.67	1.00

transhipment and onward carriage) and cares for customs clearance as a representative. Furthermore, all average raw brokerage scores as shown in Tables 6.8 and 6.9 do not necessarily add up to 1, because in the freight forwarder samples, some of the other transport chain participants act as intermediaries, too. One of many examples to this are customs house brokers, who are sometimes engaged to clear goods at customs on behalf of their clients like shippers, consignees or freight forwarders.

A first striking feature of both tables is that all freight forwarders serve neither a representative nor a local brokerage role at all *vis-a-vis* their clients, because in all freight forwarder samples b10k.B C W /-1 D *w*1k.B C W /-D 0 for all k. Furthermore, the majority of contacts vis-a-vis their clients are of the liaison type with a maximum average b0k.B C W /-D 0:57 if all tasks are done or organized for FCL-A export shipments and minimum average b0k.B C W /-D 0:93 in the case of FCL-HB import shipments, spanning across different subgroups with their clients on one hand and the other participants along the transport chain on the other.

Comparing FCL-A with FCL-D in Table 6.8, there is some striking empirical evidence that in the case of export shipments, Austrian freight forwarders seem to act merely as gatekeepers rather than as brokers of the liaison type as their German counterparts do.⁶² Especially freight forwarders from Hamburg engage themselves mostly in brokerage activity of the liaison type (with average b0k.B C W /-1 D 0:79) which at the same time results in a considerably higher average degree of vertical integration in comparison to their counterparts from Bremen and Bremerhaven.63 This is borne from the fact that most of them handle their customs paperwork in-house and book sea freight directly, whereas the other freight forwarders more often employ ship brokerage intermediaries with freight forwarders from Bremen and Bremerhaven heavily relying on carrier haulage services for precarriage.

According to Table 6.9, import shipments operations in the AIR sample come along with slightly more brokerage of the gatekeeper type than at their counterparts in the SEA sample, which results in a higher degree of vertical integration for freight forwarders in air freight import operations.⁶⁴ They more often conduct cartage service on own account, whereas for the freight forwarders in the SEA sample it seems to be more common just to dispatch incoming shipments with onward carriage organized in merchant or carrier haulage.

Comparing these empirical results with the benchmark case of a fully integrated transport chain, the situation is totally reversed: For outgoing export shipments, the integrator mainly serves as a gatekeeper for his own highly standardized logistics system, whereas import shipments coming from own subsidiaries abroad are mainly handled in-house and so brokerage of the representative or local type occurs.⁶⁵ This sums up to a maximum degree of vertical integration.

⁶²Statistically significant at the 5% level for b01k .B ^C W /-¹ according to Mann-Whitney-U (z D -2:031; - ^D 0:042) and for b0k .B ^C W /-¹ according to Mann-Whitney-U (z D -2:156; - D 0:043), see Appendix Statistical Tests.

⁶³Statistically significant for Export V Ik at the 5% level according to Mann-Whitney-U (z D -2:602; - D 0:009) and at the 10% level according to Kolmogoroff–Smirnov (z D -1:268; - D 0:080). Furthermore, statistically significant at the 10% level for Import V Ik according to Mann-Whitney-U (z D -1:8520; - D 0:069) as well as Kolmogoroff–Smirnov (z D -1:235; -D 0:095), see Appendix Statistical Tests.

⁶⁴Statistically significant at the 5% level for Import V Ik according to Mann-Whitney-U (z D -2:165; -D 0:030), see Appendix Statistical Tests.

⁶⁵Statistically significant at least at the 10% level for V Ik , b01k.B ^C W /-¹ and b0k .B ^C W /in export as well as V Ik and b0k .B ^C W /-¹ in import according to Mann-Whitney-U testing integrator benchmark case against SEA, AIR as well as total sample, see Appendix Statistical Tests for detailed test statistics. In the case of testing the integrator benchmark case against FCL-A, FCL-D, FCL-HH and FCL-HB, the sample sizes seemed to be too small to give overall statistically significant support according to Mann-Whitney-U for V Ik , b01k.B ^CW /-¹ and b0k.B ^CW /-¹ in export as well as V Ik and b0k.B ^CW /-¹ in import, but the results weakly support such assertions.

Last but not least, considering the values of C⁰ ^C .pk/ and C⁰ ^B .pk/ in Tables 6.6 and 6.7, the proposition of C⁰ ^C .integrator/>C⁰ ^C .freight forwarder/ and C⁰ ^B .integrator/>C⁰ ^B.freight forwarder/ can be supported, albeit the differences shown there turned out not to be of statistical significance as some of the freight forwarders in the sample actually integrated operations like cartage service, customs declaration or affreightment via own NVO otherwise executed by other transport chain members (see Sect. 6.3.1.3). Nevertheless, reflecting Taube's (2004) ¨ framework of social capital it can be reasoned that freight forwarder's social capital is some sort of leverage capital which stems from his ability to bring transport chain members together like bridging structural holes in a social network.

Chapter 7 Final Conclusions

In these concluding sections, first main findings from the previous parts of this book are briefly summarized before the contribution of this work to extant research and managerial implications are outlined. Finally, limitations are addressed and suggestions are given for further research.

7.1 Summarizing Main Findings

As an overall guiding principle of this work, role theory as outlined in Sect. 1.2 was employed which offers a body of definitions concerning role, position and status of a person or organization. Both entities were treated as equal in the context of this work as the freight forwarding business is a truly people-driven service industry. In the following, main findings are presented along the three research questions as initially stated in Sect. 1.2.

7.1.1 What Is the Intermediary Role of Freight Forwarders in Multimodal Transport Chains Including a Leg by Sea or Air?

This was encountered from three different perspectives as shown in Fig. 3.1. First a legal and functional perspective of freight forwarding was outlined in Chapter 2 which served as status quo or reference point for more elaborated theoretical and empirical treatments. Then an intermediary perspective was developed in Chapter 3, followed by a network perspective in Chapter 4. Finally in Chapter 6, a survey was conducted in order to show empirical evidence in the case of freight forwarders from Germany and Austria in 2003. Concerning the intermediary role of the freight forwarder in the context of this work, it can be concluded that he is 286 7 Final Conclusions

truly "The Architect of Transport" as this is reflected in all perspectives. Its roleset covers all necessary functions to organize shipments and to provide ancillary financial, transport and logistics services upon request, its position is one of a focal actor between shippers, consignors, consignees and the other participants along a multimodal transport chain and his status seems to be high.

7.1.2 Which Body of Theory May Help to Explore Its Intermediary Role Theoretically?

Two perspectives based on economic theory and/or economic sociology were thoroughly reviewed in Chapters 3 and 4 of this work, namely (1) an intermediary perspective consisting of New Institutional Economics (NIE), Market Microstructure Theory (MMT) and Intermediation Theory of the Firm (ITF) and (2) a network perspective with Network Models in Operations Research and Management Science, Strategic Networks, Dynamic Networks, Entrepreneurial Networks, Industrial Networks and Social Networks as a selection of existing network theories which were regarded as relevant in the context of this work.

Comparing these two bodies of theory with the initially made observations of the legal and functional perspective in Chapter 2 the intermediation perspective showed that a freight forwarder may be best described as a strange hybrid providing transport, logistics and financial services in addition to his basic intermediary function of organizing physical goods movements on behalf and account of his clients as being somehow the core of freight forwarding business. In terms of ITF, he has a widespread range of intermediary role-sets from being a market maker, matchmaker, information producer, -guarantor or certificator as well as agent for delegated expertise, bargaining, contracting and/or monitoring. Concerning the network perspective, it can be concluded, that a freight forwarder's intermediary role in multimodal transport chains with a leg by sea or air can at best be described as a focal company in a social network structure.

7.1.3 Which Kinds of Measures Are Needed to Catch Its Intermediary Role Empirically?

As Social Network Analysis in Chapter 4 turned out to show the highest usability in the context of this work, in Chapter 5 relevant analytical concepts for both personal network structure (e.g. network size and density) and intermediary position (e.g. centrality and prestige) as well as relationships (e.g. brokerage capacity and network constraint) were thoroughly discussed and suggestions were made how to apply these measures to empirical data in order to explore social network structures as well as social capital.

Concerning the empirical results of freight forwarder's personal network structure and intermediary position among participants of a multimodal transport chain with a leg by sea or air in Part 6, freight forwarders engaged in air cargo operations show a higher amount of social capital than their counterparts engaged in maritime shipments. Moreover, freight forwarders at Bremen or Bremerhaven tend to show the highest amount of social capital followed by their colleagues at Hamburg and the Austrian ones which may stem from the fact that they do not often act on own account but employ other multimodal transport chain participants to organize their shipments. A closer look at measures of centrality indicates that their sources of social capital are different as freight forwarders at Hamburg, Bremen and Bremerhaven are engaged in organizing local operations at seaports whereas Austrian ones exploit their position as a remote acting agent for their clients in the hinterland. These assertions are further reinforced by measures of brokerage capacity according to which freight forwarders in the sample generally serve neither a representative nor a local brokerage at all *vis-`a-vis*their clients of export shipments with the majority of their contacts being of the liaison type where they serve as an intermediary between their clients and other multimodal transport chain participants unless unrelated to each other. In addition to this, Austrian freight forwarders seem to act merely as gatekeepers rather than brokers of the liaison type whereas freight forwarders at Hamburg engage themselves mostly in brokerage activity of the liaison type as most of them handle their customs paperwork in-house and book sea freight directly, whereas the other freight forwarders more often employ ship brokerage intermediaries with freight forwarders from Bremen and Bremerhaven heavily relying on carrier haulage services for pre-carriage. Concerning import shipments, air cargo operations may come along with slightly more brokerage of the gatekeeper type than full container loads in maritime shipping. Freight forwarders engaged in air cargo shipments seem to conduct cartage service more often on own account, whereas in the case of maritime shipments it seems to be more common just to dispatch incoming cargo with onward carriage organized in merchant or carrier haulage. Finally comparing these empirical results with the benchmark case of a fully integrated transport chain, the situation is totally reversed: For outgoing export shipments, the integrator mainly serves as a gatekeeper for his own highly standardized logistics system, and import shipments coming from own subsidiaries abroad are mainly handled in-house and so brokerage of the representative or local type occurs. This leads to the conclusion that freight forwarding is far from being vertically integrated and further it can be reasoned that freight forwarder's social capital is some sort of leverage capital which stems from his ability to bring transport chain participants together like bridging structural holes in a social network.

7.2 Contribution to Extant Research

As this work came along with many reviews of already existing theory, there are several topics to highlight which may be useful even in other contexts.

288 7 Final Conclusions

In Chapter 2 a legal and functional perspective of freight forwarding was thoroughly discussed which is not too much reflected outside literature of strictly legal nature.

Developing an intermediary perspective in Chapter 3 deeply rooted in NIE led to strictly defined intermediary roles derived from MMT and ITF, which again resulted in a taxonomy of intermediaries and service providers in multimodal transport chains and logistics as shown in Sect. 3.3. But apart from just describing intermediary role-sets in the context of this work, NIE as well as MMT and ITF offer widespread applicability in law and economics as they help to clarify complex contractual relationships beyond simply legal considerations by introducing issues like bounded rationality, imperfect information structures as well as positive transaction costs.

Concerning the network perspective in Chapter 4, six relevant modelling approaches were identified. Considering their different approaches to network structures, a one-for-all approach is not available as their underlying body of theory deviates in many aspects in terms of theory objectives, drivers, actors and their linkages, line of argumentation, presumed network governance and analytical instruments offered. Especially Social Network Analysis (SNA) turned out to be an excellent approach to treat various sorts of network structures because it (1) allows to model a full set of possible actor roles as well as network governance without bias, (2) develops purely social aspects and/or information flows in network structures beyond strictly economic considerations and (3) allows to explore and/or quantify them directly.

Furthermore, various measurement concepts on a personal as well as overall network level from the wide prospering field of SNA are described en detail in Chapter 5. Among them, however, the brokerage role approach in social networks with non-overlapping subgroups seems to be a powerful measurement concept if intermediary relationships forming triads are present. This offers a new way to describe all sorts of network-like structures in transport and logistics operations as well as supply chain management, where up to now research on dyadic relationships is more common.

Overall, both an intermediary and network perspective has been added as theoretical background to conduct studies in the field of freight forwarding as well as the wider context of transport, logistics and supply chain management where several actors are needed to end in a joint product of transport and/or logistics service. Suggestion for further research in this directions are outlined below.

7.3 Managerial Implications

Even in such a globalized world of today, distance still matters in terms of time and costs spend on cargo movements (Frankel, 1999, 2002) as well as trade barriers and unfavourable business environments are still present in many countries as clearly shown e.g. by Ghemawat (2001), the Logistics Performance Index (LPI) of the Worldbank or the Enabling Trade Index (ETI) from The World Economic Forum.⁷⁸⁰ As shown in this book, freight forwarder help to overcome these obstacles of international trade as quite central player serving as a bridge between several actors along multimodal transport chains by sea or air. Furthermore, as the freight forwarding business in its core is a people's business, personal social networks truly matter. While not being conscious about this, freight forwarders deal day by day with these issues of embeddedness and social capital from social network theory. Beside this, other economic theories treated in this book like NIE, MMT or ITF are worthwhile to employ as they may lead to new insights and a further understanding of freight forwarding business which then may improve strategic management thinking in this field as e.g. Stabell and Fjeldstad (1998), Hakansson ˚ and Ford (2002), Gadde et al. (2003), Spulber (2003) or Button (2005) suggested in similar contexts.

7.4 Limitations and Suggestions for Further Research

Of course, some limitations of this work have to be noted. First, the reviews about PAT, MMT and IFT in Chapter 3 – albeit extensive – are far from being complete. Furthermore, other network theories than these six reviewed ones could have be taken into account in Chapter 4 which may have been suitable in this context, too. Moreover, a focus was set solemnly on measures on a personal and overall network level in Chapter 5.

Referring to Chapter 6, the overall sample size of the surveys got too small and can be regarded as exploratory and non-representative. But comparisons of sample characteristics with empirical data raised elsewhere showed that survey respondents and their companies can be considered as typical freight forwarders engaged in multimodal transport chains with leg by sea or air in terms of company size, fields of activity as well as company turnover generated by sea, air and international freight forwarding operations. Furthermore, the organizational arrangements made in case of typical shipments given by respondents for further investigation mirror general statements about contemporary sea and air transport operations as laid out in Chapter 2 considering that application of electronic customs procedures where nonobligatory in 2003 and cargo community systems in air cargo transport operations like TRAXON got more attention in the meantime. Therefore, customs procedures for air cargo shipments in the sample were mainly handled manually, whereas long established cargo community systems like DAKOSY and DBH were frequently used to process maritime shipments. Moreover, some more elaborated measurement concepts like prestige and network constraint – albeit initially intended – could not be applied to network structure data obtained by the surveys as it had to be reduced

⁷⁸⁰For more information about their content and methodology see http://www.worldbank.org/lpi and http://www.weforum.org/en/initiatives/gcp/GlobalEnablingTradeReport/index.htm.

290 7 Final Conclusions

to a binary graph. May be structured personal interviews would have led to better results, but this was a matter of time and financial funding during sample raising.

In addition to this, throughout this book industry dynamics were only marginally addressed like the internationalization processes and M&A activity of freight forwarding companies, issues of dis- and reintermediation along multimodal transport chains as well as impacts of overall business environment changes which offer opportunities for further research. First, it may be good to look closer on the internationalization process of freight forwarders in multimodal transport chains (see e.g. Lommelen, 2004 or Neiberger, 2007), a topic so far mainly reflected in an European context by Hertz (1993, 1996a,b), Stahl (1995a) or Ludvigsen (2000). Furthermore, this comes often with heavy M&A activity781, which follows typical patterns concerning country regions and main field of activity as noted e.g. by Schramm and Niedermaier (2008a,b). Moreover, dynamic aspects of vertical integration as well as dis- or reintermediation as shown exemplary in Sect. 4.4.2 have to be taken into consideration in the context of such multimodal transport chains.⁷⁸² Especially middlemen mainly economizing on searching and matching processes like real estate brokers, travel agents or brokers on financial as well as insurance markets as well as freight forwarders mainly engaged in matchmaking activity seem to be in danger of just being disintermediated.⁷⁸³ But at the same time, reintermediation of such middlemen does actually occur, if they adapt quickly to the new environment and differentiate themselves successfully by offering valueadding services.⁷⁸⁴ In addition to this, disintermediation may occur as a by-product of governmental action, as Carr and Crum (1995) showed in the customs brokerage business and Clott (2000) in case of US-American NVOCCs. This leads directly to topics of industry dynamics in a wider context, where e.g. political embeddedness and/or socio-cultural environment may have to be taken into account, as freight forwarding business as a specialized service industry is always exposed to such developments outside its sphere of control. As SNA is rather too descriptive, a case study approach to explore e.g. long-term strategic issues or longitudinal studies to research organizational developments and/or environmental dynamics would offer new insights in the freight forwarding business.

Last but not least, some authors like Beier (1989), Gentry (1996), Bask (1999, 2001), Larson and Gammelgaard (2001-2002) discussed the implications of triadic relationships in transport and logistics, where a carrier or a logistics provider is involved in a buyer–supplier or a consignor–consignee relationship. Furthermore,

⁷⁸¹See e.g. Dorrenbacher (1990), Plehwe (1999), St ¨ one (2001), Carbone and Stone (2005).

⁷⁸²See e.g. Schlichting (1931, pp. 64–66), Stern (1938, pp. 49–52), Kirchner (1950, pp. 136–144), Seidenfus (1973, p. 2), Davies (1981) or Miller (2002), about forward or backward oriented vertical integration in freight forwarding or maritime shipping.

⁷⁸³Cf. Amely (1994), Lewis and Talalayevsky (1997), Schmidt et al. (1998), Potzl (2000), ¨ Economides (2001), Globerman et al. (2001) or Sawyer et al. (2003).

⁷⁸⁴See e.g. Clemons and Hann (1999) in the case of travel agencies and Sawyer et al. (2003) for real estate brokerage.

recent works on triads of buyer–supplier–supplier as well as buyer–supplier–buyer relationships by Choi and Kim (2008), Li and Choi (2009) and Choi and Wu (2009) were added to the supply chain research agenda. Especially these research approaches seems to be very promising as triad relationships are fundamental building blocks of network structures and therefore can be used in the following to explore even more complex structures in the field of transport, logistics and supply chain management.

| • | | | |---|--|--| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |

Aberle G (1989) Der Spediteur als Generalist vor einer neuen Aufgabenteilung im Verkehr. In: DVWG (ed) Neue Aufgaben der Spedition im Rahmen verkehrslogistischer Entwicklungen, Schriftenreihe der Deutschen Verkehrswissenschaftlichen Gesellschaft e.V. (DVWG) No. B116, 6–15
Aberle G (2003) Transportwirtschaft: Einzelwirtschaftliche und gesamtwirtschaftliche Grundlagen, 4th edn. Oldenbourg, Munich, Vienna
Aberle G (2005) Entwicklungstendenzen auf den Ma rkten f¨ ¨ ur transportbezogene Logistikdienstleistungen. In: Lasch R, Janker CG (eds) Logistik management: innovative Logistikkonzepte, DUV, Wiesbaden, pp 379–394
Adler PS, Kwon S-W (2002) Social capital: prospects for a new concept. Acad Manag Rev 27(1):17–40
Admati AR, Pfleiderer P (1986) A monopolistic market for information. J Econ Theor 39(2): 400–438
Admati AR, Pfleiderer P (1988) Selling and trading on information in financial markets. Am Econ Rev 78(2)-Papers and Proceedings, 96–103
Admati AR, Pfleiderer P (1990) Direct and indirect sale of information. Econometrica 58(4): 901–928
Aertsen F (1993) Contracting out the physical distribution function: a trade-off between asset specificity and performance measurement. Int J Phys Distrib Logist Manag 23(1):23–30
Aghion P, Dewatripont M, Rey P (1994) Negotiation design with unverifiable information. Econometrica 62(2):257–282
Agrawal P (2002) Double moral hazard, monitoring, and the nature of contracts. J Econ [formerly Zeitschrift f¨ur Nationalokonomie] 75(1):33–61 ¨
Agrawal V, Seshadri S (2000) Risk intermediation in supply chains. IIE Transactions 32(9): 819–831
Agrell PJ, Lindroth R, Norrman A (2004) Risk, information and incentives in telecom supply chains. Int J Prod Econ 90(1):1–16
Ahuja RK, Magnanti TL, Orlin JB (1993) Network flows: theory, algorithms, and applications. Prentice-Hall, Englewood Cliffs
Akerlof GA (1970) The market for "lemons": quality uncertainty and the market mechanism. Q J Econ 84(3):488–500
Albaum G, Strandskov J, Duerr E, Dowd L (1995) International marketing and export management, 2nd edn. (reprint). Addison-Wesley, Wokingham
Albaum G, Duerr E, Strandskov J (2005) International marketing and export management, 5th ed. Prentice-Hall/Financial Times, Harlow

Albers S (2000) Nutzenallokation in Strategischen Allianzen von Linienluftfrachtgesellschaften, Arbeitsbericht Nr.101 des Seminars f¨ur Allgemeine Betriebswirtschaftslehre, Betriebswirtschaftliche Planung und Logistik der Universitat zu K ¨ o ln, Cologne ¨

Albers S, Delfmann W, Gehring M, Heuermann C (2005) Supply chain integration and supply chain integrators – towards a differentiated view. In: de Koster R, Delfmann W (eds) Supply chain management – European perspectives. Copenhagen Business School Press, Copenhagen, pp 62–91
Alchian AA (1965) Some economics of property rights. Il Politico 30:816–829
Alchian AA (1991) Property Rights. In: Eatwell J, Milgate M, Newman P (eds) The new palgrave: a dictionary of economics, vol 3 (reprint). Macmillan, New York, pp 1031–1034
Alchian AA, Demsetz H (1972) Production, information costs, and economic organization. Am Econ Rev 62(5):777–795
Alchian AA, Woodward S (1987) Reflections of the theory of the firm. J Inst Theoretical Econ 143:110–136
Alchian AA, Woodward S (1988) The firm is dead; long live the firm a review of Oliver E. Williamson's the economic institutions of capitalism. J Econ Lit 26(1):65–79
Alicke K (1999) Modellierung und Optimierung von mehrstufigen Umschlagsystemen, Institut f¨ur Fordertechnik und Logistiksysteme der Universit ¨ at Karlsruhe: Karlsruhe (also Doctoral Thesis ¨ at Universitat Karlsruhe) ¨
Allen WB (1997) The logistics revolution and transportation. In: Long SG (eds) Transport at the millenium. Sage, Thousand Oaks, pp 106–116
Allen F, Santomero AM (1998) The theory of financial intermediation. J Bank Finance 21(11):1461–1485
Amely T (1994) Beratungsintensive Dienstleistungen als Wachstumsfeld, Sparkasse 111(10): 481–483
Amihud Y, Mendelson H (1980) Dealership market marketmaking with inventory. J Financ Econ 8(1):31–53
Amiouny SV, Bartholdi III JJ, Van de Vate JH, Zhang J (1992) Ballanced loading. Manag Sci 40(2):238–246
Andersen O (1993) On the internationalization process of firms: a critical analysis. J Int Bus Stud 24(2):209–231
Anderson H, Havila V, Andersen P, Halinen A (1998) Position and role-conceptualizing dynamics in business networks. Scand J Manag 14(3):167–186
Anderson AR, Jack SL (2002) The articulation of social capital in entrepreneurial networks: a glue or a lubricant? Enterpren Reg Dev 14(4):192–210
Andersson D (1995) Logistics Alliances and Structural Change, Linkoping Studies in Science and ¨ Technology Thesis No.470, Linkoping ¨
Andersson D (1997) Third Party Logistics Outsourcing Logistics in Partnerships, Linkoping ¨ Studies in Management and Economics, Dissertations No.34, Linkoping ¨
Andreewsky E, Bourcier D (2000) Abduction in language interpretation and law making. Kybernetes 29(7-8):836–45
Anglin PM, Arnott R (1991) Residential real estate brokerage as a principal-agent problem. J R Estate Finance Econ 4(2):99–125
Anthonisse JM (1971) The rush in a directed graph, Technical Report BN 9/71. Stichting Mathematisch Centrum, Amsterdam
Araujo L, Easton G (1996) Networks in socioeconomic systems. In: Iacobucci D (ed) Networks in marketing. Sage, Thousand Oaks, pp 63–107
Armstrong JS, Overton TS (1977) Estimating nonresponse bias in mail surveys. J Market Res 14(3):396–402
Arndt K, Kinkeldei W (1978) Neue Wege der Luftfrachtabfertigung: Dezentralisierung und flankierende Maßnahmen zur Kostensenkung, Doctoral Thesis at Universitat Dortmund ¨
Arnold MA (1992) The principal-agent relationship in real estate brokerage services. J Am R Estate Urban Econ Assoc 20(1):89–106
Arnott R, Stiglitz JE (1988) The basic analytics of moral hazard. Scand J Econ 90(3):383–413

Arnott R, Stiglitz JE (1991) Moral hazard and nonmarket institutions: dysfunctional crowding or peer monitoring? Am Econ Rev 81(1):179–190

Arrow KJ (1963a) Uncertainty and the welfare economics of medical care. Am Econ Rev 53(5):941–973
Arrow KJ (1963b) Research in management controls: A critical synthesis. In: Bonini C, Jaediche R, Wagner H (eds) Management controls: new directions in basic research. McGraw-Hill, New York, pp 317–327
Arrow KJ (1969) The organization of economic activity: issues pertinent to the choice of market versus non-market allocation. In: The analysis and evaluation of public expenditures: the PBBsystem, Joint Economic Committee, 91st Congress, 1st Session, Vol.1, Washington D.C.
Arrow KJ (1985) The economics of agency. In: Pratt JW, Zeckhauser RJ (eds) The structure of business. Harvard Business School Press, Boston, pp 37–51
Arrow KJ (1986) Agency and the market. In: Arrow KJ, Intriligator MD (eds) Handbook of mathematical economics, vol III, North-Holland, Amsterdam, pp 1183–1195
Asariotis R, Hull H-J, Clarke MA, Herber R, Kiantou-Pampouki A, Moran-Bovio D, Ramberg J, ´ de Wit R, Zunarelli S (1999) Intermodal transportation and carrier liability, Office for Official Publications of the European Communities, Luxembourg
Ashford N, Stanton HPM, Moore CA (1997) Airport operations, 2nd edn. McGraw-Hill, New York Aspromourgos T (1991) 'neoclassical'. In: Eatwell J, Milgate M, Newman P (eds) The new palgrave: a dictionary of economics, vol 3 (Reprint). Macmillan, New York, p 625
Autry CW, Griffis SE (2005) A social anthropology of logistics research: exploring productivity and collaboration in an emerging science. Transport J 44(4):27–43
Autry CW, Griffis SE (2008) Supply chain capital: The impact of strucutral and relational linkages on firm execution and innovation. J Bus Logist 29(1):157–173
Avondo-Bodino G (1962) Economic applications of the theory of graphs. Blackie & Sons, London-Glasgow
Axelrod R (1984) The evolution of cooperation. Basic Books, New York
Axelsson B (1992) Network research future issues. In: Axelsson B, Easton G (eds) Industrial networks – a view of reality. Routledge, London, pp 237–251
Axelsson B, Easton G (1992) Industrial networks: a new view of reality. Routledge, London
AXS-Alphaliner (2010) AXS-Alphaliner cellular fleet forecast Nov. 2010, AXSMarine, Paris
Bachmeier S (1999) Integratoren Die Schnellen Dienste des Weltverkehrs, N¨urnberger Wirtschafts- und Sozialgeographische Arbeiten Bd. 53, Erlangen-N¨urnberg (also Doctoral Thesis at Universitat Erlangen-N¨ ¨ urnberg)
Backhouse R (1985) A history of modern economic analysis. Basil Blackwell, Oxford, New York Bader G (1959) Das Speditionswesen in der modernen Volkswirtschaft, Doctoral Thesis at Universitat Innsbruck ¨
Bagehot W (1971) The only game in town. Financ Analyst J 27(2):12–14, 22
Baird AJ (2002) Privatization trend at the world's top-100 container ports. Marit Pol Manag 29(3):271–284
Baird AF (2006) Logistics strategy in ocean container shipping. In: Heideloff C, Pawlik T (eds) Handbook of container shipping management, vol I, ISL, Bremen, pp 113–138
Baker WE (1984) The social structure of a national securities market. American Journal of Sociology 89(4):775–811
Baker GP, Hubbard TN (2000) Contractability and Asset Ownership: on-Board Computers and Governance in U.S. Trucking, NBER Working Paper No.7634. National Bureau of Economic Research, Cambridge
Baker GP, Hubbard TN (2002) Make versus buy in trucking: asset ownership, job design and information, NBER Working Paper No.8727. National Bureau of Economic Research, Cambridge
Baker GP, Hubbard TN (2003) Make versus buy in trucking: asset ownership, job design and information. Am Econ Rev 93(3):551–572
Balabanis GI (2000) Factors affecting export intermediaries' service offerings: the British example. J Int Bus Stud 31(1):83–99

Balabanis GI (2001) The relationship between diversification and performance in export intermediary firms. Br J Manag 12(1):67–84

Balabanis GI, Baker MJ (1993) Determinants and directions of trading companies' development in Europe. Eur J Market 27(9):58–76

Balderston FE (1958) Communication networks in intermediate markets. Manag Sci 4(2):154–171 Baliga S, Sjostr ¨ om T (1998) Decentralization and collusion. J Econ Theor 83(2):196–232 ¨

Baligh HH, Richartz LE (1967) Vertical market structure. Allyn & Bacon, Boston

Ball M, Barnhart C, Nemhauser G, Odoni A (2007) Air transportation: irregular operations and controll. In: Barnhart C, Laporte G (eds) Transportation. Handbook of OR & MS, vol 14, North Holland, Amsterdam, pp 1–67

Ballester C, Calvo-Armengol A, Zenou Y (2006) Who's who in networks. Wanted: the key player. ´ Econometrica 74(5):1403–1417

Bardhan P (1997) Corruption and development: a review of issues. J Econ Lit 35(3):1320–1346

Barkema HG, Bell HJ, Pennings JM (1996) Foreign entry, cultural barriers, and learning. Strat Manag J 17(2):151–166

Barkema HG, Vermeulen F (1997) What differences in the cultural background of partners are detrimental for international joint ventures? J Int Bus Stud 28(4):845–864

Barnard C (1938) The Functions of the executive. Harvard University Press, Cambridge

Barnes JA (1954) Class committees in a Norwegian island parish. Hum Relat 7(1):39–58

Barnes JA (1969) Graph theory and social networks: a technical comment on connectedness and connectivity. Sociology 3(2):215–232

Barnes JA (1972) Social networks. Addison-Wesley, Reading

Barnes JA (1979) Network analysis: orienting notion, rigorous technique, or substantive field of study? In: Holland PW, Leinhardt S (eds) Perspectives on social network research. Academic Press, New York, pp 401–423

Barney JB, Hesterly W (1996) Organizational economics: understanding the relationship between organizations and economic analysis. In: Clegg SR, Hardy C, Nord WR (eds) Handbook of organization studies. Sage, London; Thousand Oaks, New Delhi, pp 115–147

Barnhart C, Krishnan N, Kim D, Ware K (2002) Network design for express shipment delivery. Comput Optim Appl 21(3):239–262

Barnhart C, Cohn A (2004) Airline schedule planning: Accomplishments and opportunities. Manuf Serv Oper Manag 6(1):3–22

Baron DP (1982) A model of the demand for investment banking advising and distribution services for new issues. J Finance 37(4):955–976

Baron DP, Besanko D (1987) Monitoring, moral hazard, asymmetric information, and risk sharing in procurement contracting. Rand J Econ 18(4):509–532

Baron DP, Holmstrom B (1980) The investment banking contract for new issues under asymmetric ¨ information: delegation and the incentive problem. J Finance 35(5):1115–1138

Barton FL, McGehee RB (1942) Freight forwarders. Harv Bus Rev 20(3):336–347

Bartscher W (1976) Neukonzeption und Leistungsbemessung einer Container-Umschlagsanlage f¨ur den Binnenverkehr, Doctoral Thesis at Technische Hochschule Aachen

Bask AH (1999) Third party relationships in logistics services. Publications of the Helsinki School of Economics and Business Administration B-27, Helsinki

Bask AH (2001) Relationships among TPL providers and members of supply chains – a strategic perspective. J Bus Ind Market 16(6):470–486

Batt PJ (2008) Building social capital in networks. Ind Market Manag 37(5):487–491

Baumgarten H (1972) Uber technische und organisatorische M ¨ oglichkeiten zur Anpassung der ¨ Industriebetriebe an das Container-System, Doctoral Thesis at Technische Universitat Berlin ¨

Baumgarten H, Kasiske F, Zadek H (2002) Logistik-Dienstleister – Quo vadis ? – Stellenwert der Fourth Party Logistics Provider (4PL). Logistik Management 4(1):27–40

Baumgarten H, Zadek H (2002a) Struktur des Logistik-Dienstleistungsmarktes. In: Baumgarten H, Wiendahl H-P, Zentes J (eds) Logistik-Management. Springer, Heidelberg

Baumgarten H, Zadek H (2002b) Netzwerksteuerung durch Fourth-Party-Logistics-Provider (4PL). Jahrbuch der Logistik 2002. Verlagsgruppe Handelsblatt, D¨usseldorf, pp 14–20

Baumol WJ (1965) The stock market and economic efficiency. Fordham University Press, New York

Baumol WJ, Wolfe P (1958) A warehouse-location problem. Oper Res 6(2):252–263
Baumoller FW (1970) Der Einfluß des Containers auf die Verkehrstarife und die Verkehrstarifpoli- ¨ tik, Doctoral Thesis at Universitat zu Kiel ¨
Bavelas A (1948) A mathematical model for small group structures. Hum Organ 7:16–30
Bavelas A (1950) Communication patterns in task-oriented groups. J Acoust Soc Am 22(6): 725–730
Bazerman MH, Neale MA, Valley KL, Zajac EJ, Kim YM (1992) The effect of agents and mediators on negotiation outcomes. Int Behav Hum Decis Process 53(1):55–73
Beach WW, O'Driscoll Jr. GP (2003) Explaining the factors of the index of economic freedom. In: O'Driscoll Jr. GP, Feulner EJ, O'Grady MA (eds) 2003 Index of economic freedom. Heritage Foundation. Washington
Beamer-Downie D (1999) Freight forwarders' liability during international multimodal transportation. Thesis at McGill University, Montreal
Beauchamp MA (1965) An improved index of centrality. Behav Sci 10(2):161–163
Beckermann W (1956) Distance and the pattern of intra-European trade. Rev Econ Stat 38(1):31–40
Becker S (1999) Reorganisation der Luftfrachtkonsolidierung eine Analyse strategischer Allianzen von Luftfrachtspeditionen, Vandenhoeck & Ruprecht: Gottingen (also Doctoral ¨ Thesis at Universitat M¨ ¨ unster 1998)
Beddow M (2001) Take your partners. Containerization International No.11/2001, 57–61
Behrens U (1986) Landbr¨ucken als multimodale Transportsysteme und ihr Einfluß auf den Seetransport unter besonderer Ber¨ucksichtigung der Transsibirischen Containerroute, Doctoral Thesis at Universitat Rostock ¨
Beier FJ (1989) Transportation contracts and the experience effect: A framework for future research. J Bus Logist 10(2):73–89
Bellmore M, Nemhauser GL (1968) The traveling salesman problem: a survey. Oper Res 16(3):538–558
Bello DC, Williamson NC (1985a) Contractual arrangement and marketing practices in the indirect export channel. J Int Bus Stud 16(2):65–82
Bello DC, Williamson NC (1985b) The American export trading company: Designing a new international marketing institution. J Market 49(4):60–69
Benacchio M, Ferrari C, Musso E (2007) The liner shipping industry and EU competition rules. Transport Pol 14(1):1–10
Bender W (2000) Flughafen. In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom], Vahlen, Munich Beneviste LM, Marcus AJ, Wilhelm WJ (1992) What's special about the specialist? J Financ Econ 32(1):61–86
Benjamin JD, Jud GD, Sirmans G (2000a) What do we know about real estate brokerage? J R Estate Res 20(1-2):5–30
Benjamin JD, Jud GD, Sirmans G (2000b) Real estate brokerage and the housing market: an annotated bibliography. J R Estate Res 20(1-2):217–278
Benson D, Bugg R, Whitehead G (1994) Transport and logistics. Woodhead-Faulkner, Hertfordshire
Benston GJ, Smith jr. CW (1976) A transactions cost approach to the theory of financial intermediation. J Finance 31(2):215–231
Beplat K (1970) Probleme der Einf¨uhrung des Containerverkehrs. Verlag Weltarchiv, Hamburg
Berendt G (1961), Die Entwicklung der Marktstruktur im internationalen Luftverkehr. Duncker & Humblot, Berlin (also Doctoral Thesis at Universitat Hamburg) ¨
Berge C (1958) Theorie des graphes et ses applications. Dunod, Paris [English translation as: Berge ´ C (1962) The Theory of Graphs and its Applications, Methuen, London]
Berge C, Ghouila-Houri A (1962) Programmes, jeux et resaux de transport. Dunod, Paris [English ´ translation as: Berge C, Ghouila-Houri A (1965) Programming, games and transportation

networks. Methuen, London and Wiley, New York and German translation as: Berge C, Ghouila-Houri A (1967) Programme, Spiele, Transportnetze. Teubner, Leipzig]

Bergen M, Dutta S, Walker Jr. OC (1992) Agency relationships in marketing: a review of the implications and applications of agency and related theories. J Market 56(3):1–24
Berglund M (2000) Strategic positioning of the emerging third-party logistics providers. Linkoping ¨ Studies in Management and Economics, Dissertations No.45, Linkoping ¨
Berglund M, van Laarhoven P, Sharman G, Wandel S (1999) Third-party logistics: is there a future? Int J Logist Manag 10(1):59–70
Berle AA, Means GC (1932) Modern corporation and private property. Macmillan, New York
Bertsch L (2000a) The Air Cargo Tarif (TACT). In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, Munich
Bertsch L (2000b) Vertriebssysteme im Luftverkehr. In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, Munich
Bess IA (1996) Global logistics: a Study of restructuring in the international air express sector and its role in the distribution strategies of firms. Queen's University, Kingston
Bey I (1972) Simulationstechnische Analyse der Luftfrachtabfertigung, Doctoral Thesis at Universitat Karlsruhe ¨
Bhattacharya U, Yavas A (1993) In search of the right middleman. Econ Lett 42(4):341–347
Bhattacharya S, Lafontaine F (1995) Double-sided moral hazard and the nature of share contracts. Rand J Econ 26(4):761–781
Bichou K, Bell MGH (2007) Internationalisation and consolidation of the container port industry: assessment of channel structure and relationships. Marit Econ Logist 9(1):35–51
Biddle BJ (1986) Recent development in role theory. Annu Rev Sociol 12:67–92
Thomas EJ, Biddle BJ (1966a) The nature and history of role theory. In: Biddle BJ, Thomas EJ (eds) Role theory: concepts and research. Wiley, New York, pp 3–19
Thomas EJ, Biddle BJ (1966b) Basic concepts for classifying the phenomena of role. In: Biddle BJ, Thomas EJ (eds) Role theory: concepts and research. Wiley, New York, pp 23–45
Biebig P, Althof W, Wagnener N (2004) Seeverkehrswirtschaft (3rd rev. ed.). Oldenburg, Munich Bienstock CC, Mentzer JT (1999) An experimental investigation of the outsourcing decision for motor carrier transportation. Transport J 39(1):42–59
BIFA (2000) Best practice for the effective use of BIFA 2000 STCs. British International Freight Association (BIFA), Feltham, see: http://mipsweb.martex.uk.com/bifabeta/mips/a stcs/ pdfs/BestPrac.pdf
Biglaiser G (1993) Middlemen as experts. Rand J Econ 24(2):212–223
Biglaiser G, Friedman JW (1994) Middlemen as guarantors of quality. Int J Ind Organ 12(4): 509–531
Biglaiser G, Friedman JW (1999) Adverse selection with competitive inspection. J Econ Manag Strat 8(1):1–32
Bischoff EE, Ratcliff MSW (1995) Issues in the development of approaches to container loading. Omega 23(4):337–390
Bitran GR, Gurumurthi S, San SL (2007) The need for third party coordination in supply chain governance. Sloan Management Review 48(3):29–37
Bjelicic B (2000) Integrator. In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, Munich Blank R (1929) Die Kalkulation in der Binnenschiffahrts-Spedition, Doctoral Thesis at Technische Hochschule M¨unchen
Blankenburg D (1992) Kopplade Relationer i Industriella Natverk, Licentiate thesis, Uppsala ¨ University Department of Business Studies: Uppsala
Blattmann R (1977) Entwicklungsaspekte des Luftfrachtverkehrs unter besonderer Ber¨ucksichtigung des Luftfrachtpotentials im Bundesgebiet, Wahl: Karlsruhe (also Doctoral Thesis at Universitat Karlsruhe) ¨
Blau PM (1964) Exchange and power. Wiley, New York
Blau PM, Scott WR (1963) Formal organizations: a comparative approach. Routledge and Kegan Paul, London

Blettgen HG (2001) Die Regelung des multimodalen Transports nach dem Transportrechtsreformgesetz (TRG). In: Riedel E (ed) Multimodal Transport und Binnenschiffahrt. Nomos, Baden Baden, pp 53–63

Bloch F, Ryder H (2000) Two-sided search, marriages, and matchmakers. Int Econ Rev 41(1): 93–115

Blois KJ (1972) Vertical quasi-integration. J Ind Econ 20(3):253–272

Bohme H (2000a) Weltseeverkehr: Mit Zuversicht in das neue Jahrtausend, Kieler Diskussions- ¨ beitrage No.364/365, ¨ Institut f¨ur Weltwirtschaft Kiel: Kiel

Bohme H (2000b) Reederei. In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, Munich ¨ Bohme H (2000c) Seeschiffahrt. In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, ¨ Munich

Bohme H (2000d) Seeschiffahrtskonferenzen. In: Bloech J, Ihde GB (eds) Logistik total [CD- ¨ Rom]. Vahlen, Munich

Bolland JM (1988) Sorting out centrality: an analysis of the performance of four centrality models in real and simulated networks. Soc Network 10(3):233–253

Bollinger F ( ¨ 1930) Die Spedition in ihrem Verhaltnis zur Reichsbahn in der neusten Zeit ¨ unter besonderer Ber¨u cksichtigung des Wettbewerbskampfes der Eisenbahn mit den ¨ubrigen Verkehrsmitteln. Ludwig, Bonn (also Doctoral Thesis at Universitat zu K ¨ oln) ¨

Bolumole YA (2001) The supply chain role of third-party logistics service providers. Int J Logist Manag 14(2):93–107

Bolumole YA (2003) Evaluating the supply dhain role of logistics service providers. Int J Logist Manag 12(2):87–102

Bonacich P (1972a) Technique for analyzing overlapping memberships. Socio Meth 4:178–185

Bonacich P (1972b) Factoring and weighting approaches to status scores and clique identification. J Math Sociol 2:113–120

Bonacich P (1987) Power and centrality: a family of measures. Am J Sociol 92(5):1170–1182

Bonacich P (2007) Some unique properties of eigenvector centrality. Soc Network 29(3):555–564 Bonacich P, Lloyd P (2001) Eigenvector-like measures of centrality for asymmetric relations. Soc Network 23(3):191–201

Bonacich P, Lloyd P (2004) Calculating status with negative relations. Soc Network 26(4):331–338 Bonacich P, Oliver A, Snijders TAB (1998) Controlling for size in centrality scores. Soc Network 20(2):135–141

Bond EW, Gresik TA (2006) Efficient Delegation by an Informed Principal, working paper, see: http://www.nd.edu/˜tgresik/research.htm

Bontekoning YM, Macharis C, Trip JJ (2004) Is a new applied transport research field emerging? – A review of intermodal rail-truck freight transport literature. Transport Res Part A 38(1):1–34

Boorman SA, White HC (1976) Social structure from multiple networks. II. Role structures. Am J Sociol 81(6):1384–1446

Bopp W (1976) Wirtschaftlichkeitsberechnung nuklear angetriebener Containerschiffe: Faktoren, Verfahren, Modelle, Doctoral Thesis at Technische Universitat Braunschweig ¨

Border KC, Sobel J (1987) Samurai accountant: a theory of auditing and plunder. Rev Econ Stud 54(180):525–540

Borgatti SP (1995) Centrality and AIDS. Connections 18(1):112–114

Borgatti SP (2005) Centrality and network flow. Soc Network 27(1):55–71

Borgatti SP (2006) Identifying sets of key players in a social network. Comput Math Organ Theor 12(1):21–34

Borgatti SP, Carley KM, Krackhardt D (2006) On the robustness of centrality measures under conditions of imperfect data. Soc Network 28(2):124–136

Borgatti SP, Everett MG (2006) A graph-theoretic perspective on centrality. Soc Network 28(4):466–484

Borgatti SP, Foster PC (2003) The network paradigm in organizational research: a review and typology. J Manag 29(6):991–1013

Borgatti SP, Jones C, Everett MG (1998) Network measures of social capital. Connections 21(2):27–36

Borgatti SP, Li X (2009) On social network analysis in a supply chain context. J Supply Chain Manag 45(2):5–22

Bortfeld A (1995) Informierte Graphensuchverfahren und genetische Algorithmen zur Losung von ¨ Containerbeladeproblemen. Koester, Berlin (also Doctoral Thesis at Freie Universitat Berlin) ¨

Bortfeldt A (2000) Eine Heuristik f¨ur Multiple Containerladeprobleme. OR Spektrum 22(2): 239–261

Bortfeldt A, Gehrig H (1998) Ein Tabu Search-Verfahren f¨ur Containerbeladeprobleme mit schwach heterogenem Kistenvorrat. OR Spektrum 20(4):237–250

Bott E (1957) Family and social network. Travistock, London [2nd edn. as: Bott E (1971) Family and social network. Free Press, New York]

Bottler S (2006) Truck statt Turbine. Verkehrsrundschau 36:46–49

Bourdieu P (1980) Le capital social: notes provisoires. Actes de la Recherche en Sciences Sociales 31(2):2–3

Bourdieu P (1983) Okonomisches Kapital, kulturelles Kapital, soziales Kapital. In: Kreckel R (ed) ¨ Soziale Ungleichheiten – Soziale Welt. Schwartz, Gottingen, pp 183–198 ¨

Bourdieu P (1985) The forms of capital. In: Richardson JG (ed) Handbook of theory and research for the sociology of education. Greenwood, New York, pp 241–258

Bowen J, Leinbach T (2004) Market concentration in the air freight forwarding industry. Tijdschrift voor Economische en Sociale Geographie 95(2):174–188

Bowersox DJ, Cooper MB (1992) Strategic marketing channel management. McGraw-Hill, New York

Bowersox DJ, Closs DJ (1996) Logistical management: the integrated supply chain process. McGraw-Hill, New York

Boyd JH, Prescott EC (1986) Financial intermediary-coalitions. J Econ Theor 38(2):211–232

Bradach JL, Eccles RG (1989) Price, authority, and trust: from ideal types to plural forms. Annu Rev Sociol 15:97–118

Bramel J, Simchi-Levi D (1997) The logic of logistics: theory, algorithms, and apllications for logistics management. Springer, New York

Branch AE (1998) Maritime economics – marketing and management, 3rd edn. Stanley Thornes, Cheltenham

Branch AE (2000) Export practice and management, 4th edn. Thomson Learning, London

Brandes U, Erlebach T (2005) Fundamentals. In: Brandes U, Erlebach T (eds) Network analysis – methodological foundations. Springer, Heidelberg, pp 5–15

Brasch JJ (1978) Export management companies. J Bus Stud 9(1):59–71

Brasch JJ (1981) Using export specialists to develop overseas sales. Harv Bus Rev 59(3):6–8

Breid V (1995) Aussagefahigkeit agencytheoretischer Ans ¨ atze im Hinblick auf die Verhal- ¨ tenssteuerung von Entscheidungstra gern. Zeitschrift f¨ ¨ ur betriebswirtschaftliche Forschung 47(9):821–854

Breitzmann K-H (2000) Seehafen. In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, Munich

Bretzke W-R (1989) To buy or not to buy – Kriterien f ¨ur die Entscheidung zwischen Selbsterstellung und Fremdbezug logistischer Leistungen. Internationales Verkehrswesen 6:389–395

Bretzke W-R (1998) Der Einfluß von "Transaktionskosten" auf die Entscheidung ¨uber den Fremdbezug von logistischen Leistungen. In: Matschke MJ, Schildbach T (eds) Unternehmensberatung und Wirtschaftspr¨ufung. Schaffer-Poeschel, Stuttgart, pp 3–24 ¨

Briggs P (1997) Paperless solutions streamline exports. Can Plast 5:37–40

Broers P (1974) Die Strukturwandlungen in der Seeschiffahrt. Duncker und Humblot, Berlin (also Doctoral Thesis at Universitat Bonn 1973) ¨

Brooks MR (2004) The governance structure of ports. Rev Netw Econ 3(2):168–183

Brooks MR, Blunden RG, Bidgood CI (1993) Strategic alliances in the global container transport industry. In: Culpan R (eds) Multinational strategic alliances. Haworth Press, Binghampton, pp 221–248

Brothers LA (1954) Operations analysis in the United States air force. J Oper Res Soc Am 2(1): 1–16

Brown TA (1984a) Shippers' agents and the marketing of rail intermodal service. Transport J 23(3):44–52

Brown TA (1984b) Freight brokers and general commodity trucking. Transport J 24(4):4–14

Bryan DL, O'Kelly ME (1999) Hub-and-spoke networks in air transportation: an analytical review. J Reg Sci 39(2):275–295

Bryant J, Wallace N (1980) Open-market operations in a model of regulated, insured intermediaries. J Polit Econ 88(1):146–173

BSL (2000) Zahlen Daten Fakten aus Spedition und Logistik 2000, Bundesverband Speditions und Logistik e.V. (BSL), Bonn

Buchanan JM (1975) The limits of liberty: between anarchy and leviathan. University of Chicago Press, Chicago

Buchwald P (1973) Hauptprobleme des heutigen und k¨unftigen Luftverkehrs, Eigenverlag: Cologne - Vienna (also Doctoral Thesis at Hochschule f¨ur Welthandel Wien 1972)

Bumstead J, Cannons K (2002) From 4PL to managed supply-chain operations. Logist Transport Focus 4(4):18–23

Burckhardt PC, Elhence S, van Rooijen MCW (1998), European freight forwarders: which way to turn? McKinsey Quarterly 2:84–95

Burg G van den (1975) Containerisation and other unit transport, 2nd rev. edn. Hutchinson Benham, London

Buro T (2000) Gestaltung globaler Luftfracht-Transportsysteme mit Hilfe des OSTO-TCO Ansatzes, VDI-Verlag: D ¨usseldorf (also Doctoral Thesis Technische Hochschule Aachen)

Burt RS (1976) Positions in networks. Social Forces 55(1):93–122

Burt RS (1977a) Positions in multipe network systems, part one: a general conception of stratification and prestige in a system of actors cast as a social typology. Social Forces 56(1):106–131

Burt RS (1977b) Power in a social typology. Soc Sci Res 6(1):1–83

Burt RS (1980a) Models of network structure. Annu Rev Sociol 6:79–141

Burt RS (1980b) Autonomy in a social typology. Am J Sociol 85(4):892–925

Burt RS (1982) Toward a structural theory of action. Academic Press, New York

Burt RS (1983) Range, in: Burt RS, Minor MJ (eds.) Applied network analysis: a methodological introduction. Sage, Beverly Hills, pp 176–194

Burt RS (1992) Structural holes: the social structure of competition. Harvard University Press, Cambridge

Burt RS (1997) The contigent value of social capital. Admin Sci Q 42(2):339–365

Burt RS (1998) The gender of social capital. Rationality and Society 10:5–46

Burt RS (2000) The network structure of social capital. Res Organ Behav 22:345–423

[with the appendix hosted at: http://gsb.uchicago.edu/fac/ronald.burt/]

Button K (2001) Deregulation and liberalization of European air transport markets. Innovation 14(3):255–275

Button K (2005) Shipping Economics: where we are and looking ahead from an institutional economics perspective. Marit Pol Manag 32(1):39–58

Button K, Stough R (2000) Air transport networks: Theory and policy implications. Elgar, Cheltenham

Cahn AS (1948) The warehouse problem [Abstract]. Bull Am Math Soc 54(11):1073

Campbell TS, Kracaw WA (1980) Information production, market signalling, and the theory of financial intermediation. J Finance 35(4):863–882

Cao B (1991) Operations Research in der Container-Stellplatz-Verwaltung, Doctoral Thesis at Universitat Hamburg ¨

Carbone V, Stone MA (2005) Growth and relational strategies used by the European logistics service providers: rationale and outcomes. Transport Res Part E 41(6):495–510

Cardonne Molina MI (1981) Some considerations on the organization of the Internal Movement of Containers in the Republic of Cuba, Doctoral Thesis at Universitat Rostock ¨

Carr CMJ, Crum MR (1995) The U.S. Customs modernization and informed compliance act: implications for the logistics pipeline. Int J Logist Manag 6(2):67–81

Carter R, Hodgson GM (2006) The impact of empirical tests of transaction cost economics on the debate on the nature of the firm. Strat Manag J 27(5):461–476

Carter CR, Ellram LM, Tate W (2007a) The use of social network analysis in logistics research. J Bus Logist 28(1):137–168
Carter CR, Leuschner R, Rogers DS (2007b) A social network analysis of the Journal of Supply Chain Management. J Supply Chain Manag 43(2):15–28
Cartwright D, Harary F (1956) Structural balance: a generalization of Heider's theory. Psychol Rev 63(5):277–293
Casson M (1986) The role of vertical integration in the shipping industry. J Transport Econ Pol 20(1):7–29
Casson M (1990) Enterprise and competitiveness: a systems view of international business. Clarendon Press, Oxford
Casson M (1993) Internationalization theory and beyond. In: Dunning JH (ed) United nations library on transnational corporations, vol 1: the theory of transnational corporations. Routledge, London, pp 361–386
Casson M (1996) The nature of the firm reconsidered: information synthesis and entrepreneurial organisation. Manag Int Rev Special Issue 1:55–94
Casson M (1997a) Information and organization: a new perspective on the theory of the firm. Clarendon Press, Oxford
Casson M (1997b) Entrepreneurial networks in international business. Bus Econ Hist 26(2): 811–823
Casson M (1997c) Institutional economics and business history: a way forward? Bus Hist 34(4):151–171
Casson M (1998) The economic analysis of multinational trading companies. In: Jones G (ed) The multinational traders. Routledge, London, pp 22–47
Casson M (1999) The organisation and evolution of the multinational enterprise: an information cost approach. Manag Int Rev Special Issue 1:77–121
Casson M (2000a) The boundaries of firms: a global systems perspective. In: Casson M (ed) Economics in international business: a new research agenda. Elgar, Cheltenham, pp 58–93
Casson M (2000b) Entrepreneurship and the international business system: Developing the perspective of Schumpeter and the Austrian school. In: Casson M (ed) Economics in international business: a new research agenda. Elgar, Cheltenham, pp 228–253
Casson M (2000c) Entrepreneurial networks in international business. In: Casson M (ed) Economics in international business: a new research agenda. Elgar, Cheltenham, pp 254–277
Casson M, Cox H (1997) An economic model of inter-firm networks. In: Ebers M (ed) The formation of inter-organizational networks. Oxford University Press, Oxford, pp 174–196
Casson M, Wadeson N (1998) Communication costs and the boundaries of the firm. Int J Econ Bus 5(1):5–27
Casti JL (1995) The theory of networks. In: Batten D, Casti JL, Thord R (eds) Networks in action: communication, economics and human knowledge. Springer, Berlin, pp 3–24
Cazzaniga Francesetti D, Foschi AD (2001) The impact of 'Hub and Spokes' port networks on transport systems. http://papers.ssrn.com/sol3/papers.cfm?abstract id=275156
Cecchetti SG (1999) The future of financial intermediation and regulation: An overview. J Curr Issues Finance Bus Econ 5(8):1–5
Chan Y-S (1983) On the positive role of financial intermediation in allocation of venture capital in a market with imperfect information. J Finance 38(5):1543–1568
Chari VV, Jagannathan R (1988) Banking panics, information, and rational expectations equilibrium. J Finance 43(3):749–761
Charisse K (2000) Umzugsverkehr (M ´ obelverkehr). In: Bloech J, Ihde GB (eds) Logistik total ¨ [CD-Rom]. Vahlen, Munich
Charnes A, Cooper WW (1955) Generalizations of the warehousing model. Oper Res Q 6(4): 131–172
Cheung SNS (1983) The contractual nature of the firm. J Law Econ 26(1):1–21

Chevalier D (1989) Freight forwarders – How to use them in foreign trade. International Trade Forum 25(3):4–32

Child J, Ng SH, Wong C (2002) Psychic distance and internationalization Evidence of Hong Kong firms. Int Stud Manag Organ 32(1):36–56
Chinese Culture Connection (1987) Chinese values and the search for culture-free dimensions of culture. J Cross Cult Psychol 18(2):143–164
Cho DS (1984) The anatomy of the Korean general trading company. J Bus Res 12(2):241–255
Choi TY, Dooley KJ, Rungtusanatham M (2001) Supply networks and complex adaptive systems: control versus emergence. J Oper Manag 19(3):351–366
Choi TY, Kim Y (2008) Structural embeddedness and supplier management: a network perspective. J Supply Chain Manag 44(4):5–13
Choi TY, Wu Z (2009) Triads in supply networks: theorizing buyer-seller-supplier relationships. J Supply Chain Manag 45(1):8–25
Christiansen M, Fagerholt K, Ronen D (2004) Ship routing and scheduling: Status and perspectives. Transport Sci 38(1):1–18
Christiansen M, Fagerholt K, Nygreen B, Ronen D (2007) Maritime transportation. In: Barnhart C, Laporte G (eds) Transportation, Handbook of OR & MS vol 14. North Holland, Amsterdam, pp 189–284
Christie PMJ, Levary RR (1998) Virtual corporations. Ind Manag 7-8:7–11
Christofides N (1975) Graph theory: an algorithmic approach. Academic Press, New York
Chu W, Chu W (1994) Signaling quality by selling through a reputable retailer: An example of renting the reputation of another agent. Market Sci 13(2):117–189
Chung TY (1991) Incomplete contracts, specific investments, and risk sharing. Review of Economic Studies 58(5):1031–1042
Churchman CW, Ackoff RL, Arnoff EL (1957) Introduction to operations research. Wiley, NewYork
Cisic D, Kesic B, Jakomin L (2000a) Research in the power in the supply chain. Economics Working Paper Archive (WoPEc), http://econwpa.wustl.edu/eprints/it/papers/0004/0004002.abs
Cisic D, Komadina P, Kesic B (2000b) Investigating the power in freight transport communication network structure. Economics Working Paper Archive (WoPEc), http://ideas.repec.org/p/wpa/wuwpit/0004003.html
Clarke RI (1997) An analysis of the international ocean shipping conference system. Transport J 36(4):17–24
Clarke MA (2002) Multimodal transport in the new millennium. WMU J Marit Aff 1:71–84
Clemons EK, Hann I-H (1999) Rosenbluth international: strategic transformation of a successful enterprise. J Manag Inform Syst 16(2):9–27
Clott CB (2000) Ocean freight intermediaries: an analysis of non-vessel operating common carriers (NVOCC's) and maritime reform. Transport J 40(2):17–26
Coase RH (1937) The nature of the firm. Economica (New Series) 4(16):386–405
Coase RH (1960) The problem of social cost. J Law Econ 3(3):1–44
Coase RH (1990) The firm, the market, and the law. University of Chicago Press, Chicago
Cohen KJ, Maier SF, Schwartz RA, Witcomb DK (1979) Market makers and the market spread: a review of recent literature. J Financ Quant Anal 14(4):813–835
Coleman JS (1972) Systems of social exchange. J Math Sociol 2:145–163
Coleman JS (1973) The mathematics of collective action. Adline, Chicago
Coleman JS (1988) Social capital in the creation of human capital. Am J Sociol 94(Suppl): S94–S120
Coleman JS (1990) Foundations of social theory. Harvard University Press, Cambridge
Coleman JS, Katz E, Menzel H (1966) Medical innovation: a diffusion study. Bobbs-Merrill, Indianapolis
Collin S-O (1997) Financial intermediation through markets and organizations: an informationboundary argument for financial organizations. Scand J Manag 13(2):175–189
Commons JR (1924) The legal foundation of capitalism. University of Wisconsin Press, Madison Commons JR (1931) Institutional Economics. Am Econ Rev 21(4):648–657

Commons JR (1934) Institutional economics. Macmillan, New York

Conlisk J (1996) Why bounded rationality? J Econ Lit 34(2):669–700

Contractor FJ (1990) Contractual and cooperative forms of international business: Towards a unified theory of modal choice. Manag Int Rev 30(1):31–54

Conway P (2005a) Revolution delayed. Airline Business 1:46–49

Conway P (2005b) Dominant force. Airline Business 11:52–54

Cook KS (1982) Network structures from an exchange perspective. In: Marsden PV and Lin N (eds) Social structure and market analysis. Sage, Beverly Hills, pp 177–199

Cook KS, Emerson RM (1978) Power, equity and commitment in exchange networks. Am Socio Rev 43(5):721–739

Cook KS, Emerson RM, Gillmore MR, Yamagishi T (1983) The distribution of power in exchange networks: theory and experimental results. Am J Sociol 89(2):275–305

Cook KS, Whitmeyer JM (1992) Two approaches to social structure: exchange theory and network analysis. Annu Rev Sociol 18:109–127

Cook KS, Yamagishi T (1992) Power in exchange networks: a power-dependence formulation. Soc Network 14(3-4):245–265

Cooper R, Ross TW (1985) Product warranties and double moral hazard. Rand J Econ 16(1): 103–113

Copeland TE, Galai D (1983) Information effects on the bid-ask spread. J Finance 38(5): 1457–1469

Corsten H, Gossinger R (2000) Einf¨ ¨ u hrung in das Supply-Chain-Management. Oldenbourg, M¨unchen-Wien

Corsten D, Lenz M, Klose M (2002) Logistics service providers and information-based logistics services – an exploratory study. Logistik Management 4(1):45

Cosimano TF (1996) Intermediation. Economica 63:131–143

Crainic TG (2000) Service network design in freight transport. Eur J Oper Res 122(2):272–288

Crum MR (1985) The expanded role of motor freight brokers in the wake of regulatory reform. Transport J 24(4):5–15

Cullen T (2006) Germanic giants. Containerisation International 9:40–43

Cummins JD, Doherty NA (2006) The economics of insurance intermediaries. J Risk Insur 73(3):359–396

Curti H (2003) Agents and Brokers as Intermediaries: Their Regulation in Germany. German Working Papers in Law and Economics No.17, http://www.bepress.com/gwp/default/vol2003/ iss1/art17

Czerny AI, Mitusch K (2005) Cooperation and competition in the cargo liner shipping industry. Internationales Verkehrswesen 57(12):553–557

Daganzo CF (2005) Logistic Systems Analysis, 4th edn. Springer, Berlin

Dahrendorf R (1960) Homo sociologicus: Ein Versuch zu Geschichte, Deutung und Kritik der Kategorie der sozialen Rolle, 2nd edn. Westdeutscher Verlag, Koln ¨

D'Amato A, D'Amato J (1977) The changing role of the foreign freight forwarder. J Acad Manag Sci 5(Suppl):143–145

Damsgaard J (1999) Global logistics system Asia Co., Ltd. J Inform Tech 14(3):303–314

Dantzig GB (1951) Application of the simplex method to a transportation problem. In: Koopmans TC (ed) Activity analysis of production and allocation: proceedings of a conference. Cowles Commission Monograph No.13. John Wiley, New York, pp 359–373

Dantzig GB (1963) Linear programming and extensions. Princeton University Press, Princeton

Dantzig GB (2002) Linear programming. Oper Res 50(1):42–47

Dantzig GB, Fulkerson DR (1954) Minimizing the number of tankers to meet a fixed schedule. Nav Res Logist Q 1(3):217–222

Dantzig GB, Fulkerson R, Johnson B (1954) Solution of a large-scale traveling-salesman problem. J Oper Res Soc Am 2(4):393–410

Dantzig, GB; Ramser, JH (1959) The truck dispatching problem, Management Science 6(1), 80–91 Danzas Lotse (2002) Danzas-Lotse: Leitfaden f¨ur Logistik, eCommerce, Spedition, Land-, Luftund Seeverkehre, Zoll, Außenwirtschaft, Danzas Deutschland Holding GmbH, D¨usseldorf

Dautel P (2000) Trucking. In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, Munich Davidov WH, Malone MS (1993 [1992]) Das virtuelle Unternehmen. Campus, Frankfurt [German edn. of Davidov WH, Malone MS (1992) The virtual corporation. HarperCollins, New York]

David RJ, Han S-K (2004) A systematic assessment of the empirical support for transaction cost economics. Strat Manag J 25(1):39–58
Davies GJ (1981) The role of exporter and freight forwarder in the United Kingdom. J Int Bus Stud 12(3):99–108
Davis K (1966) Status and related concepts. In: Biddle BJ, Thomas EJ (eds) Role theory: concepts and research. Wiley, New York, pp 67–74
Davis JA (1967) Clustering and structural balance in graphs. Hum Relat 20(2):181–187
Davis JA, Leinhardt S (1972) The structure of positive interpersonal relations in small groups. In: Berger J (ed) Sociological theory in progress, vol 2. Houghton Mifflin, Boston, pp 218–251
De Alessi L (1980) The economics of property rights: a review of the evidence. Res Law Econ 2:1–47
Debrie J, Gouvernal E (2006) Intermodal rail in Western Europe: actors and services in a new regulatory environment. Growth Change 37(3):444–459
Dehn WH (1987) Qualitative Wachstumschancen der See- und Luftfrachtspedition in der interkontinentalen Distributionslogistik der Investitionsg¨uterindustrie, Vandenhoeck und Ruprecht: Gottingen (also Doctoral Thesis at Universit ¨ at M¨ ¨ unster 1986)
Dejax PJ, Crainic TG (1987) A review of empty flows and fleet management models in freight transportation. Transport Sci 21(4):227–248
De la Motte H (1998a) Das Transportrechtsreformgesetz mit Kurzkommentierung. In: Korf W (ed) Lorenz – Der G¨u terverkehr nach neuem Recht. DVZ, Hamburg, pp 106–163
De la Motte H (1998b) Allgemeine Deutsche Spediteursbedingungen. In: Korf W (ed) Lorenz Der G¨u terverkehr nach neuem Recht. DVZ, Hamburg, pp 163–233
Delos MD (1972) Absatzwirtschaftliche Uberlegungen beim Einsatz von Containern unter ¨ Ber¨ucksichtigung verschiedener Containerarten und Großen, Doctoral Thesis at Hochschule ¨ f¨ur Welthandel Wien
De Meza D (1998) Coase theorem. In: Newman P (ed) The new palgrave dictionary of economics and the law, vol 1. Macmillan, London, pp 270–282
Dempsey PS (2000) The law of intermodal transportation: what it was, what it is, what it should be. National Center for Intermodal Transportation, Denver. See: http://www.ie.msstate.edu/ ncit/RESEARCH.html
Demsetz H (1967) Toward a theory of property rights. Am Econ Rev 57(2):347–359
Demsetz H (1968) The cost of transacting. Q J Econ 82(1):33–53
Demski JS, Sappington DEM (1987) Delegated expertise. J Account Res 25(1):68–89
Demski JS, Sappington DEM (1991) Resolving double moral hazard problems with buyout agreements. Rand J Econ 22(2):232–240
Denis J-E, Mallette-LaFreniere H (1987) Trading companies and freight forwarders as export ` intermediaries. In: Rosson PJ, Reid D (eds), Managing export entry and expansion. Praeger, New York, pp 269–281
De Noble AF, Castaldi RM, Moliver DM (1989) Export intermediaries: small business perceptions of services and performance. J Small Bus Manag 27(2):33–41
Deo NF (1974) Graph theory with applications to engineering and computer science. Prentice-Hall, Englewood Cliffs
Dequech D (2003) Cognitive and cultural embeddedness: combining institutional economics an economic sociology. J Econ Issues 37(2):461–470
Detrez W (1973) Konzentration und Kooperation im ¨uberseeischen Containerverkehr unter wettbewerbspolitischen Gesichtspunkten, Doctoral Thesis at Universitat Marburg ¨
Diamond DW (1984) Financial intermediation and delegated monitoring. Rev Econ Stud 51(3):393–414
Diamond DW (1996) Financial intermediation as delegated monitoring: a simple example. Econ Q 82(3):51–68

Diamond DW, Dybvig PH (1983) Bank runs, deposit insurance, and liquidity. J Polit Econ 91(3):401–419

Diaz-Bone R (1997) Egozentrierte Netzwerkanalyse und familiale Beziehungssysteme. Deutscher Universit ats Verlag, Wiesbaden ¨
Diederich H (2000) Logistische Betriebe (Logistikunternehmen). In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, Munich
Diehl G (1973) Voraussetzung, Moglichkeiten und Realisierung des Computereinsatzes in Spedi- ¨ tionsbetrieben, Doctoral Thesis at Universitat Mannheim ¨
Diehl G (1996) Angebotsstrategien logistischer Dienstleister. In: Jahrbuch der G¨uterverkehrswirtschaft 1996. Servicegesellschaft f¨ur Spedition und Logistik, Frankfurton-Main, pp 76–81
Dietrich R (1929) Die Entwicklung des Dresdner Speditionsgewerbes. Werkgemeinschaft, Leipzig (also Doctoral Thesis at Technische Hochschule Dresden 1928)
Dietz H (1935) Die Organisation einer Speditionsunternehmung im G¨uterfernverkehr mit Kraftwagen. Mauser, Forchheim (also Doctoral Thesis at Technische Hochschule M¨unchen)
Dinger F (2004) The future of liner conferences in Europe. Peter Lang, Frankfurt-on-Main (also Doctoral Thesis at Universitat zu Kiel 2003) ¨
Diruf G (1999) Modelle und Methoden der Transportplanung. In: Weber J, Baumgarten H (eds) Handbuch Logistik. Springer, Berlin, pp 368–391
Dischinger A, Daigl H, Dreykorn W, Hiedl W, Rebhan U (2005) Speditionsbetriebslehre, 5th edn. Winklers, Darmstadt
Domschke W, Drexl A (1995) Logistik, Band 1: Transport, 4th edn. Oldenbourg, Munich-Vienna Domschke W, Drexl A (1996) Logistik, Band 3: Standorte, 4th edn. Oldenbourg, Munich-Vienna
Domschke W, Drexl A (1997) Logistik, Band 2: Rundreisen und Touren, 4th edn. Oldenbourg, Munich-Vienna
Domschke W, Drexl A (2002) Einf¨uhrung in Operations-Research, 5th edn. Springer, Berlin
Donndorf G (1963) Die Entwicklung des Speditionsgewerbes in Leipzig von 1800–1920, Doctoral Thesis at Hochschule f¨ur Verkehrswesen, Dresden
Dorrenbacher C (1990) Corporate reorganisation in the European transport and logistic sector in ¨ the 1990s. International Economics, vol 3, Lit Verlag, M¨unster
Dow D (2000) A note on psychological distance and export market selection. J Int Market 8(1): 51–64
Drechsler W (1988) Markteffekte logistischer Systeme: Auswirkungen von Logistik- und unternehmens¨ubergreifenden Informationssystemen im Logistikmarkt, Vandenhoeck & Ruprecht: Go ttingen (also Doctoral Thesis at Universit ¨ at M¨ ¨ unster)
Drews KH (2003) Zum anwendbaren Recht beim multimodalen Transport. Transportrecht 1:12–19 DSLV (2005) Zahlen Daten Fakten aus Spedition und Logistik 2005. Deutscher Speditions- und Logistikverband e.V. (DSLV), Bonn
Dubois A, Gadde L-E (2002) Systematic combining: an abductive approach to case research. Journal of Business Research 55(7) 553 560
Dunlop JT (1991) Arbitration. In: Eatwell J, Milgate M, Newman P (eds) The new palgrave: a dictionary of economics, Vol1 (Reprint). Macmillan, New York, pp 110–111
DVB (2001) The Global Air Cargo Industry, Deutsche Verkehrsbank (DVB) Industrial Research No.3/2001, Frankfurt-on-Main
DVB (2002) M&A Trends in Global Transportation, DVB Group Industrial Research: Frankfurt-on-Main
Dybvig PH, Lutz NA (1993) Warranties, durability, and maintenance: two-sided moral hazard in a continous-time model. Rev Econ Stud 60(204):575–597
Dyckhoff H, Finke U (1992) Cutting and packing in production and distribution. Physica, Heidelberg
Dye RA (1986) Optimal monitoring policies and agencies. Rand J Econ 17(3):339–350
Dyer JH (1996) Specialized supplier networks as a source of competitive advantage: evidence from the auto industry. Strat Manag J 17(4):271–299

Easley D, O'Hara M (1995) Market microstructure. In: Jarrow RA, Maksimovic V, Ziemba WT (eds) Handbooks in operations research and management science, vol 9: Finance. Elsevier, Amsterdam, pp 357–383

Easterwood JC, Morgan GE (1991) Eroding market imperfections, reintermediation and disintermediation. J Financ Res 16(4):345–358
Easton G (1992) Industrial networks: a review. In: Axelsson B, Easton G (eds) Industrial networks – A view of reality. Routledge, London, pp 3–27
Easton G, Araujo L (1994) Market exchange, social structures and time. Eur J Market 28(3):72–84 Eaton J (1994) The crazy economics of air freight. Intereconomics 29(1):33–37
Ebers M (2001) Interorganisational relationships and networks. In: Smelser NJ (ed) International encyclopedia of the social and behavioral sciences. Elsevier, Amsterdam, pp 7855–7860
Eccles RG (1981) The quasifirm in the construction industry. J Econ Behav Organ 2(4):335–357
Eckstein WE (1985) Kooperative Systembildung in der Transportwirtschaft: Innovationschancen mittelstandischer Betriebe. von Loeper, Karlsruhe (also Doctoral Thesis at Universit ¨ at¨ Karlsruhe)
Economides N (2001) The impact of the internet on financial markets. J Financ Transform 1(1): 8–13
Edlin AS, Hermalin BE (2000) Contract renegotiation and options in agency problems. J Law Econ Organ 16(2):395–423
Egert W (1972) Die volkswirtschaftliche Bedeutung des Container-Verkehrs f¨ur Europa unter vergleichender besonderer Ber¨ucksichtigung der Verhaltnisse in der Bundesrepublik Deutschland, ¨ in der Schweiz und in Osterreich, Doctoral Thesis at Hochschule f¨ ¨ ur Welthandel, Wien
Eggli W (1942) Das Basler Speditionsgewerbe. Sauerlander & Co., Aarau (also Doctoral Thesis at ¨ Universitat Basel) ¨
Ehret E (1951) Organisationsprobleme der Speditionsbetriebe, Doctoral Thesis at Universitat¨ M¨unchen
Eiras AI, Schaefer BD (2003) The 2003 index of economic freedom: the countries. In: O'Driscoll Jr. GP, Feulner EJ, O'Grady MA (eds) 2003 Index of economic freedom. Heritage Foundation, Washington
Eiselt HA, Sandblom C-L (2000) Integer programming and network models. Springer, Berlin
Eisenhardt KM (1989) Agency theory: an assessment and eeview. Acad Manag Rev 14(1):57–74 Elg U, Johansson U (1993) The institutions of industrial governance. Int Stud Manag Organ 23(1):29–46
Eller D (1996) The credit jigsaw. Containerization International 2:64–65
Eller D (2001) Liner agents chart survival. Containerization International 11(Suppl), 3
Ellis PD (2003) Social structure and intermediation: Market-making strategies in international exchange. J Manag Stud 40(7):1683–1708
Emerson RM (1962) Power-dependence relations. Am Socio Rev 27(1):31–41
Emerson RM (1972a) Exchange theory part I: a psychological basis for social exchange. In: Berger J, Zelditch M, Anderson B (eds) Sociological theory in progress. Houghton-Mifflin, Boston, pp 38–57
Emerson RM (1972b) Exchange theory part II: exchange relations and networks. In: Berger J, Zelditch M, Anderson B (eds) Sociological theory in progress. Houghton-Mifflin, Boston, pp 58–87
Emmermann M, Zadek H, Kieffer D (2003) Wem passt welche Jacke? Deutsche Verkehrszeitung 21(18.02.2003):26–27
Emons W (1988) Warranties, moral hazard, and the lemons problem. J Econ Theor 46(1):16–33 Encyclopædia Britannica (2002) Encyclopædia Britannica 2002 – Standard edition CD [CD-Rom]. britannica.co.uk, London
Engelsleben T (1999) Marketing f¨ur Systemsanbieter: Ansatze zu einem Relationship Marketing- ¨ Konzept f¨ur das logistische Kontraktgeschaft. Gabler, Wiesbaden (also Doctoral Thesis at ¨ Universitat zu K ¨ oln) ¨
Erdmann M (1999) Konsolidierungspotentiale von Speditionskooperationen: eine simulationsgest¨utzte Analyse. Gabler, Wiesbaden (also Doctoral Thesis at Universitat zu K ¨ oln) ¨

Erlei M, Leschke M, Sauerland D (1999) Neue Institutionenokonomik. Schaeffer-Poeschel, ¨ Stuttgart

Ernst E (1982) Fahrplanerstellung und Umlaufdisposition im Containerschiffsverkehr, Lang: Frankfurt-on-Main - Bern (also Doctoral Thesis at Technische Hochschule Aachen)
Esch E (1923) Die Geschaftsbedingungen der europ ¨ aischen Spediteure, insbesondere die ¨ Beschrankung ihrer Haftpflicht. Zeitschrift f¨ ¨ ur Verkehrswissenschaft 2(1):3–15
Etgar M, Zusman P (1982) The marketing intermediary as an information seller: A new approach. J Bus 55(4):505–515
Etschmaier MM, Mathaisel DF (1985) Airline scheduling: an overview. Transport Sci 19(2): 127–138
Eurostat (2003) Glossary for transport statistics, 3rd edn. Office for Offical Publications of the European Commission, Luxembourg
Evangelista P, Morvillo A (1999) Alliances in liner shipping: an instrument to gain operational efficiency or supply chain integration? Int J Logist Res Appl 2(1):21–38
Evan WM (1965) Toward a theory of inter-organizational relations. Manag Sci 11(10): B217–B230 Everett M, Borgatti SP (2005) Ego network betweenness. Soc Network 27(1):31–38
Ewert K (2006) Cooperation and concentration in the container shipping industry. In: Heideloff C,
Pawlik T (eds) Handbook of container shipping management, vol I. ISL, Bremen, pp 141–152 Exler M (1996) Containerverkehr: Reichweiten und Systemgrenzen in der Weltwirtschaft. Selbstverlag, Erlangen-N¨u rnberg (also Doctoral Thesis at Universitat Erlangen-N¨ ¨ urnberg)
Faber D (1996) The problems arising from multimodal transport. Lloyd's Maritime and Commercial Law Quarterly 503–518
Fafchamps M, Minten B (1999) Relationships and traders in Madagascar. J Dev Stud 35(6):1–35 Falk J (1995) Ein Multi-Agentensystem zur Transportplanung und -steuerung bei Speditionen mit
Trampverkehr. Infix, St. Augustin (also Doctoral Thesis at Universitat Erlangen-N¨ ¨ urnberg) Faller P (1999) Strategische Allianzen im Bereich Transport, Verkehr, Logistik. DVWG, Bergisch Gladbach
Fama EF (1980) Agency problems and the theory of the firm. J Polit Econ 88(2):288–307
Fama EF, Jensen MC (1983a) Separation of ownership and control. J Law Econ 26(2):301–325
Fama EF, Jensen MC (1983b) Agency problems and residual claims. J Law Econ 26(2):327–349
Faure-Grimaud A, Laffont J-J, Martimort D (2003) Collusion, delegation and supervision with soft information. Rev Econ Stud 70(2):253–279
Faynzilberg PS, Kumar P (2000) On the generalized principal-agent problem: Decomposition and existence results. Rev Econ Des 5(1):23–58
Ferguson AR, Dantzig GB (1955) The problem of routing aircraft a mathematical solution. Aeronaut Eng Rev 14(4):51–55
Ferguson AR, Dantzig GB (1956) The allocation of aircraft to routes an example of linear programming under uncertain demand. Manag Sci 3(1):45–73
Fernandez A, Arrunada B, Gonz ´ alez M (2000) Quasi-integration in less-than-truckload trucking. ´ In: Menard C (eds.) Institutions, contracts, and organizations: Perspectives from new institu- ´ tional economics. Elgar, Cheltenham-Northampton, pp 293-312
Ferry J (2004) Flextronics: staying real in a virtual world. Strat Bus 37(Winter Issue):64–73
Fershtman C, Gneezy U (2001) Strategic delegation: An experiment. Rand J Econ 32(2):352–368 Fershtman C, Judd KL (1987) Equilibrium incentives to oligopoly. Am Econ Rev 77(5):927–940
Fershtman C, Judd KL, Kalai E (1991) Observable contracts: strategic delegation and cooperation. Int Econ Rev 32(3):551–559
Fershtman C, Kalai E (1997) Unobserved delegation. Int Econ Rev 38(4):763–774
FIATA (1975) The Forwarder The Architect of Transport. Fed´ eration Internationale des ´ Associations de Transitaire et Assimiles (FIATA), Z¨ ´ urich
FIATA (2004) Subject: freight forwarders adopt an official description of "freight forwarding and logistics services". Doc. CL04/06 of 2004-10-29. see: http://www.fiata.com/index.php?id=116
Fingleton J (1997a) Competition among middlemen when buyers and sellers can trade directly. J Ind Econ 45(4):405–427

Fingleton J (1997b) Competition between intermediated and direct trade and the timing of disintermediation. Oxf Econ Paper 49(4):543–556

Fiotakis D (2005) The impact of information technology upon the shipbroking profession. Tulane Marit Law J 29:237–253

Fischer G, Reuther O, Mayer L (1930) Der G ¨uter- und Personenverkehr. Gloeckner, Leipzig

Flament C (1963) Applications of graph theory to group structure. Prentice-Hall, Englewood Cliffs Fleming DK (2002) Reflections on the history of US cargo liner service (Part I). Int J Marit Econ 4(4):369–389

Flood MM (1954) Application of transportation theory to scheduling a military tanker fleet. J Oper Res Soc Am 2(2):150–162

Flood MM (1956) The traveling-salesman problem. Oper Res 4(1):61–75

Foglar VP (2009) Schweizerisches transportrecht. Transportrecht 7-8:290–298

Ford R (2001) The value proposition in international freight: the contribution of the freight forwarder to the global logistics triad, PhD-Thesis at Cranfield University

Ford LR, Fulkerson DR (1956) Solving the transportation problem. Manag Sci 3(1):24–32

Ford LR, Fulkerson DR (1957) A primal-dual algorithm for the capacitated hitchcock problem. Nav Res Logist Q 4(1):47–54

Ford LR, Fulkerson DR (1962) Flows in networks. Princeton University Press, Princeton

Ford D, Hakansson H (2006) IMP – Some things achieved: much more to do. Eur J Market ˚ 40(3-4):248–258

Foreign Policy (2003) Measuring globalization: Who's up, who's down? Foreign Pol 134(January/February Issue):60–72

Forest J, Mehier C (2001) John R. Commons and Herbert A. Simon on the concept of rationality. J Econ Issues 35(3):591–605

Fortuin L and van Beek P (1996) Industrial applications. In: Gass SI, Harris CM (eds) Encyclopedia of operations research and management science. Kluwer, Boston, pp 295–299

Foster TA (1999) Who's in charge around here? Logist Manag Distrib Rep 38(6):61–64

Franck B, Bunel J-C (1991) Contestability, competition and regulation – The case of liner shipping. Int J Ind Organisat 9(2):141–159

Frankle A (2001) Untersuchungen zur Wettbewerbsf ¨ ahigkeit eines schienengebundenen ¨ Luftfrachtersatzverkehrs. Verlag Praxiswissen, Dortmund (also Doctoral Thesis at Universitat¨ Dortmund)

Frankel EG (1999) The economics of total trans-ocean supply chain management. Int J Marit Econ 1(1):61–69

Frankel EG (2002) The economics of international trade logistics and shipping transactions. In: Grammenos CT (ed) Handbook of maritime economics and business. LLP, London, pp 877-897 Frankel JA, Romer D (1999) Does trade cause growth? Am Econ Rev 89(3):379–399

Frankel JA, Rose AK (2000) Estimating the effect of currency unions on trade and output, NBER Working Paper Series No. 7857. National Bureau of Economic Research, Cambridge

Frankel JA, Rose A (2002) An estimate of the effect of common currencies of trade and income. Q J Econ 117(2):437–466

Freeman LC (1977) A set of measures of centrality based on betweenness. Sociometry 40(1):35–41 Freeman LC (1979) Centrality in social networks conceptual clarification. Soc Network 1(2): 215–239

Freeman LC (2004) The development of social network analysis. Empirical Press, Vancouver

Freeman LC, Borgatti SP, White DR (1991) Centrality in valued graphs: a measure of betweenness based in network flow. Soc Network 13(2):141–154

Freeman S (1996a) Clearinghouse banks and banknote over-issues. J Monetary Econ 38(1): 101–116

Freeman S (1996b) The payments systems, liquidity, and rediscounting. Am Econ Rev 86(5): 1126–1138

Freese S (1975) Die Funktionen des Speditionsbetriebes im Lichte des Allokationspostulats, Doctoral Thesis at Hochschule f¨ur Welthandel, Wien

Freichel SL (1992) Organisation von Logistikservice-Netzwerken: theoretische Konzeption und empirische Fallstudien. Schmidt, Berlin (also Doctoral Thesis at Technische Hochschule Darmstadt)

Fremont A (2009) Shipping lines and logistics. Transport Rev 29(4):537–554 ´
Fricke M, Dussoye SS (2000) Luftverkehrstechnik. In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, Munich
Friedkin, N.E. (1991) Theoretical foundations for centrality measures. Am J Sociol 96(6): 1478–1504
Friedman RA (1992) The culture of mediation: Private understandings in the context of public conflict. In: Kolb DM, Bartunek JM (eds) Hidden conflict in organizations: Uncovering behindthe-scenes disputes. Sage, Newbury Park, pp 143–164
Fritz A-H (1955) Wesen und Grenzen der Spedition als Verkehrsvermittler und Verkehrstrager, ¨ Doctoral Thesis at Universitat T¨ ¨ ubingen
Frye H (2003) Flachenbezogene Optimierung von Luftfrachtterminals. Verlag Praxiswissen, ¨ Dortmund (also Doctoral Thesis at Universitat Dortmund 2002) ¨
Fuhrmann R (1993) Logistiksysteme bei G¨u terverkehrsunternehmen: Gestaltungsansatze aus ¨ Sicht der Produktion von Dienstleistungen. Verlag Praxiswissen, Dortmund (also Doctoral Thesis at Universitat Dortmund 1992) ¨
Fuhrmann R (2000) Korrespondenzspediteur. In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, Munich
Fulconis F, Saglietto L, Pache G (2007) Strategy dynamics in the logistics industry: a transactional ´ center perspective. Manag Decis 45(1):104–117
Furubotn EG, Pejovich S (1972) Property rights and economic theory: a survey of recent literature. J Econ Lit 10(4):1137–1152
Fusfeld DR (1991) Methodenstreit. In: Eatwell J, Milgate M, Newman P (eds) The new palgrave: a dictionary of economics, vol 3 (Reprint). Macmillan, New York, pp 454–455
Gadde L-E, Huemer L, Hakansson H (2003) Strategizing in industrial networks. Ind Market Manag ˚ 32(5):357–364
Gadde L-E, Hulthen K (2009) Improving l ´ ogisitics outsourcing through increasing buyer-provider interaction. Ind Market Manag 38(5):633–640
Galaskiewicz J, Wasserman S (1993) Social network analysis Concepts, methology, and directions for the 1990s. Socio Meth Res 22(1):3–22
Gallai T (1990) Denes K ´ onig: a biographical sketch. In: K ¨ onig D (1990) Theory of finite and ¨ infinite graphs. Birkhauser, Boston, pp 423–426 ¨
Garella P (1989) Adverse selection and the middlemen. Economica 56:395–399
Garella PG, Peitz M (1999) Intermediation can replace certification, WP AD 99-04. Institutio Valenciano de Investigaciones Economicas, Valenica
Garella, PG, Peitz M (2000) Intermediation can replace certification. J Econ Manag Strat 9(1):1–24 Garman MB (1976) Market microstructure. J Financ Econ 3:257–275
Gass W (1991) Der Speditionsvertrag im internationalen Handelsverkehr unter besonderer Ber¨ucksichtigung der Deutschen Spediteurbedingungen (ADSp), der englischen Standard Trading Conditions (STC) und der franzoischen Conditions G ¨ en´ erales. Druckerei Brenner, ´ T¨ubingen (also Doctoral Thesis at Eberhardt-Karls-Universitat T¨ ¨ ubingen 1991)
Gass W (1999) Das neue Transport- und Speditionsrecht. Beck, Munich
Gattorna J (1998) Fourth-party logistics. In: Gattorna J (ed) Strategic supply chain alignment. Gower, Aldershot, pp 425–445
Gehrig T (1993) Intermediation in search markets. J Econ Manag Strat 2(1):97–120
Gehrig T (1996) Natural oligopoly and customer networks in intermediated markets. Int J Ind Organ 14(1):101–118
Gentry JJ (1996) The role of carriers in buyer-supplier strategic partnerships: a supply chain management approach. J Bus Logist 17(2):35–55
Gerckens C (1978) Containerisierung im Hamburger Hafen 1968–1975: strukturelle Auswirkungen der staatlichen und privaten Anpassung, Doctoral Thesis at Universitat Konstanz ¨

Gertler M (1988) Financial structure and aggregate economic activity: an overview. J Money Credit Bank 20(3):549–588

Geuther R (1990) Der Wettbewerb in der Containerlinienschiffahrt – Der Transportmarkt Bundesrepublik Deutschland – Fernost. Vandenhoeck & Ruprecht, Gottingen (also Doctoral Thesis ¨ at Universitat M¨ ¨ unster)

Ghemawat P (2001) Distance still matters – The hard reality of global expansion. Harv Bus Rev 79(8):137–147

Gibbons R (2005) Incentives between firms (and within). Manag Sci 51(1):2–17

Giddens A (1984) The constitution of society. Polity Press, Cambridge

Gillis C (2008) Past in the present. Am Shipper 6:80–81

Gilman S (1977) Optimal shipping technologies for routes to developing countries. J Transport Econ Pol 11(1):24–44

Globerman S, Roehl TW, Standifird S (2001) Globalization and electronic commerce: interferences from retail banking. J Int Bus Stud 32(4):749–768

Glosten LR, Milgrom PR (1985) Bid, ask and transaction prices in a specialist market with heterogenously informed traders. J Financ Econ 14(1):71–100

Glover F, Klingman D, Philips NV (1992) Network models in optimization and their applications in practice. Wiley, New York

Gluckman M (1955) The juridical process among the Barotse of Northern Rhodesia. Manchester University Press, Manchester

Gobel E (2002) Neue ¨ Institutioneno konomik. Lucius und Lucius, Stuttgart ¨

Goerres W (1909) Das deutsche Speditionsgesch aft im nationalen und internationalen Verkehre. ¨ Aachener Verlags- und Druckges, Aachen (also Doctoral Thesis at Universitat T¨ ¨ ubingen)

Goldberg VP (1976a) Toward an expanded economic theory of contract. J Econ Issues 10(1):45–61

Goldberg VP (1976b) Regulation and administered contracts. Bell J Econ 7(2):426–448 Gollnick R (1992) Organisatiorische Gestaltung von G¨uterkraftverkehrsbetrieben bei steigender Unternehmenskonzentration. Ducker und Humblot, Berlin (also Doctoral Thesis at Technische

Universitat Berlin) ¨ Gompf G (1994) Linien-Luftverkehrsgesellschaften in Luftfracht-Logistikketten: Strategien von Luftfracht-Linienverkehrsgesellschaften unter besonderer Ber¨u cksichtigung der Schnittstellenproblematik in Luftfrachtlogistikketten am Beispiel der Deutschen Lufthansa AG, Doctoral

Thesis at Universitat Giessen ¨ Gopfert I (1994) Interlini ¨ ng, Code-Sharing und Trucking. Wirtschaftswissenschaftliches Studium No.4/1994, 460–462

Gorton G, Pennacchi G (1988) Financial intermediaries and liquidity creation. J Finance 45(1): 49–71

Gorton G, Winton A (2002) Financial intermediation, NBER Working Paper No.8929. National Bureau of Economic Research, Cambridge

Goss RO (2002) An early history of maritime economics. Int J Marit Econ 4(4):390–404

Gould RV (1989) Power and social structure in community elites. Social Forces 68(2):531–552

Gould RV, Fernandez RM (1989) Structures of mediation: A formal approach to brokerage in transaction networks. Socio Meth 19:89–126

Grabellus, W. (1989) Bordmessungen des Leistungsbedarfs und Fahrtoptimierung eines Containerschiffes, Doctoral Thesis at Technische Universitat Berlin ¨

Graf G, Leßmann B (2000) Luftfrachtinformationssysteme. In: Bloech J, Ihde GB (eds) Logistik ¨ total [CD-Rom]. Vahlen, Munich

Graham B (1995) Geography and air transport. Wiley, Chichester

Graham A (2002) Managing Airports – An International Perspective, 2nd edn. Butterworth-Heinemann, Oxford

Gran D (1996) Probleme der Luftfrachtspedition. Transportrecht 4:138–142

Grandjot H-H (2000) Luftfracht. In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, Munich

Grandjot H-H (2002) Leitfaden Luftfracht, 2nd edn. Huss-Verlag, Munich

Grandori A, Soda G (1995) Inter-firm networks: antecedents, mechanisms and forms. Organ Stud 16(2):183–214

Granovetter M (1973) The strength of weak ties. Am J Sociol 78(6):1360–1380
Granovetter M (1982) The strength of weak ties. In: Marsden PV, Lin N (eds) Social structure and market analysis. Sage, Beverly Hills, pp 105–130
Granovetter M (1985) Economic action and social structure: the problem of embeddedness. Am J Sociol 91(3):481–510
Granovetter M (1992) Problems of explanation in economic sociology. In: Nohria N, Eccles E (eds) Networks and organizations. Harvard Business School Press, Boston, pp 25–56
Grass K (1927) Binnenschiffahrtspedition: Die Organisation und Betriebstechnik im Speditions-Großbetrieb der Rheinschiffahrt. Spaeth & Linde, Berlin (also Doctoral Thesis at Universitat¨ Frankfurt am Main 1928)
Gretzschel H (1968) Die Organisation von Sammelladungen f¨ur Außenhandelsg¨uter der DDR im Export und Import, Doctoral Thesis at Hochschule f¨ur Verkehrswesen "Friedrich List", Dresden
Griffis SE, Goldsby TJ, Cooper M (2003) Web-based and mail surveys: a comparison of response, data, and cost. J Bus Logist 24(2):237–258
Grigsby DW anf Bigoness WJ (1982) Effects of mediation and alternative forms of arbitration on bargaining behavior: a laboratory study. J Appl Psychol 67(5):549–554
Gronroos C (1999) Internationalization strategies for services. J Serv Market 13(4-5):290–297 ¨
Gross N, Mason WS, McEachern AW (1958) Explorations in role analysis. Wiley, New York
Grossman SJ, Hart OD (1983) An analysis of the principal-agent problem. Econometrica 51(1): 7–45
Grossman SJ, Hart OD (1986) The costs and benefits of ownership: a theory of vertical and lateral integration. J Polit Econ 94(4):691–719
Grossman SJ, Miller MH (1988) Liquidity and market structure. J Finance 43(3):617–637
Gubi E, Arlbjørn JS, Johansen J (2003) Doctoral dissertations in logistics and supply chain management: a review of Scandinavian contributions from 1990 to 2001. Int J Phys Distrib Logist Manag 33(10):854–885
Gudmundsson SV, Walczuck R (1999) The development of electronic markets in logistics. Int J Logist Manag 10(2):99–113
G¨unther H-O, Kim KH (2005) Preface: logistics control issues of container terminals and automated transportation systems. In: G¨unther HO, Kim KH (eds) Container Terminals and Automated Transport Systems – Logistics Control Issues and Quantitative Decision Support. Springer, Berlin
G¨ursel T (1996) Containerstauung und-sicherung bei unterschiedlichen Containerschiffstypen. Koster, Berlin (also Doctoral Thesis Technische Universit ¨ at Berlin) ¨
Guesnerie R, Laffont J-J (1984) A complete solution to a class of principal-agent problems with an application to the control of a self-managed firm. J Publ Econ 25(3):329–369
Guesnerie R, Picard P, Rey P (1988) Adverse selection and moral hazard with risk neutral agents. Eur Econ Rev 33(4):807–823
Guhl T, Koller A, Schnyder AK, Druey JN (2000) Das Schweizer Obligationenrecht, 9th edn. Schulthess, Z¨urich
Gupta S, Romano RE (1998) Monitoring the principal with multiple agents. Rand J Econ 29(2):427–442
Gurley JG, Shaw E (1955) Financial aspects of economic development. Am Econ Rev 45(4): 515–538
Gurley JG, Shaw E (1956) Financial intermediaries and the saving-investment process. J Finance 11(2):257–276
Gutthal S (1999) Chancen und Risiken des Expressgutmarktes. DUV, Wiesbaden (also Doctoral Thesis at Technische Universitat Berlin) ¨
Gwartney J, Lawson R (2003) Economic freedom of the world 2003 Annual report. Fraser Institute, Vancouver
Habib M, Zurawicki L (2002) Corruption and foreign direct investment. J Int Bus Stud 33(2): 291–307

Hackett SC (1992) A comparative analysis of merchant and broker intermediation. J Econ Behav Organ 18(3):299–315

Hadjikhani A, Hakansson H (1996) ˚ Political actions in business networks a Swedish case. Int J Res Market 13(5):431–447
Hadjikhani A, Ghauri PN (1998) The behavior of international firms in socio-political environments in the European Union. J Bus Res 52(3):263–275
Hagg I, Johanson J (1982) F ¨ oretag i n ¨ atwerk [in Swedish]. SNS, Stockholm ¨
Haigh RW (1994) Thinking of exporting? Export management companies would be the answer. Columbia J World Bus 29(4):66–81
Hakansson H (1982) International marketing and purchasing of industrial goods An interaction ˚ approach. John Wiley, Chichester
Hakansson H, Ford D (2002) How should companies interact in business networks? J Bus Res ˚ 55(1):133–139
Hakansson H, Johanson J (1992) A model of industrial networks. In: Axelsson B, Easton G (eds) ˚ Industrial networks – A view of reality. Routledge, London, pp 28–34
Hakansson H, Johanson J (1993) The network as a governance structure Interfirm cooperation ˚ beyond markets and hierarchies. In: Grabher G (ed) The embedded firm – On the socioeconomics of industrial networks. Routledge, London, pp 35–51
Hakansson H, Snehota I (1995) Developing business relationships. Routledge, London ˚
Hald K (1951) Handbuch des Speditionsgewerbes, 2nd edn. Stoytscheff, Nuremberg
Halinen A, Tornroos J- ¨ A(1998) The role of embeddedness in the evolution of business networks. ˚ Scand Manag J 14(3):187–205
Hall ET (1960) The silent language in overseas business. Harv Bus Rev 5-6:87–96
Hall ET (1976) Beyond culture. Anchor Press, New York
Hall RW (1989) Configuration of an overnight package air network. Transport Res A 23A(2): 139–149
Halldorsson A, Skjøtt-Larsen T (2003) Dynamics of Relationship Governance in TPL Arrange- ´ ments – A Dyadic Perspective, Work-in-Progress-Paper presented at 15th Annual Conference for Nordic Researchers in Logistics, Oulu 12-13th June 2003
Halldorsson A, Kotzab H, Mikkola JH, Skjøtt-Larsen T (2004) Designing and managing supply ´ chains: theoretical foundation and application. In: Aronson H (ed) NOFOMA 2004, challenging boundaries with logistics. UniTryk, Linkoping, pp 247–262 ¨
Halldorsson A, Kotzab H, Mikkola JH, Skjøtt-Larsen T (2005) How inter-organisational theories ´ contribute to supply chain management. In: de Koster R, Delfmann W (eds) Supply chain management – European perspectives. Copenhagen Business School Press, Copenhagen, pp 37–61
Halldorsson A, Kotzab H, Mikkola JH, Skjøtt-Larsen T (2007) Complementary theories to supply ´ chain management. Supply Chain Manag Int J 13(4):284–296
Hamilton WH (1919) The institutional approach to economic theory. Am Econ Rev 9(1-Suppl):309–318
Hammarkvist KO, Hakansson H, Mattsson L-G (1982) Marknadsf ˚ oring f ¨ or konkurrenskraft ¨ [in Swedish]. Liber, Malmo¨
Hanlon P (2007) Global airlines: competition in a transnational industry, 3rd edn. Butterworth-Heinemann, Oxford
Hanna JB, Maltz A (1998) LTL expansion into warehousing: a transaction cost analysis. Transport J 38(2):5–17
Hansen G (1928) Die Entwicklung und Bedeutung des Kieler Speditionsgewerbes, Doctoral Thesis at Universitat Kiel ¨
Harary F, Norman RZ, Cartwright D (1965) Structural models. Wiley, New York
Haralambides HE, Cariou P, Benacchio M (2002) Costs, benefits and pricing of dedicated container terminals. Int J Marit Econ 4(1):21–34
Harland CM, Lamming RC, Zheng J, Johnsen TE (2001) A taxonomy of supply networks. J Supply Chain Manag 37(4):21–27

Harris M, Raviv A (1978) Some results on inventive contracts with applications to education and employment, health insurance, and law enforcement. Am Econ Rev 68(1):20–30

Harris M, Raviv A (1979) Optimal incentive contracts with imperfect information. J Econ Theor 20(2):231–251

Harris TE, Ross FS (1955) Fundamentals of a method for evaluating rail net capacities, Research Memorandum RM-1573, The Rand Corportation, Santa Monica

Hart O, Moore J (1988) Incomplete contracts and renegotiation. Econometrica 56(4):755–785

Hart O, Moore J (1990) Property rights and the nature of the firm. J Polit Econ 98(6):1119–1158

Hausmaninger H (2000) The Austrian legal system, 2nd edn. Manz, Wien

Hausser R und Abele E (2003) Aktuelle Probleme der Speditionsversicherung. Transportrecht 1: ¨ 8–12

Hautau U (2002) Seeverkehrsmarkte im wettbewerbspoltischen Wandel: dargestellt am ¨ Beispiel der Container-Linienschiffahrt auf der Nordatlantik- und Fernost-Route. Lang, Frankfurt-on-Main (also Doctoral Thesis at Universitat Bremen 2002) ¨

Havila V (1996) International Business-Relationship Analysis, Doctoral Thesis No.64, Department of Business Studies. Uppsala University, Uppsala

Havila V, Johanson J, Thilenius P (2004) International business-relationship triads. Int Market Rev 21(2):172–186

Hayek FA v (1937) Economics and knowledge. Economica (New Series) 4:35–54

Hayek FA v (1960) The constitution of liberty. Routledge & Kegan Paul, London

Hayek FA v (1973) Law, legislation, and liberty. Routledge, Chicago

Heaver TD (1992) The role of customs administration in the structure and efficiency of international logistics: an international comparison. Int J Logist Manag 3(1):63–72

Heaver TD (2002a) The evolving roles of shipping lines in international logistics. Int J Marit Econ 4(3):210–230

Heaver TD (2002b) Supply chain and logistics management: implications for liner shipping. In: Grammenos CT (ed) Handbook of maritime economics and business. LLP, London, pp 375–396

Heaver T, Meersman H, Moglia F, van de Voorde E (2000) Do mergers and alliances influence European shipping and port competition? Marit Pol Manag 27(4):363–373

Heaver T, Meersman H, van de Voorde E (2001) Co-operation and competition in international container transport: strategies for ports. Marit Pol Manag 28(3):293–305

Hebeler H (1998) Is box standardsation promoting or limiting intermodal growth? Hansa 135(9):260–263

Hebert O (1992) Strukturwandel der Spedition unter marktwirtschaftlichen Bedingungen in den neuen Bundeslandern der BRD, Doctoral Thesis at Hochschule f¨ ¨ ur Verkehrswesen, Dresden

Heckmann M (2002) DV-gest¨utztes Gescha ftsprozeßmanagement in der Luftfrachtlogistik: ¨ Konzeption eines Informationssystems f¨ur eine unternehmens¨ubergreifende Planung und Steuerung im Luftfrachtersatzverkehr. Shaker, Aachen (also Doctoral Thesis at Universitat zu ¨ Koln 2001) ¨

Hector B (1987) Aus der Entwicklungsgeschichte der Kurier-, Paket- und Expreßdienste. In: Transportmarkte: Kurier,- Expreß- und Paketdienste. Spiegel-Verlagsreihe M ¨ arkte im Wandel, ¨ Band 13, pp 8–21

Heider F (1946) Attitudes and cognitive organizations. J Psychol 21(1):107–112

Heider F (1958) The psychology of interpersonal relations. Wiley, New York

Heierli S (1995) Strategisches Management der Informatik als Erfolgsfaktor Schweizer Speditionsbetriebe: empirische U berpr¨ ¨ ufung eines Kausalmodells. Shaker, Aachen (also Doctoral Thesis at Hochschule St. Gallen 1991)

Hein A-P (2007) Consolidation and batching strategies for international transportation chains. Verlag Praxiswissen, Dortmund (also Doctoral Thesis at Universitat Dortmund) ¨

Heinrich R (1913) Der Spediteur in der modernen deutschen Volkswirtschaft, Doctoral Thesis at Universitat Heidelberg ¨

Heinrich HD (1966) Die wirtschaftliche Entwicklung und Bedeutung des Weltluftfrachtverkehrs – Unter besonderer Ber¨ucksichtigung des grenz¨uberschreitenden Luftfrachtverkehrs der Bundesrepublik Deutschland, Doctoral Thesis at Universitat zu K ¨ oln ¨

Heinzl A, Brandt A (1999) Simulationsmodelle. In: Weber J, Baumgarten H (eds) Handbuch logistik. Springer, Berlin, pp 392–411

Helf A (2005) Entscheidungsorientiertes Speditionsmarketing dargestellt am Beispiel des Marktes f¨ur Gefahrgut-Logistik. Driesen, Taunusstein (also Doctoral Thesis at Universitat¨ Bremen)
Hellberg R, Sannes R (1991) Customs clearance and electronic data interchange A study of Norwegian freight forwarders using EDI. Int J Prod Econ 24(1–2):91–101
Hellwig MF (2000) Financial intermediation with risk aversion. Rev Econ Stud 67(234):719–742 Herber R (1998) Die Neuregelung des deutschen Transportrechts. Neue Jurist Wochenschr 51:3297–3376
Herber R (1999) Seehandelsrecht: systematische Darstellung. de Gruyter, Berlin
Herrebosch M (2008) De economische impact van de wetgeving inzake samenwerkingsverbanden in de lijnvaart. Eur Transport Law 43(1):17–60
Hertz S (1993) The Internationalization Process of Freight Transport Companies Towards a Dynamic Network Model of Internationalization, Doctoral Thesis at Stockholm School of Economics
Hertz S (1996a) The dynamics of international strategic alliances. Int Stud Manag Organ 26(2):104–130
Hertz S (1996b) Drifting closer and drifting away in networks. In: Iacobucci D (ed) Networks in marketing. Sage, Thousand Oaks, pp 179–204
Hertz S, Alfredsson M (2003) Strategic development of third party logistics providers. Ind Market Manag 32(2):139–149
Hess T (1999) Implikationen der Prinzipal-Agent-Theorie f¨ur das Management von Unternehmensnetzwerken, Arbeitsbericht Nr.3/1999, Institut f¨ur Wirtschaftsinformatik, Abteilung Wirtschaftsinformatik 2, Georg-August-Universitat G ¨ ottingen ¨
Hess SW, Wong-Martinez CG (1996) Politics. In: Gass SI, Harris CM (eds) Encyclopedia of operations research and management science. Kluwer, Boston, pp 496–498
Heuer K (2003) Die Allgemeinen Deutschen Spediteursbedingungen nach dem Ableben der Speditionsversicherung gema ß Ziff. 29 ADSp. Transportrecht 1:1–8 ¨
Hildebrand W-C (2008) Management von Transportnetzwerken im containerisierten Seehafenhinterlandverkehr, Univ.-Verlag der Technischen Universitat: Berlin (also Doctoral Thesis at ¨ Technische Universitat Berlin) ¨
Hill RL (1988) The third-party role in intermodal transportation. In: McKeon JE (ed) Partnerships: A natural evolution in logistics. Leaseway Transportation Corp., Cleveland, pp 87–96
Hill WL (1990) Cooperation, opportunism, and the invisible hand: Implications for transaction cost theory. Acad Manag Sci 15(3):500–513
Hinricher M (1993) Die Zukunftschancen kleiner und mittlerer Seehafen. Vandenhoeck & ¨ Ruprecht, Gottingen (also Doctoral Thesis at Universit ¨ at M¨ ¨ unster)
Hinterhuber HH, Levin BM (1994) Strategic networks the organization of the future. Long Range Plann 27(3):43–53
Hirsch-Kreinsen H (2002) Unternehmensnetzwerke revisited. Zeitschrift f¨ur Soziologie 31(2):106–124
Hitchcock FL (1941) The distribution of a product from several sources to numerous localities. J Math Phys 20:224–230
Ho T, Stoll HR (1980) On dealer markets under competition. J Finance 35(2):259–268
Hobbs JE (1996) A transaction cost approach to supply chain management. Supply Chain Management: An International Journal 1(2):15–27
Hodgson GM (1998) The approach to institutional economics. J Econ Lit 36(1):166–192
Hodgson GM (2003) John R Commons and the foundation of institutional economics. J Econ Issues 37(3):547–576
Hoff G (1972) Entwicklung und Bedeutung des Luftfrachtverkehrs in der Bundesrepublik Deutschland. Duncker und Humblot, Berlin (also Doctoral Thesis at Universitat Bonn) ¨
Hoppel P (1925) Die Spe ¨ dition und ihre Stellung im Wirtschaftsleben, Doctoral Thesis at Universitat Rostock ¨

Hofstede G (1980) Culture's consequences: International differences in work-related values. Sage, Beverly Hills

Hofstede G (1991) Cultures and organizations: Software of the mind. McGraw-Hill, London

Hofstede G (2001) Culture's consequences: Comparing values, behaviors, insitutions, and organizations arcoss nations. Sage, Thousand Oaks

Hofstede G (2002) Dimensions do not exist: A reply to Brendan McSweeney. Hum Relat 55(11):1355–1361

Hofstede G, Bond MH (1988) The confucius connection: From cultural roots to economic growth. Organ Dynam 16(4):5–21

Holland PW, Leinhardt S (1970) A method of detecting structure in sociometric data. Am J Sociol 76(3):492–513

Holland PW, Leinhardt S (1971) Transitivity in structural models of small groups. Comp Groups Studies 2(2):107–124

Holland PW, Leinhardt S (1972) Holland and Leinhardt reply: Some evidence on the transitivity of positive interpersonal sentiment. Am J Sociol 77(6):1205–1209

Holmstrom B (1979) Moral hazard and observab ¨ ility. Bell J Econ 10(1):74–91

Holmstrom B, Roberts J (1998) The boundaries of the firm revisited. J Econ Perspect 12(4):73–94 ¨

Holmstrom B, Tirole J (1998) Private and public supply of liquidity. J Polit Econ 106(1):1–40 ¨ Holmen E, Pedersen A-C (2000) Avoiding Triadic Reductionism: Serial Tetrads – An Useful Concept for Studying Connected Relationships? paper presented at the 16th IMP-Conference, Bath, see: http://www.impgroup.org/uploads/papers/67.pdf

Holtappels P (2006) Zum Rechtsverhaltnis zwischen dem Hamburger Containerschiffsmakler und ¨ seinem deutschen Prinzipal. Transportrecht 5:184–191

Homans GC (1950) The human group. Harcourt, Brace & World, New York

Homans GC (1961) Social behaviour: Its elementary forms. Routledge and Kegan Paul, London Ho T, Stoll HR (1981) Optimal dealer pricing und transactions and return uncertainty. J Financ Econ 9(1):47–73

Howard DG, Maskula JM (1988) Will American export trading companies replace traditional export management companies? Int Market Rev 5(4):41–50

Nordbank HSH (2007) Sector report international air cargo. HSH Nordbank AG, Hamburg-Kiel

Hubbard TN (1998) Why are process monitoring technologies valuable? The use of on-board information technology in the trucking industry. NBER Working Paper No.6482. National Bureau of Economic Research, Cambridge

Hubbard TN (1999) How wide is the scope of hold-up-based theories? Contractual form and market thickness in trucking, NBER Working Paper No.7347. National Bureau of Economic Research, Cambridge

Hubbard TN (2000) The demand for monitoring technologies: the case of trucking. Q J Econ 115(2):533–560

Hubbard TN (2001) Contractual form and market thickness in trucking. Rand J Econ 32(2): 369–386

Hubbel CH (1965) An input-output approach to clique identification. Sociometry 28(4):377–399 Huch J (1973) Leertransporte im Containerverkehr, Doctoral Thesis at Universitat Mannheim ¨

Hume D ([1739/40] 1978) A treatise of human nature: being an attempt to introduce the experimental method of reasoning into moral subjects, Reprint, 2nd edn. Oxford University

Press, New York Hummel HJ, Sodeur W (1987) Strukturbeschreibung von Positionen in sozialen Beziehungsnetzen. In: Pappi FU (ed) Techniken der empirischen Sozialforschung: Methoden der Netzwerkanalyse.

M ¨unchen, Oldenbourg, pp 177–202 Hummon NP, Carley K (1993) Social networks as normal science. Soc Network 15(1):71–106

Hutchison TW (1984) Institutional economics old and new. J Inst Theor Econ 140(1):20–29

IATA (2004) European air cargo programme handbook, Launch edn. International Air Transport Association, Geneva

Ihde G (1989) Wettbewerbsstrategien der Spedition im Spannungsfeld zwischen (De)Regulierung und Strukturwandel. Zeitschrift f¨ur Verkehrswissenschaft 60:129–140

IMD (2003) IMD world competitiveness yearbook 2003. Institute for Management Development (IMD), Lausanne

Inglehart R (2000) Globalization and postmodern values. Wash Q 23(1):215–22
Inglehart R, Baker WE (2000) Modernization, cultural change, and the persistence of traditional values. Am Socio Rev 65(1):19–51
Inglehart R, Basanez M, Moreno A (1999) Human values and beliefs a cross-cultural sourcebook. ˜ University of Michigan Press, Ann Arbor
Iri M (1969) Network flow, transportation and scheduling. Academic Press, New York-London
Irwin DA, Tervio M (2000) Does trade raise income? Evidence from the twentieth century, NBER ¨ Working Paper Series No.7745. National Bureau of Economic Research: Cambridge
Isermann H (1995) Verkehrsintegration als logistische Aufgabe. Internationales Verkehrswesen 47(10):602–608
ITZ (2002) Ladungskontrolle: Alles sehen, alles wissen. Internationale Transport Zeitschrift No.13/2002, 36–37
Jack SL, Anderson AR (2002) The effects of embeddedness on the entrepreneurial process. J Bus Venturing 17(5):467–487
Jacklin CJ, Bhattacharya S (1988) Distinguishing panic and information-based bank runs: welfare and policy implications. J Polit Econ 96(3):568–592
Jackman RW (2001) Social capital. In: Smelser NJ (ed) International encyclopedia of the social and behavioral sciences. Elsevier, Amsterdam, pp 14216–14218
Jacob E (1996) KECIS ein kostenorientiertes Entscheidungs- und Controllingsystem im Speditionswesen, Shaker: Aachen (also Doctoral Thesis at Universitat Saarbr¨ ¨ ucken)
Jaeger G, Laudel H (2003) Transportmanagement: Die Fachkunde des G¨uterverkehrs f¨ur Spedition, Handel und Industrie Band 1: Lehrbuch, Feldhaus: Hamburg
Jaggi F (2000) Gestaltungsempfehlungen f¨ ¨ ur Hub-and-Spoke-Netzwerke im europaischen ¨ Luftverkehr – ein resourcenbasierter Ansatz, Difo-Druck: Bamberg (also Doctoral Thesis at Universitat St. Gallen) ¨
Jansen D (2003) Einf¨uhrung in die Netzwerkanalyse: Grundlagen, Methoden, Forschungsbeispiele, Leske und Budrich: Opladen
Jansson JO, Shneerson D (1982) Port economics. MIT Press, Cambridge
Jarillo JC (1988) On strategic networks. Strat Manag J 9(1):31–41
Jarillo JC (1993) Strategic networks creating the borderless organization. Butterworth-Heinemann, Oxford
Jarillo JC, Ricart JE (1987) Sustaining networks. Interfaces 17(5):82–91
Jarke M (1981) Uberwachung und Steuerung von Container-Transportsystemen, Gabler: ¨ Wiesbaden (also Doctoral Thesis at Universitat Hamburg 1980) ¨
Jehle C-U (1981) Umschlagoptimierung in Containerterminals bei vier- und mehrlagiger Stapelung, Duncker & Humblot: Berlin (also Doctoral Thesis at Technische Universitat Berlin ¨ 1980)
Jennings B, Holcomb MC (1996) Beyond containerization: the broader concept of intermodalism. Transport J 35(3):5–13
Jensen MC (1983) Organization theory and methodology. Account Rev 58(2):319–339
Jensen MC, Meckling WH (1976) Theory of the firm Managerial behaviour, agency costs and ownership structure. J Financ Econ 3(4):305–360
Johannisson B (1987) Beyond process and structure: social exchange networks. Int Stud Manag Organ 17(1):3–23
Johanson J, Mattsson L-G (1987) Interorganizational relations in industrial systems: a network approach compared with the transaction-cost approach. Int Stud Manag Organ 17(1):34–48
Johanson J, Mattsson LG (1988) Internationalisation in industrial systems a network approach. In: Hood N, Vahlne J-E (eds) Strategies in global competition. Croom Helm: Beckenham, pp 287–214
Johanson J, Mattsson L-G (1992) Network positions and strategic action an analytical framework. In: Axelsson B, Easton G (eds) Industrial networks: a new view of reality. Routledge, London, pp 204–261

Johanson J, Vahlne J-E (1977) The internationalization process of the firm: a model of knowledge development and increasing foreign commitments. J Int Bus Stud 8(1):23–32

Johanson J, Wiedersheim-Paul F (1975) The internationalization of the firm four Swedish cases. J Manag Stud 12(2):305–322
Johnson J (1995) Links, arrows, and networks: fundamental metaphors in human thought. In: Batten D, Casti JL, Thord R (eds) Networks in action: communication, economics and human knowledge. Springer, Berlin, pp 25–48
Johnson DF, Pruitt DG (1972) Preintervention effects of mediation versus arbitration. J Appl Psychol 56(1):1–10
Johnson JC, Schneider KC (1995a) Licensed transportation brokers: their joys and frustrations. Transport J 34(4):38–45
Johnson JC, Schneider KC (1995b) Outsourcing in distribution: the growing importance of transportation brokers. Bus Horizon 38(6):40–48
Johnson JC, Wood DF (1996) Contemporary logistics, 6th edn. Prentice-Hall, New York
Jones G (1998) Multinational trading companies in history and theory. In: Jones G (ed) The multinational traders. Routledge, London, pp 1–21
Jones WB, Cassady CR, Bowden Jr. RO (2000) Developing a standard definition of intermodal transportation. Transport Law J 27:345–352
Jones C, Hesterly WS, Borgatti SP (1997) A general theory of network governance: exchange conditions and social mechanisms. Acad Manag Rev 22(4):911–945
Jones GR, Pustay MW (1988) Interorganizational coordination in the airline industry, 1925–1938: a transaction cost approach. J Manag 14(4):529–546
Juhel MH (2001) Globalisation, privatisation and restructuring of ports. Int J Marit Econ 3(2): 139–174
Jung C (1999) Strategische Allianzen in der Verkehrswirtschaft. Zeitschrift f¨ur Verkehrswissenschaft 70(1):61–86
Jungnickel D (1994) Graphen, Netzwerke und Algorithmen, 3rd edn. BI-Wissenschafts-Verlag, Mannheim
Kah A (1993) Profitcenter-Steuerung: ein Beitrag zur theoretischen Fundierung des Controlling anhand des Principal-Agent-Ansatzes, Schaffer-Poeschel: Stuttgart ¨
Kahler HD (1975) Das Konzept des sozialen Netzwerks: Eine Einf¨ ¨ uhrung in die Literatur. Zeitschrift f¨ur Soziologie 4(3):283–290
Kaltenbrunner R (2005) Die Bedeutung des Internets f¨ur die traditionelle Luftfrachtspedition, Schriftenreihe des Instituts f¨ur Transportwirtschaft und Logistik Nr.2 (2005 VER), Institut f¨ur Transportwirtschaft und Logistik: Wien
Kaluza B, Dullnig H, Malle F (2003) Principal-Agent-Probleme in der Supply Chain Problemanalyse und Diskussion von Losungsvorschl ¨ agen, Diskussionsbeitr ¨ age des Instituts f¨ ¨ ur Wirtschaftswissenschaften der Universitat Klagenfurt No.2003/03 ¨
Kambhu J (1982) Optimal product quality under asymmetric information and moral hazard. Bell J Econ 13(2):483–492
Kantorovich LV (1939) Matematicheskie methody organizatsii i planirovaniya proizvodstva, Leningradskii Universitet: Leningrad [in Russian, translated into English and reprinted (1960) ´ as: Mathematical methods of organizing and planning production. Manag Sci 6(4):366–422]
Kantorovich LV (1942) O peremeshchenii mass, Doklady Akademii Nauk SSSR 37(7-8):227–230 [in Russian, English translation (1942) as: On the translocation of masses, Comptes Rendus (Doklady) de l'Academie des Sciences de l'U.R.S.S. No.37, 199–201 and reprinted (1958) in ´ Manag Sci 5(1):1–4]
Kantorovich LV, Gavurin MK (1949) Primenenie matematicheskikh metodov v voprosakh analiza gruzopotokov. In: Problemy povysheniya effectivnosti raboty transporta, Akademiia Nauk SSSR: Moscow-Leningrad, pp 110–138 [in Russian]
Kaselau R (1971) Das optimale Transportkonzept in der Linienseeschiffahrt: Entscheidungstheo- ¨ retische Analyse und Kostenvergleiche, Vandenhoeck und Ruprecht: Gottingen (also Doctoral ¨ Thesis at Universitat Hamburg) ¨

Kaspar C (1999) Strategische Allianzen im Verkehr als wichtige Instrumente des Kooperationsmanagements. O sterreichische Zeitschrift f¨ ¨ ur Verkehrswissenschaft 45(1-2):25–31

Kasprzak B (1997) Losung des Containerstauproblems mit Graphensuchverfahren und genetischen ¨ Algorithmen, P und H: Berlin (also Doctoral Thesis at Freie Universitat Berlin) ¨

Katz L (1953) A new status index drived from sociometric analysis. Psychometrika 18(1):39–43 Katz MJ (1991) Game-playing agents: unobservable contracts as precommitments. Rand J Econ 22(3):307–328

Kehagiaras AD (2004) NVOCC: secret agent? U.S.F. Marit Law J 17(2):207–240

Kehrberger HP (1996) IATA und IATA Agentur – Rechtliche und organisatorische Struktur. Transportrecht No.4/1996, 131–138

Keller M (1966) Betriebswirtschaftliche Faktoren der Wettbewerbsfahigkeit von Speditionsun- ¨ ternehmungen mit eigenem Fahrzeugeinsatz, Doctoral Thesis at Universitat Freiburg, Fribourg ¨ Kephart WM (1950) A quantitative analysis of intragroup relationships. Am J Sociol 55(6): 544–549

Kim WC (1986) Global diffusion of the general trading company concept. Sloan Manage Rev 27(4):35–42

Kindleberger CP (1993) Multinational ownership of shipping activities. In: Sauvant KP, Mallampally M (eds) United nations library on transnational corporations vol 12: transnational corporations in services. Routledge, London, pp 168–185

Kirchner PH (1950) Die Bedeutung der internationalen Spedition f¨ur die osterreichische ¨ Wirtschaft, Doctoral Thesis at Hochschule f¨ur Welthandel Wien

Kirschnick P (1969) Der Wandel in der o konomischen Bedeutung der großen europ ¨ aischen ¨ Seehafen im 20. Jahrhunde ¨ rt, Institut f¨ur Weltwirtschaft: Kiel (also Doctoral Thesis at Universitat zu Kiel) ¨

Kirzner IM (1991) Austrian school of economics. In: Eatwell J, Milgate M, Newman P (eds) The new palgrave: a dictionary of economics, vol 1 (Reprint). Macmillan, New York, pp 145–151

Klampferer W (1996) Einsatz lokaler Suchalgorithmen und genetischer Algorithmen bei der Luftfrachtcontainerbeladung, Doctoral Thesis at Technische Universitat Wien ¨

Klaus P (1999) Logik der Fusionen? Logistikmanagement 1(2):109–121

Kleer M (1991) Gestaltung und Kooperationen zwischen Industrie- und Logistikunternehmen: Ergebnisse theoretischer und empirischer Untersuchungen, Schmidt Verlag: Berlin (also Doctoral Thesis at Technische Hochschule Darmstadt 1990/91)

Klein B (1980) Transaction cost determinants of "unfair" contractual arrangements. Am Econ Rev 70(2):356–362

Klein B (1998) Hold-up problem. In: Newman P (ed) The new palgrave dictionary of economics and the law, vol 2. Macmillan, London, pp 241–244

Klein B, Crawford RG, Alchian AA (1978) Vertical integration, appropriable rents, and the competititve contracting process. J Law Econ 21(1):297–326

Klein B, Leffler KB (1981) The role of market forces in assuring contractual performance. J Polit Econ 89(4):615–641

Klippel B (2000) Non vessel operator (NVO). In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, Munich

Kloetz W (1969) Aufgaben und rechtliche Gestaltung der Tatigkeit der sozialistischen interna- ¨ tionalen Spedition der DDR unter Ber¨ucksichtigung des internationalen Verkehrs mit westeurop aischen kapitalistischen L ¨ andern, Doctoral Thesis at Universit ¨ at Halle, Halle an der Saale ¨

Knee R (2001) Mergers in transit. Logist Manag Distrib Rep 40(9):55–58

Knepper L (1978) Einsatz und Auslegung von Hochlagersystemen f¨ur Container, dargestellt am Beispiel einer Seehafenumschlagsanlage, Doctoral Thesis at Technische Hochschule Aachen

Knight FH (1921) Risk, uncertainty and profit. Houghton Mifflin Company, Boston, New York

Knoke D, Burt RS (1983) Prominence. In: Burt RS, Minor MJ (eds) Applied network analysis: a methodological introduction. Sage, Beverly Hills, pp 195–222

Knoke D, Kuklinski JH (1982) Network analysis. Sage Publications, Newbury Park

Knoke D, Yang S (2008) Social network analysis, 2nd edn. Sage, Thousand Oaks

Knudsen V (1998) Tiefgreifende Veranderungen im Schiffsmaklergewerbe. In: Probleme im Seev- ¨ erkehr – 6. Kieler Seminar zu aktuellen Fragen der See- und K¨ustenschiffahrt, Schriftenreihe der Deutsche Verkehrswissenschaftlichen Gesellschaft e.V. (DVWG) No.B207, 150–157

Kobelnig E (1952) Die historisch-wirtschaftliche Entwicklung und die daraus resultierende Funktionen des Speditionsgewerbes, Doctoral Thesis at Universitat Innsbruck ¨
Koc¸kesen L, Ok EA (2004) Strategic delegation by unobservable incentive contracts. Rev Econ Stud 71(2):397–424
Koberlein C (1997) Verkehrslexikon. Oldenbourg, Munich ¨
Kogut B, Singh H (1988) The effect of national-culture on the choice of entry mode. J Int Bus Stud 19(3):411–432
Kohler W (1921) In ¨ telligenzpr¨ufungen an Menschenaffen, Springer: Berlin [1st ed. in English as: Kohler W (1925) The men ¨ tality of apes. Harcourt, New York]
Kojima K, Ozawa T (1993) Japan's general trading companies: merchants of economic development. In: Sauvant KP, Mallampally M (eds), United nations library on transnational corporations vol 12: transnational corporations in services. Routledge, London, pp 361–386
Kolb DM (1983) The Mediators. MIT-Press, Cambridge
Kolmogoroff AN (1933) Sulla determinazione empirica di una legge di distribuzione. Giornale dell' Istitute Italiano degli Attuari vol 4, 83–91
Kolmogoroff AN (1941) Confidence limits for an unkown distribution function. Ann Math Stat 12(4):461–463
Konig D (1936) Theorie der endlichen und unendlichen Graphen, Akademische Verlagsge- ¨ sellschaften, Leipzig [2nd ed. as: Konig D (1950) Theorie der endlichen und unendlichen ¨ Graphen, Chelsea: New York and English translation with commentary by Tutte WT as: Konig ¨ D (1990) Theory of finite and infinite graphs. Birkhauser, Boston] ¨
Koopmans TC (1949) Optimum utilization of the transportation system. Econometrica 17(Suppl), 136–146
Koopmans TC (1951) Introduction. In: Koopmans TC (ed) Activity analysis of production and allocation: proceedings of a conference, Cowles Commission Monograph No.13. Wiley, New York, pp 1–12
Koopmans TC, Reiter S (1951) A model of transportation. In: Koopmans TC (ed) Activity analysis of production and allocation: proceedings of a conference, Cowles Commission Monograph No.13. Wiley, New York, pp 222–259
Kosch¨utzki D, Lehmann KA, Peeters L, Richter S, Tenfelde-Podehl D, Zlotowski O (2005) Centrality indices. In: Brandes U, Erlebach T (eds) Network analysis – methodological foundations. Springer, Heidelberg, pp 16–61
Kotowitz Y (1990) Moral hazard. In: Eatwell J, Milgate M, Newman P (eds) The new palgrave: allocation information and markets. Macmillan, London, pp 207–213
Kovacs G, Spens KM (2005) Abductive reasoning in logistics research. Int J Phys Distrib Logist ´ Manag 35(2):132–144
Kowtha NR (1997) Skills, incentives, and control An integration of agency and transaction cost analysis. Group Organ Manag 22(1):53–86
Krackhardt D (1992) The strength of strong ties: the importance of philos in organizations. In: Nohria N, Eccles RG (eds) Networks and organizations: structure, form, and action. Harvard Business School Press, Cambridge, pp 216–239
Krajewska MA (2008) Potentials for efficency increase in modern freight forwarding. Gabler, Wiesbaden (also Doctoral Thesis at Universitat Bremen 2007) ¨
Kramer H (2004) Privatfinanzierungen von Container-Terminalinfrastrukturen als Alternative zu staatlichen Finanzierungen: Entwicklungen und deren Ursachen im Containerhafensektor in der Europaischen Union, ISL: Bremen (also Doctoral Thesis at Universit ¨ at Bremen) ¨
Kranz W (1973) Die Dynamik des Containerverkehrs und Moglichkeiten seiner Steuerung mit ¨ Datenverarbeitungsanlagen, Doctoral Thesis at Hochschule f¨u r Verkehrswesen, Dresden
Krasa S, Villamil AP (1992) Monitoring the monitor: an incentive structure for a financial intermediary. J Econ Theor 57(1):197–221

Krass R (1984) Kooperation zwischen Verlader und Spedition: Konzept zur Effizienzsteigerung der Marketing-Logistik, Toeche-Mittler: Darmstadt (also Doctoral Thesis at Technische Hochschule Darmstadt, 1983)

Krebs J (1974) Die unterschiedlichen Entwicklungstendenzen im Frachtversand der deutschen Flughafen und ihre Bestimmungsfaktoren, Doctoral Thesis at Universit ¨ at Freiburg, Freiburg ¨ im Breisgau

Krejci H (2005) Handelsrecht, 3rd edn. Manz, Vienna

Kreps D (1998) Bounded rationality. In: Newman P (ed) The new palgrave dictionary of economics and the law vol 1. Macmillan, London, pp 168–173

Kreuter H (1931) Der Luftpost- und Luftfrachtverkehr unter besonderer Ber¨ucksichtigung der Deutschen Handelsluftfahrt, Bernecker: Melsungen (also Doctoral Thesis at Universitat Frank- ¨ furt am Main)

Krieger W (1984) Computergest¨utzte Auswahl interkontinentaler Distributionsverfahren, Duncker und Humblot: Berlin (also Doctoral Thesis at Technische Universitat Berlin 1983) ¨

Kronman AT (1985) Contract law and the state of nature. J Law Econ Organ 1(1):5–32

Krugman P (1995) Growing world trade: causes and consequences. Brookings Paper Econ Activ 1:327–377

Kruskal WH (1952) A nonparametric test for the several sample problem. Ann Math Stat 23(4):525–540

Kruskal WH, Wallis WA (1952) Use of ranks in one-criterion analysis of variance. J Am Stat Assoc 47(260):583–621 [errata ibid 48(264):907–911]

Kuby MJ, Gray RG (1993) The hub network design problem with stopovers and feeders: the case of Federal Express. Transport Res A, 27A(1):1–12

Kuhn HW (1955) The Hungarian method for the assignment problem. Nav Res Logist Q 2(1-2):83–97

K¨uhnbaum H (1976) Vervollkommnung der Informationsbeziehungen zur Sicherung der Leistungsprozesse im internationalen Containerverkehr der Mitgliedslander des RGW, Doctoral ¨ Thesis at Hochschule f¨ur Verkehrswesen "Friedrich List", Dresden

K¨uhner M (2002) Fourth party service provider. Supply Chain Manag 2(3):21–25

K¨urbs F (1920) Der deutsche Speditionsverkehr nach Frankreich, Univ Doctoral Thesis at Universitat zu K ¨ oln ¨

Kummer S (2006) Einf¨uhrung in die Verkehrswirtschaft. UTB: Wien

Kummer S, Schramm H-J, Sudy I (2010) Internationales Transport- und Logistikmanagement (2nd edn.), UTB: Wien

Kunz M (1999) Entb¨undelter Zugang zu Flughafen: Zur Liberalisierung der Bodenverkehrsdienste ¨ auf europaischen Flugh ¨ afen. Zeitschrift f¨ ¨ ur Verkehrswissenschaft 70(2):206–232

Kunzmann M (1988) Ein Beitrag zur Optimierung des Containerumschlages in Seehafen beim ¨ Einsatz frei beweglicher Flurf orderzeuge, Institut f¨ ¨ ur Fordertechnik: Karlsruhe (also Doctoral ¨ Thesis Universitat Karlsruhe) ¨

Kylaheiko K, Cisic D, Komadina P (2000) Application of transaction costs to choice of transport corridors, Economics Working Paper Archive (WoPEc), see: http://netec.mmc.ac.uk/ WoPEc/data/Papers/wpawuwpit0004001.html

Kyle AS (1984) Market structure, information, futures markets and price formation. In: Storey GG, Schmitz A, Sarris A (eds) International agricultural trade: advanced reading in price formation, market structure, and price instability. Westview Press, Boulder, pp 45–64

Kyle AS (1985) Continuous auctions and insider trading. Econometrica 53(6):1315–1336

Laage-Hellman J (1989) Technological Development in Industrial Networks, PhD thesis, Uppsala University, Department of Business Studies: Uppsala

Laarhoven P van, Berglund M, Peters M (2000) Third-party logistics in Europe – five years later. Int J Phys Distrib Logist Manag 30(5):425–442

Laarhoven P van, Sharman G (1994) Logistics alliances: the European experience. McKinsey Quarterly 1:39–49

Ladany SP (1996) Sports. In: Gass SI, Harris CM (eds) Encyclopedia of operations research and management science. Kluwer, Boston, pp 639–641

Laffont J-J, Martimort D (1998) Collusion and delegation. Rand J Econ 29(2):280–305
Laffont J-J, Martimort D (2002) The theory of incentives. Princeton University Press, Princeton
Laffont JJ, Maskin ES (1990) The efficient market hypothesis and insider trading on the stock market. J Polit Econ 98(1):70–93
Lafontaine F, Masten SE (2002) Contracting in the Absence of Specific Investments and Moral Hazard: Understanding Carrier-Driver Relations in US Trucking, University of Michigan, John M. Ohlin Center for Law & Economics, Paper No.02-002, Ann Arbor
Lai KH, Cheng TCE (2004) A study of the freight forwarding industry in Hong Kong. Int J Logist Res Appl 7(2):71–84
Lal R (1990) Improving channel coordination through franchising. Market Sci 9(4):299–318
Lambert DM, Cooper MC, Pagh JD (1998) Supply chain management: implementation issues and research opportunities. Int J Logist Manag 9(2):1–19
Lambert DM, Stock JR (1999) Strategic logistics management, 3rd edn. Irwin, Chicago
Lambsdorff J Graf (2003) Framework Document 2003: Background Paper to the 2003 Corruption Perception Index, Transparency International and University of Passau, see: http://www. transparency.org/cpi/index.html#cpi
Lambsdorff J Graf (2004) Corruption Perception Index 2003. In: Transparency International (ed) Global Corruption Report 2004. Pluto Press, London, pp 282–286
Lammich K (2000) Frachtdokumente. In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, Munich
Lamming R, Johnsen T, Zheng J, Harland C (2000) An initial classification of supply networks. Int J Prod Manag 20(6):675–691
Lamprecht H (1969) Wettbewerbsbeschra nkungen der gewerblichen Verkehrswirtschaft ¨ Osterreichs, Doctoral Thesis at Universit ¨ at Innsbruck 1970 ¨
Langen PW de, Welters HWH (1999) Innovations in the freight transport market. In: Faller P (eds) Transportwirtschaft im Umbruch. Linde, Vienna, pp 169–181
Langlois RN, Cosgel MM (1993) Frank Knight on risk, uncertainty, and the firm: a new interpretation. Econ Inq 31(3):456–465
Large R, Stolzle W (1999) Logistikforschung im Spiege ¨ l wissenschaftlicher Publikationen. Eine empirische Untersuchung auf Basis betriebswirtschaftlicher und ingenieurwissenschaftlicher Dissertationen. In: Pfohl HC (ed) Logistikforschung – Entwicklungsz¨uge und Gestaltungsansatze. Schmidt Verlag, Berlin, pp 3–35 ¨
Larson PD (2005) A note on mail surveys and response rates in logistics research. J Bus Logist 26(2):211–222
Larson PD, Gammelgaard B (2001-2002) The logistics triad: survey and case study results. Transport J 41(2-3):71–83
Larson PD, Poist RF (2004) Improving reponse rates to mail surveys: a research note. Transport J 43(4):67–74
Lassar WM, Kerr JL (1996) Strategy and control in supplier-distributor relationships: an agency perspective. Strat Manag J 17(8):613–632
Lassar W, Zinn W (1995) Informal channel relationships in logistics. J Bus Logist 16(1):81–106
Lasser R (1984) Die Bedeutung der transsibirischen Container Line (TSCL) als Wettbewerber im Far East Trade. Hussverlag: M¨unchen (also Doctoral Thesis at Universitat Mannheim) ¨
Lauenroth L (2001) Das Wissen als U berlebensgarantie Quartiersmann ist ein ¨ ¨ uber die Ware definierter Fachspediteur. Deutsche Verkehrszeitung No.51/2001 (28.04.2001), 21
Lauf JO (2008) IT-Sicherheitsarchitektur zur Container¨uberwachung, Doctoral Thesis at Technische Universitat Hamburg-Harburg ¨
Lawler EL, Wood DE (1966) Branch-and-bound methods: a survey. Oper Res 14(4):699–719
Lazzarini SC, Chaddad FR, Cook ML (2001) Integrating supply chain and network analyses: the study of netchains. J Chain Netw Sci 1(1):7–22
Lederer PJ, Nambimadom RS (1998) Airline network design. Oper Res 46(6):785–804
Lee R (2009) Social capital and business and management: Setting a research agenda. Int J Manag Rev 11(3):247–273

Lehmann H (1949) Die Entwicklung von Großhandel und Spedition in Duisburg im 19. Jahrhundert, Doctoral Thesis at Universitat K ¨ oln ¨

Lehrner H (1943) Der Donaug¨uterverkehr, unter besonderer Ber¨ucksichtigung der Stellung des internationalen Binnenschiffahrtspediteurs, Doctoral Thesis at Hochschule f¨ur Welthandel, Wien

Leland HE, Pyle DH (1977) Informational asymmetries, financial structure and financial

intermediation. J Finance 32(2):371–387 Lemoine W, Dagnæs L (2003) Globalisation strategies and business of organisation of a network

of logistics service providers. Int J Phys Distrib Logist Manag 33(3):209–228 Lennarz G (2001) Renaissance des Schieneng¨u terverkehrs? Bus & Bahn No.12/2001, 14–16

LeRoy SF, Singell Jr. LD (1987) Knight on risk and uncertainty. J Polit Econ 95(2):394–406

Levinson M (2006) The box: how the shipping container made the world smaller and the world economy bigger. Princeton University Press, NJ

Levinthal D (1988) A survey of agency models of organizations. J Econ Behav Organ 9(2):153–185 Lewin K (1936) Principles of topological psychology. McGraw Hill, New York

Lewin K (1951) Field theory in the social sciences. Harper, New York

Lewis I, Talalayevsky A (1997) Travel agents: threatened intermediaries? Transport J 36(3):26–30 Li Y (1998) Middlemen and private information. J Monetary Econ 42(1):131–159

Li PP (2007) Social tie, social capital, and social behavior: Toward an integrative model of informal exchange. Asia Pacific J Manage 24(2):227246

Li M, Choi TY (2009) Triads in services outsourcing: bridge, bridge decay and bridge transfer. J Supply Chain Manag 45(3):27–39

Liao J, Welsch H (2003) Social capital and entrepreneurial growth aspiration: a comparison of technology- and non-technology-based nascent entrepreneuers. J High Tech Manag Res 14(1):149–170

Lieb TC (1990) Wettbewerbliche Perspektiven von Weltcontainerhafen unter ver ¨ anderten see- ¨ und landseitigen Rahmenbedingungen: eine systemanalytische Untersuchung – dargestellt am Beispiel der nordwesteuropaischen H ¨ afen, Lang: Frankfurt-on-Main - Bern (also Doctoral ¨ Thesis at Universitat Mannheim 1989) ¨

Lieb T (2000) Seehafenhinterlandverkehr. In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, Munich

Lilliefors HW (1967) On the Kolmogorov-Smirnov test for normality with mean and variance unknown. J Am Stat Assoc 62(318):399–402

Lim C (1981) Risk pooling and intermediate trading agents. Can J Econ 14(2):261–275

Lin N (1976) Foundations of social research. McGraw-Hill, New York

Lin N (1999) Building a network theory of social capital. Connections 22(1):28–51

Lincoln JR (1982) Intra-(and inter-)organizational networks. Res Sociol Org 1:1–38

Linden W (1966) Dr. Gablers Verkehrs-Lexikon. Gabler, Wiesbaden

Lindner A, Benz MU (2002) Die Herren des Netzwerks, Deutsche Verkehrszeitung No.91/2002 (01.08.2002), 3

Linnemann H (1966) An econometric study of international trade flows. North-Holland, Amsterdam

Linton R (1936) The study of man. Appleton-Century-Crofts, New York

Linz M (2008) Airfreight supply chain collaboration – wettbewerbliche Potenziale und Wirkung auf die logistische Effizienz. Gabler, Wiesbaden (also Doctoral Thesis at European Business School Oestrich-Winkel)

Lizzeri A (1999) Information relevation and certification intermediaries. Rand J Econ 30(2): 214–231

Llewellyn KN (1931) What price contract? – An essay in perspective. Yale Law J 40:704–751

Logan MS (2000) Using agency theory to design successful outsourcing relationships. Int J Logist Manag 11(1):21–32

Logistics Europe (2002) The 3PL Mix, Logistics Europe July 2002, 26–32

Lohatepanont M, Barnhart C (2004) Airline schedule planning: integrated models and algorithms for schedule design and fleet assignment. Transport Sci 38(1):19–32

Lohre D (2005) Umweltmanagement und Qualifizierung in Speditionen, Kovac: Hamburg (also Doctoral Thesis at Universitat Duisburg-Essen) ¨

Lommelen T (2004) A learning perspective of internationalization: progression in the internationalization of logistics services providers. Doctoral Thesis at Limburgs Universitair Centrum: Limburg
Lopez E (2003) How ocean carriers organize the empty containers reposition activity in the USA? Marit Pol Manag 30(4):339–355
Lorenz W (2004) Leitfaden f¨ur Spediteure und Logistiker in Ausbildung und Beruf (Band 2), Deutscher Verkehrs-Verlag: Hamburg
Lorenz W (2005) Leitfaden f¨ur Spediteure und Logistiker in Ausbildung und Beruf (Band 1), Deutscher Verkehrs-Verlag: Hamburg
Lorrain F, White HC (1971) Structural equivalence of individuals in social networks. J Math Soc 1:49–80
Losch N (1972) Langfristige Unternehmensplanung unter Ber¨ ¨ ucksichtigung des Speditionsgewerbes, Doctoral Thesis at Universitat Innsbruck ¨
Lovaglia MJ, Skvoretz J, Willer D, Markovsky B (1995) Negotiated exchanges in social networks. Soc Forces 74(1):123–155
Love J (2004) 3PL/4PL Where Next? Logist Transport Focus 6(3):18–21
Lu Y, Dinwoodie J (2002) Comparative perspectives in international freight forwarder services in China. Transport J 42(2):17–27
Lublow R (1995) Konzeption eines telematikgest ¨utzten Speditionsleitstandes f¨ur den Sammelgutumschlag, Verlag Praxiswissen: Dortmund (also Doctoral Thesis at Universitat¨ Dortmund)
Lucke HJ, Eisenkopf A, L¨usch J, Milczarek N, Stabenau H, Woda A, Wolfel B (2001) Verkehrsun- ¨ ternehmen als Logistikdienstleister. In: Krampe H, Lucke H-J (eds), Grundlagen Logistik: Einf¨uhrung in Theorie und Praxis logistischer Systeme. Huss-Verlag, Munich, pp 251–300
Ludvigsen J (2000) The international networking between european logistical operators. EFI: Stockholm School of Economics (also PhD-Thesis at Stockholm School of Economics)
Lunnan R, Reve T (1995) Governance mechanisms in a network organization the case of a Norwegian industrial shipping company. In: Hakansson H, Snehota I (eds) Developing business ˚ relationships. Routledge, London, pp 357–368
Lutz NA (1995) Ownership rights and incentives in franchising. J Corp Finance 2:103–131
Lutze W (1982) Probleme der Einf¨uhrung des Containerverkehrs in die seewartigen Außenhandel- ¨ stransporte der Entwicklungslander, Doctoral Thesis at Universit ¨ at Rostock ¨
Maaß M (1927) Die Unternehmungen im Luftverkehr, ihre Entstehung, Entwicklung sowie ihre Formen, Verbande und Zusammenschl¨ ¨ usse bis zum Jahre 1926 unter besonderer Ber¨ucksichtigung der deutschen Verhaltnisse, Hofer: G ¨ ottingen (also Doctoral Thesis at Univer- ¨ sitat G ¨ ottingen) ¨
Macaulay S (1963) Non-contractual relations in business: a preliminary study. Am Socio Rev 28(1):55–67
MacDonald GM (1984) New directions in the economic theory of agency. Can J Econ 17(3): 415–440
Macharis C, Bontekoning YM (2004) Opportunities in intermodal freight transport research: a review. Eur J Oper Res 153(2):400–416
Macho-Stadler I, Perez-Castrillo JD ( ´ 1998) Centralized and decentralized contracts in a moral hazard environment. J Ind Econ 46(4):489–510
MacMillan K, Farmer D (1979) Redefining the boundaries of the firm. J Ind Econ 27(3):277–285
Macneil IR (1974) The many futures of contracts. South Calif Law Rev 47:691–816 Macneil IR (1978) Contracts: adjustment of long-term economic relations under classical, neoclassical, and relational contract law. Northwest Univ Law Rev 72(6):854–905
Macneil IR (1980) The new social contract an inquiry into modern contractual relations. Yale University Press, New Haven
Macneil IR (1983) Values in contract: internal and external. Northwest Univ Law Rev 78(2): 340–418

Madhavan A (2000) Market microstructure: a survey. J Financ Market 3(3):205–258

Maetz H (1937) Die wirtschaftlichen Grundlagen des Luftverkehrs, Doctoral Thesis at Hochschule f¨ur Welthandel Wien (published in parts as Maetz, H. (1937) Die wirtschaftlichen Grundlagen des Luftverkehrs – Teildruck, Kmoch: Vienna)

Magnanti TL (1996) Network optimization. In: Gass SI, Harris CM (eds) Encyclopedia of operations research and management science. Kluwer, Boston, pp 431–437

Magnanti TL, Wong RT (1984) Network design and transportation: models and algorithms. Transport Sci 18(1):1–55

Magretta J (1998) Fast, global, and entrepreneurial: supply chain management, Hong Kong style. Harv Bus Rev 76(5):102–115

Malchow U (1996) Schneller G¨uterverkehr ¨uber See: Analyse der technischen und okonomischen ¨ Realisierbarkeit von Schiffskonzepten und Einsatzfeldern am Beispiel der Containerschiffahrt, Doctoral Thesis at Technische Universitat Berlin ¨

Maloni MJ, Jackson EC (2005) North American container port capacity: a literature review. Transport J 44(2):16–36

Maloni MJ, Carter CR (2006) Opportunities for research in third-party logistics. Transport J 45(2):23–38

Maltz A (1993) Private fleet use: a transaction cost model. Transport J 32(3):46–53

Maltz A (1994) Outsourcing the warehousing function: economic and strategic considerations. Logist Transport Rev 30(3):245–265

Mann HB, Whitney DR (1947) On a test whether one of two random variables is stochastically larger than the other. Ann Math Stat 18(1):50–60

Mann DP, Wissink JP (1988) Money-back contracts with double moral hazard. Rand J Econ 19(2):285–292

Marasco A (2008) Third-party logistics: a literature review. Int J Prod Econ 113(1):127–147

Margolin BA (1993) An Economic Theory of the Brokerage Contract, PhD Thesis at George Mason University, Fairfax (UMI 9824210)

Markides V, Holweg M (2006) On the diversification of international freight forwarders. Int J Phys Distrib Logist Manag 36(5):336–359

Markovsky B, Skvoretz J, Willer D, Lovaglia M, Erger J (1993) The seeds of weak power: an extension of network exchange theory. Am Socio Rev 58(2):197–209

Markovsky B, Willer D, Patton T (1988) Power relations in exchange networks. Am Socio Rev 53(2):220–236

Marschak J (1974) Economic information, decision and prediction. D. Reidel, Dordrecht

Marsden PV (1981) Introducing influence processes into a system of collective decisions. Am J Sciol 86(6):1203–1235

Marsden PV (1982) Brokerage behavior in restricted exchange networks. In: Marsden PV, Lin N (eds) Social structure and market analysis. Sage, Beverly Hills, pp 201–218

Marsden PV (1983) Restricted access in networks and models of power. Am J Sociol 88(4): 686–717

Marsden PV (1990) Network data and measurement. Annu Rev Sociol 16:435–463

Marsden PV (2002) Egocentric and sociocentric measures of network centrality. Soc Network 24(4):407–422

Marsden PV, Campbell KE (1984) Measuring tie strength. Soc Forces 63(2):482–501

Martiensen J (2000) Institutionenokonomik: Die Analyse der Bedeutung von Regeln und Organi- ¨ sationen f¨ur die Effizienzokonomischer Tauschbeziehungen, Vahlen: Munich ¨

Martin JL (2003) What is field theory? Am J Sociol 109(1):1–49

Martin J, Thomas BJ (2001) The container terminal community. Marit Pol Manag 28(3):279–292 Marvel H-P, McCafferty S (1984) Resale price maintenance and quality certification. Rand J Econ 15(3):346–359

Masten SE (1995) Case studies in contracting and organization. Oxford University Press, New York Masten SE, Meehan JW, Snyder EA (1991) The costs of organisation. J Law Econ Organ 7(1):1–25

Mathissen G (1969) Die seewartige Verkehrslage einer Region: Verschiebungen im Zeichen des ¨ Containerverkehrs, Vandenhoeck & Ruprecht: Gottingen (also Doctoral Thesis at Universit ¨ at¨ Hamburg)

Mathaisel DFX (1996) Decision support for airline system operations control and irregular operations. Comput Oper Res 23(11):1083–1098
Mathaisel DFX (1997) Decision support for airline schedule planning. J Combin Optim 1(3): 251–275
Matthai J (1979) Die Spe ¨ dition im Strukturwandel, Doctoral Thesis at Hochschule f¨ur Verkehrs wesen "Friedrich List", Dresden
Matthai J (1981) Merkmale zur Charakterisierung des Strukturwandels in der Spedition, ¨ Wissenschaftliche Zeitschrift der Hochschule f¨ur Verkehrswesen "Friedrich List" in Dresden, 28(3):507–517
Matthai J (1985) Sped ¨ ition Grundlagen Heft 1, Schriftenreihe der Zentralstelle f¨ur Bildung des Ministeriums f¨ur Verkehrswesen: Berlin
Matthai J (1991) Verkehrswirtscha ¨ ftlicher Strukturwandel und Spedition vom Vermittler zum Operator und Integrator logistischer Ketten. Verkehrswissenschaft aktuell 1(1):9–15
Mattsson L-G (1985) An application of a network approach to marketing: defining and changing market positions. In: Dholakia A, Arndt J (eds) Changing the course of marketing: alternative paradigms for widening marketing theory. JAI Press, Greenwich
Mattsson L-G (1986) Indirect relationships in industrial networks: a conceptual analysis of their significance, 3rd IMP International Seminar at IRE, Lyon
Mattsson L-G (1995) Relationships and networks. In: Baker MJ (ed) Companion encyclopedia of marketing. Routledge, London, pp 202–212
Mattsson LG (2003) Reorganization of distribution in globalization of markets: the dynamic context of supply chain management. Supply Chain Management: An International Journal 8(5):416–426
Maurer P (2003) Luftverkehrsmanagement, 3rd edn. Oldenbourg, M¨unchen
Mayer L (1933) Speditionsgeschaft und Speditionsbetrieb, Heymanns: Berlin or Oestereichischer ¨ Wirtschaftsverlag: Vienna
Mayer G (2001) Strategische Logistikplanung von Hub&Spoke-Systemen, Gabler: Wiesbaden (also Doctoral Thesis at Technische Universitat Darmstadt 2000) ¨
Mayercordt W (1958) Die volkswirtschaftliche Bedeutung des Behalterverkehrs, Doctoral Thesis ¨ at Universitat Bonn 1959 ¨
McAfee RP, McMillan J (1987) Competition for agency contracts. Rand J Econ 18(2):296–307
McAfee RP, McMillan J (1995) Organizational diseconomies of scale. J Econ Manag Strat 4(3):399–426
McCalla RJ, Slack B, Comtois C (2004) Dealing with globalisation at the regional and local level: the case of contemporary containerization. Can Geogr 48(4):473–487
McCarty C, Wutich A (2005) Conceptual and empirical arguments for including or excluding ego from structural analyses of personal networks. Connections 26(2):82–88 McGillcuddy NB, Welton GL, Pruitt DG (1987) Third-party intervention: a field experiment
comparing three different models. J Pers Soc Psychol 53(1):104–112
McSweeney B (2002a) Hofstede's model of national cultural differences and their consequences: a triumph of faith – a failure of analysis. Hum Relat 55(1):89–118
McSweeney B (2002b) The essentials of scholarship: a reply to Geert Hofstede. Hum Relat 55(11):1363–1372
Mead G (1934) Mind, self and society. University of Chicago Press, Chicago
Meersmans PJM (2002) Optimization of container handling systems, Tinbergen Institute Research Series No.271, Tinbergen Institute: Rotterdam - Amsterdam (also PhD Thesis at Erasmus Universiteit Rotterdam)
Meersmans PJM, Dekker R (2001) Operations research supports container handling, Technical Report EI 2001-22, Econometric Institute at Erasmus University Rotterdam
Meier R (1977) Die Aufteilung des Frachttransportes auf Luft- und Oberflachenverkehr, Doctoral ¨ Thesis at Eidgenossische Technische Hochschule Z¨ ¨ urich

Meier LH (2008) Koordination interdependenter Planungssysteme in der Logistik: Einsatz multiagentenbasierter Simulation im Planungsprozess von Container-Terminals im Hafen, Gabler: Wiesbaden (also Doctoral Thesis at Univeristat G ¨ ottingen) ¨

Melumad N, Mookherjee D, Reichelstein S (1995) Hierarchical decentralization of incentive contracts. Rand J Econ 26(4):654–672
Melumad N, Mookherjee D, Reichelstein S (1997) Contract complexity, incentives, and the value of delegation. J Econ Manag Strat 6(2):257–289
Melumad N, Reichelstein S (1987) Centralization versus delegation and the value of communication. J Account Res 25(Suppl):1–18
Menard C (2000) Enforcement procedures and governance structures: what relationship? In: ´ Menard C (ed) Instituti ´ ons, contracts, and organizations. Elgar, Cheltenham, pp 234–253
Mengen A (1993) Konzeptgestaltung von Diensleistungsprodukten: eine Conjoint-Analyse im Luftfrachtmarkt unter Ber ¨ucksichtigung der Qualitatsunsicherheit beim Dienstleistungskauf, ¨ Schaffer-Poeschel: Stuttgart (also Doctoral Thesis at Wissenschaftliche Hochschule f¨ ¨ ur Unternehmensf¨uhrung Koblenz-Vallendar 1992)
Mentzer JT, Kahn KB (1995) A framework of logistics research. J Bus Logist 16(1):231–250
MergeGlobal (2008a) Forwarder momentum opportunities for value creation in freight forwarding. Am Shipper 3:36–53
MergeGlobal (2008b) Insomnia why challenges facing the world containershipping industry make for more nightmares than they should. Am Shipper 7:68–85
MergeGlobal (2008c) End of an era? Why the 'super spike' in fuel prices may signal the end of 'super growth' in air freight. Am Shipper 8:32–46
Merton RK (1957) The role-set: problems in sociology. Br J Scoiol 8(2):106–120
Mester B (1986) Marktchancen der Binnenschiffahrt im Container-Hinterlandverkehr der Hafen der Hamburg-Antwerpen-Range, Verkehrs-Verlag: D¨ ¨ usseldorf (also Doctoral Thesis at Universitat K ¨ oln) ¨
Metzger O (1922) Die internationale Spedition. In: Esch E (ed) Die Spedition Vortrage gehalten ¨ im Sonderlehrgang f¨ur Spedition 1922, Universiats-Buchandlung Oskar M¨ ¨ uller: Cologne, 89–99
Meyer E (1922) Die Bedeutung des Speditionsgewerbes im deutschen Wirtschaftsleben, Doctoral Thesis at Universitat Leipzig ¨
Meyer R (1957) Der Spediteur in seinem Verha ltnis zu Eisenbahn und Kraftverkehr, DVZ: ¨ Hamburg (also Doctoral Thesis Universitat zu K ¨ oln 1956) ¨
Meyer G (1975) Untersuchungen ¨uber die Zuverlassigkeit von Transportketten, dargestellt am ¨ Containertransport, Doctoral Thesis at Hochschule f¨ur Verkehrswesen, Dresden
Michaletz T (1994) Wirtschaftliche Transportketten mit modularen Containern, Huss: Munich Midoro R, Musso E, Parola F (2005) Maritime liner shipping and the stevedoring industry: market
structure and competition strategies. Marit Pol Manag 32(2):89–106 Miles RE, Snow CC (1984) Fit, failure and the hall of fame. Calif Manag Rev 26(3):10–28
Miles RE, Snow CC (1986) Organizations: new concepts of firms. Calif Manag Rev 27(3):62–73
Miles RE, Snow CC (1994) Fit, failure and the hall of fame. Simon & Schuster, New York
Miles RE, Snow CC, Meyer AD, HL Coleman Jr. (1978) Organizational strategy, structure and process. Acad Manag Rev 3(3):546–562
Miller MB (2002) Ship agents in the twentieth century. In: Boyce G, Gorski R (eds) Resources and infrastructures in the maritime economy 1500-2000. International Maritime Economic History Association, St. John's, pp 5–22
Miller MB (2003) The business trip: maritime networks in the twentieth century. Bus Hist Rev 77(1):1–32
Millon MH, Thakor AV (1985) Moral Hazard and Information Sharing: A Model of Financial Information Gathering Agencies. The Journal of Finance 40(5):1403-1422
Min S, Kim SK, Chen H (2008) Developing social identity and social capital for supply chain management. J Bus Logist 29(1):283–304
Minoux M (2003) Optimum network design models and algorithms in transportation and communication. Int J Logist Res Appl 6(1-2):5–15

Mintz B, Schwartz, M (1985) The power structure of American business. University of Chicago Press

Mira MSE (1993) Organization and management of container load centre: (the case of Port-Said), Doctoral Thesis at Universitat Rostock ¨

Mirrlees JA (1976) The optimal structure of incentives and authority within an organization. Bell J Econ 7(1):105–131

Mirrlees JA (1999) The theory of moral hazard and unobservable behaviour: part I. Rev Econ Stud 66(1):3–21

Mitchell WC (1913) Business cycles. University of California Press, Berkeley

Mitchell WC (1925) Quantitative analysis in economic theory. Am Econ Rev 15(1):1–12

Mitchell WC (1927) Business cycles: the problem and its setting. National Bureau of Economic Research, New York

Mitchell JC (1969) Soc network in urban situations: analysis of personal relationships in central African towns. Manchester University Press, Manchester

Mitchell JC (1974) Social networks. Annu Rev Anthropol 3:279–299

Mitchell JC (1979) Networks, algorithms, and analysis. In: Holland PW, Leinhardt S (eds) Perspectives on social network research. Academic Press, New York, pp 425–452

Mitroussi K (2003) Third party ship management: the case of separation of ownership and management in the shipping context. Marit Pol Manag 30(1):77–90

Mitroussi K (2004) The ship owners' stance on third party ship management: an empirical study. Marit Pol Manag 31(1):31–45

Mittendorf H und Oelfke W (1974) G¨uterverkehr und Spedition – Verkehrsbetriebslehre. Gehlen, Bad Homburg

Mizruchi WS, Mariolis P, Schwartz M, Mintz B (1986) Techniques for disaggregating centrality scores in social networks. Soc Meth 16:26–48

Mizruchi MS, Potts BB (1998) Centrality and power revisited: actor success in group decisison making. Soc Network 20(4):353–387

Mollekopf HG (1956) Luftfracht und Verkehrsstatistik, Doctoral Thesis at Universitat T¨ ¨ ubingen

Monch L (1972) Factoring und Leasing als Organisationsinstrumente im Containerverkehr, ¨ Doctoral Thesis at Universitat M¨ ¨ unster

Monzel J (1974) Die Anderung des Wettbewerbsverh ¨ altnisses im St¨ ¨ uckgutumschlag der Seehafen: ¨ eine Untersuchung der Antwerpen-Hamburg-Range unter besonderer Ber¨u cksichtigung des Containerverkehrs, Fundament-Verlag Sasse: Hamburg (also Doctoral Thesis at Universitat¨ Hamburg)

Mookherjee D, Tsumagari M (2004) The organization of supplier networks: effects of delegation and intermediation. Econometrica 72(4):1179–1219

Moral Garcia J (1984) Bedingungen f¨ur den rationellen Umschlag von Containern in kubanischen Seehafen, untersucht am Beispiel des Hafens Havanna, Doctoral Thesis at Universit ¨ at Rostock ¨

Morash EA (1986) Customer service, channel separation, and transportation intermediaries. J Bus Logist 7(1):89–107

Morash EA (1987) Using transportation intermediaries for industrial purchasing decisons. J Bus Ind Market 2(4):15–27

Moreno JL (1934) Who shall survive? A new approach to the problem of human interrelations. Nervous and Menthal Disease Publishing, Washington D.C.

Moreno JL (1953) Who shall survive? Foundations of sociometry, groups psychotherapy, and sociodrama. Beacon House, Beacon

Morse PM, Kimball GE (1951) Methods of operations research, Wiley, New York

Mosler GF (1993) Strukturveranderungen in der Luftfracht durch Ersatzverkehr – Auswirkungen ¨ des "Trucking" auf Airlines und Flughafen. Internationales Verkehrswesen 45(9):507–512 ¨

Mosler GF (1999) Emery worldwide – Spediteur, airline oder "Integrated Forwarder"? Transportmarkt 8-9:52–54

Mosler GF (2001) Schulterschluß der Aircargo-Agenten, Deutsche Verkehrszeitung No.121/2001 (09.10.2001), 20

Mosler GF (2002) GSA – das unbekannte Wesen, Deutsche Verkehrszeitung Nr.111/2002 (17.09.2002), 27–28

M¨uglich A (2002) Transport- und Logistikrecht. Vahlen, Munich
M¨uller N (1990) Chronologische Bibliographie der deutschsprachigen speditionswissenschaftlichen Literatur (1841–1989), Duisburg
M¨uller S (1991) Die Psyche des Managers als Determinante des Exporterfolges: Eine kulturvergleichende Studie zur Auslandsorientierung von Managern aus sechs Landern, M&P: Stuttgart ¨
M¨uller S (2001) Ursachen und Konsequenzen von Korruption, Wirtschaftswissenschaftliches Studium 31(9):492–496
M¨uller S (2002) Soziale, kulturelle und psychische Distanz: Der Raum als Gegenstand der Marketing-Forschung, Wissenschaftliche Zeitschrift der Technischen Universitat Dresden ¨ 51(4-5):144–150
M¨uller-Feldhammer R (1994) Die Standarddokumente des kombinierten Transports Hinwendung zum System der modifizierten Einheitshaftung. Transportrecht 7-8:272–277
M¨uller-Feldhammer R (1996) Die Haftung des Unternehmers beim multimodalen Transport f¨ur G¨uterschaden und G¨ ¨ uterverluste aus dem Beforderungsvertrag, Lang: Frankfurt-on-Main (also ¨ Doctoral Thesis at Universitat Leipzig 1995) ¨
M¨uller-Rostin W (1996) Die Luftfrachtersatzbeforderung. Transportrecht No.6/1956:217–222 ¨
M¨uller S, Koglmayr HG (1986) Die psychische Distanz zu Auslandsm ¨ arkten: Ein verkanntes ¨ Exporthemmnis, Zeitschrift f¨ur betriebswirtschaftliche Forschung 38(9):788–804
M¨uller S, Kornmeier M (2002) Strategisches Internationales Management. Vahlen, Munich
Murphy PR, Daley JM, Dalenberg DR (1992) Profiling international freight forwarders: a benchmark. Int J Phys Distrib Logist Manag 22(1):35–41
Murphy PR, Daley JM (1994) Logistics issues in international sourcing: an exploratory study. Int J Purch Mater Manag 30(3):22–27
Murphy PR, Daley JM (1995) International freight forwarders: current acitivities and operational issues. Int J Purch Mater Manag 31(3):22–27
Murphy PR, Daley JM (1996a) International freight forwarder perspectives on electronic data interchange and information management issues. J Bus Logist 17(1):63–84
Murphy PR, Daley JM (1996b) A preliminary analysis of the strategies of international freight forwarders. Transport J 35(4):5–11
Murphy PR, Daley JM, Hall PK (1998) EDI issues in logistics: a user and carrier perspective. J Bus Logist 19(2):89–102
Murphy PR, Daley JM (2000) An empirical study of internet issues among international freight forwarders. Transport J 39(4):5–13
Murphy PR, Daley JM (2001) Profiling international freight forwarders: an update. Int J Phys Distrib Logist Manag 31(3):152–168
Murphy PR, Poist RF (2000) Third-party logistics: some user versus provider perspectives. J Bus Logist 21(1):121–133
Murr A (1979) Export/Import traffic management and forwarding, 6th edn. Cornell Maritime Press, Centreville
Myerson RB (1982) Optimal coordination mechanisms in generalized principal-agent problems. J Math Econ 10(1):67–81
Myerson RB, Satterthwaite MA (1983) Efficient mechanisms for bilateral trading. J Econ Theor 29(2):265–281
Nadel SF (1957) The theory of social structure. Cohen & West, London
Nahapiet J, Ghoshal S (1998) Social capital, intellectual capital and the organizational advantage. Acad Manag Rev 23(2):242–256
Nakao K (1990) Distribution of measures of centrality: enumerated distributions of Freeman's graph centrality measures. Connections 13(3):10–22
Neiberger C (2007) Vernetzungsunternehmen der Weltwirtschaft. Zeitschrift f¨ur Wirtschaftsgeographie 51(2):124–126

Neiberger C (2008) The effects of deregulation, changed customer requirements and new technology on the organisation and spatial patterns of the air freight sector in Europe. J Transport Geogr 16(4):247–256

Neuhof B (2000) Feeder-Dienste. In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, Munich
New S, Payne P (1995) Research frameworks in logistics: three models, seven dinners and a survey. Int J Phys Distrib Logist Manag 25(10):60–77
Newcomb TM (1961) The aquaintance process. Holt, Rinhart & Winston, New York
Newman MEJ (2005) A measure of betweenness centrality based on random walks. Soc Network 27(1):39–54
Nguyen D-H (1973) Untersuchung ¨uber die zweckmassige Gestaltung eines St¨ ¨ uckgut-Seehafens unter Ber¨ucksichtigung einer schrittweisen Umwandlung in einen Container-Terminal, Doctoral Thesis at Hochschule f¨ur Verkehrswesen, Dresden
Nickenig (1996) Struktur, Aufbau- und Ablauforganisation des Spediteursammelgutverkehrs, in: Jahrbuch der G¨u terverkehrswirtschaft 1996, Servicegesellschaft f¨ur Spedition und Logistik: Frankfurt-on-Main, 92–97
Nickerson JA, Silverman BS (2003a) Why arent't all truck drivers owner-operators? Asset ownership and the employment relation in interstate for-hire trucking. J Econ Manag Strat 12(1):91–118
Nickerson JA, Silverman BS (2003b) Why firms want to organize efficiently and what keeps them from doing so: inappropriate governance, performance, and adaption in a deregulated industry. Admin Q J 48(3):433–466
Niebuer A (1996) Qualitatsmanagement f¨ ¨ ur Logistikunternehmen, Gabler: Wiesbaden (also Doctoral Thesis Universit at zu K ¨ oln) ¨
Niegel JU (1991) Logistische Konzepte von Speditionsunternehmen unter besonderer Ber¨ucksichtigung der Entwicklung des Speditionsgewerbes in den neuen Landern der BRD, ¨ Doctoral Thesis at Hochschule f¨ur Verkehrswesen "Friedrich List", Dresden
Nielsen J (2000) Anforderungen an die Ausstellung von Seekonnossementen durch NVOCC-Linien und Rechtsmißbrauch. Transportrecht No.5/2000, 196–203
Niemeijer R (1973) Some applications of the notion of density. In: Boissevain J and Mitchell JC (eds) Network analysis: studies in human interaction. Mouton, The Hague, pp 45–64
Nieminen J (1974) On centrality in a graph. Scand J Psychol 15(1):322–336
Nilakant V, Rao H (1994) Agency theory and uncertainty in organizations: an evaluation. Organ Stud 15(5):649–672
Ninnemann J (2005) Seehafenwettbewerb in Europa: eine empirische Analyse der Wettbewerbsdeterminanten am Beipsiel ausgewahlter Containerh ¨ afen der Nordrange und im Mittelmeer, ¨ Kovac: Hamburg (also Doctoral Thesis at Universitat Hamburg) ¨
Nissen V, Bothe M (2002) Fourth party logistics ein Uberblick. Logistik Management 4(1):16–26 ¨ Nocker G (2002) Die beleglose Spedition, Lit: M¨ ¨ unster - Hamburg - London (revised Version of Doctoral Thesis at Universitat Osnabr¨ ¨ uck 1999)
Noh JD, Rieger H (2004) Random walks on complex networks. Phys Rev Lett 92(11): 118701–118704
Noldeke G, Schmidt KM (1995) Option contracts and renegotiations: a solution to the hold-up ¨ problem. Rand J Econ 26(2):163–179
Noltemeier H (1976) Graphentheorie: mit Algorithmen und Anwendungen, de-Gruyter: Berlin New York
North DC (1981) Structure and change in economic history. Norton, New York
North DC (1991) Institutions. J Econ Perspect 5(1):97–112
Notteboom TE (2002) Consolidation and contestability in the European container handling industry. Marit Pol Manag 29(3):257–269
Notteboom TE (2007) Spatial dynamics in the container load centers of the Le Havre-Hamburg range. Zeitschrift f¨ur Wirtschaftsgeographie 51(2):108–123
Notteboom TE, Merckx F (2006) Freight integration in liner shipping: a strategy serving global production networks. Growth Change 37(4):550–569

Notteboom TE, Winkelmans W (2001) Structural changes in logistics: how will port authorities face the challenge? Marit Pol Manag 28(1):71–89

Obergfell G, Senghas E (2000) Messelogistik. In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, Munich
Ochsenbauer C (1989) Organisatorische Alternativen zur Hierarchie: Uberlegungen zur ¨ Uberwindung der Hierarchie in Theorie und Praxis ¨ der betriebswirtschaftlichen Organisation, GBI-Verlag: Munich (also Doctoral Thesis at Universitat M¨ ¨ unchen 1988)
Odrich P (1998) Modernes Logistikmanagement, Frankfurt/Main
Offergeld J (1984) Spezialisierung des Verkehrsangebotes und Strukturwandel in der verladenden Wirtschaft – dargestellt am Beispiel von Strukturwandlungen in der Flachglasindustrie und ihren Auswirkungen auf den Verkehrssektor, Vandenhoeck und Ruprecht: Go ttingen (also ¨ Doctoral Thesis at Universitat M¨ ¨ unster 1983)
O'Grady S, Lane HW (1996) The psychic distance paradox. J Int Bus Stud 27(2):309–333
O'Hara M (1995) Market microstructure theory. Blackwell, Cambridge
O'Hara M, Oldfield GS (1986) The microeconomics of market making. J Financ Quant Anal 21(4):361–376
Ohling H (1950) Export Import Spedition: Handbuch f¨ur die Praxis und Technik des Außenhandels, 3rd rev. edn. Hemmerich & Lesser, Hamburg
Ohmae K (1985) Triad power the coming shape of global competition. MacMillan, New York
Ohmae K (1989) Managing in a borderless world. Harv Bus Rev 67(3):152–161
Ojala LM (1991) Strategic management of port operations. Publications from the Center for Maritime Studies University of Turku A8, Turku
Ojala LM (1994) A transaction cost analysis of Finnish, Swedish and Norwegian shipping. Marit Pol Manag 21(4):273–294
Ojala LM (1995) Logistics management in Finnish foreign trade transport. Publications of the Turku School of Economics and Business Adminstration Series A-3:1995, Turku
Okeafor US (1982) The role of the international freight forwarder in the U.S. export market, Ph.D. Thesis at Ohio State University, Columbus
Oldenburg B (2002) Blick ¨uber die eigene Reling. Logistik inside No.18/2002, 38–40
Oliver AL, Ebers M (1998) Networking network studies: an analysis of conceptual configurations in the study of inter-organizational networks. Organ Stud 19(4):549–583
Olivier D, Parola F, Slack B, Wang JJ (2007) The time scale of internationalisation: the case of the container port industry. Marit Econ Logist 9(1):1–34
Olivo A, Zuddas P, di Francesco M, Manca A (2005) An operational model for empty container management. Marit Econ Logist 7(3):199–222
Omta SWF, Trienekens JH, Beers G (2001) Chain and network science: a research framework. J Chain Netw Sci 1(1):1–6
Opatz A (1994) Das Hypothesen-Konzept der "kontestablen Markte" in Anwendung auf die ¨ Container-Linienschiffahrt, Doctoral Thesis at Universitat Rostock ¨
Opgenhoff L (1997) Kundenbindung von Luftfrachtunternehmen eine theoretische und empirische Untersuchung, Shaker: Aachen (also: Doctoral Thesis Universitat Saarbr¨ ¨ ucken 1997)
Ordemann F (1996) Beurteilung alternativer Hafenanlaufstrategien im Containerseeverkehr am Beispiel der Hamburg-Antwerpen-Range, Institut f¨ur Seewirtschaft und Logistik: Bremen (also Doctoral Thesis at Universitat Bremen) ¨
Orden A (1956) The transhipment problem. Manag Sci 2(3):276–285
Otto A (1993) Das Management der Qualitat von Transportdienstleistungen, GVB: Nuremberg ¨ (also Doctoral Thesis at Universitat Erlangen-N¨ ¨ urnberg)
Ouchi WG (1980) Markets, bureaucracies, and clans. Admin Q 25(1):129–141
Ozsomer A, Mitri M, Cavusgil ST (1993) Selecting international freight forwarders: an expert systems application. Int J Phys Distrib Logist Manag 23(3):11–21
Pagano M (1989) Trading volume and asset liquidity. Q J Econ 104(2):255–274
Palay TM (1984) Comparative institutional economics: the governance of rail freight contracting. J Leg Stud 13(2):265–287

Paldam M (2000) Social capital: one or many? Definition and measurement. J Econ Surv, 14(5):629–653

Palmer D (1983) Broken ties:interlocking directorates and intercorporate coordination. Administrative Science Quarterly 28(1):40-55
Palmhofer R (1944) Der Einfluß der Verkehrsentwicklung auf den Speditionsbetrieb, Doctoral Thesis at Hochschule f¨ur Welthandel Wien
Panayides PM (1999) International ship management market analysis and strategic opportunities, Cargo Systems Report, IIR Publications: London
Panayides PM (2002) Economic organization of intermodal transport. Transport Rev 22(4): 401–414
Panayides PM, Cullinane KPM (2002) The vertical disintegration of ship management: choice criteria for third party selection and evaluation. Marit Pol Manag 29(1):45–64
Panayides PM, Gong SXH (2002) Consolidation, mergers and acquisitions in the shipping industry. In: Grammenos CT (ed) Handbook of maritime economics and business. LLP, London, pp 598–620
Pankratz G (2002) Speditionelle Transportdisposition: Modell- und Verfahrensentwicklung unter Ber¨u cksichtigung von Dynamik und Fremdvergabe, DUV: Wiesbaden (also Doctoral Thesis at Fernuniversitat Hagen) ¨
Pappi FU (1987) Methoden der Netzwerkanalyse. Oldenbourg, Munich
Passas N, Jones K (2007) The regulation of non-vessel-operating common carriers (NVOCC) and customs brokers. J Financ Crime 14(1):84–93
Patton T, Willer D (1990) Connection and power in centralized exchange networks. J Math Sociol 16(1):31–49
Pedersen A-C (1996) Utvikling av Leverandørrelasjoner i Industrielle Nettverk, Doctoral thesis no. 107, Norwegian University of Science and Technology: Trondheim
Peng MW, Ilinitch AY (1998) Export intermediary firms: a note on export development research. J Int Bus Stud 29(3):609–620
Peng MW, York AS (2001) Behind intermediary performance in export trade: transactions, agents and resources. J Int Bus Stud 32(2):327–346
Perrow C (1986) Complex organizations: a critical essay. Random House, New York
Perry AC (1992) US international trade intermediaries: a field study investigation. Int Market Rev 9(2):7–20
Pesenti R (1995) Hierarchical resource planning for shipping companies. Eur J Oper Res 86(1): 91–102
Peters HJF (2001) Developments in global sea trade and container shipping markets: their effects on the port industry and private sector involvement. Int J Marit Econ 3(1):3–26
Peters AK (2002) Lead Logistics Provider treten das Erbe des 4PL an. Logistik inside No.19/2002, 40–41
Peters MJ, Lieb RC, Randall HL (1998) The use of third-party logistics services by European industry. Transport Logistics 1(3):167–179
Petersen T (1993) Recent developments in the economics of organization: the principal-agent relationship. Acta Sociologica 36(3):277–293
Pflugbeil P (2000) Bodendienste (Ground handling services). In: Bloech J, Ihde GB (eds) Logistik Total [CD-Rom]. Vahlen, Munich
Pfohl H-C (2003) Logistiksysteme Betriebswirtschaftliche Grundlagen, 7th edn. Springer, Berlin Pfohl H-C, Buse HP (2000) Interorganizational logistics systems in flexible production networks: an organizational capabilities perspective. Int J Phys Distrib Logist Manag 30(5):388–408
Pfohl H-C, Large R (1992) Gestaltung interorganisatorischer Logistiksysteme auf der Grundlage der Transaktionskostentheorie. Zeitschrift f¨ur Verkehrswissenschaft 63(1):15–51
Philipp J (1980) Die Einf¨uhrung moderner Transporttechnologien in der Linienseeschiffahrt mit Entwicklungslandern, Doctoral Thesis at Technische Universit ¨ at Berlin ¨
Phillips DM, Phillips JK (1998) A social network analysis of business logistics and transportation. Int J Phys Distrib Logist Manag 28(5):328–349

Picard P (1987) On the desgin of incentive schemes und moral hazard and adverse selection. J Publ Econ 33(3):305–331

Picot A (1982) Transaktionskostenansatz in der Organisationstheorie: Stand der Diskussion und Aussagewert. Die Betriebswirtschaft 42(2):267–284

Picot A, Dietl H, Franck E (2002) Organisation: eine okonomische Perspektive, 3rd edn. ¨ Scha ffer-Poeschel: Stuttgart ¨

Piore MJ, Sabel CF (1985 [1984]) Das Ende der Massen Produktion, Wagenbach: Berlin [translated and revised ed. of M.J. Piore and C.F. Sabel (1984) The second industrial divide – possibilities for prosperity. Basic Books, New York]

Pirath C (1934) Die Grundlagen der Verkehrswirtschaft. Springer, Berlin

Pirrong SG (1993) Contracting practices in bulk shipping markets: a transactions cost explanation. J Law Econ 36(2):937–976

Plehwe D (1999) Why and how do national monopolies go "Global"? Discussion Paper FS I 99 – 102, Wissenschaftszentrum Berlin f¨ur Sozialforschung, Berlin

Pocuca M, Cisic D, Komadina P (2000) Modeling the documentation flow in transport supply chain – results form the simulation experiments, Economics Working Paper Archive (WoPEc), see: http://netec.mmc.ac.uk/WoPEc/data/Papers/wpawuwpit0004004.html

Podolny JM (1994) Market uncertainty and the social character of economic exchange. Admin Sci Q 39(3):458–483

Podolny JM (2001) Networks as the pipes and prisms of the market. Am J Sociol 107(1):33–60

Ponta R (1997) Optimierung der Containerlogistik in Linienreedereien, Doctoral Thesis at Universitat Paderborn ¨

Pontow L (1922) Die betriebswirtschaftliche Organisation in Speditionsbetrieben. Springer, Berlin (also Doctoral Thesis at Universitat zu K ¨ oln 1992) ¨

Pope DJ, Thomchick EJ (1985) U.S. foreign freight forwarders and NVOCCs. Transport J 24(3):26–36

Popp A (2000) Swamped in information but starved of data: information and intermediaries in clothing supply chains. Supply Chain Manag Int J 5(3):151–161

Porter ME (1980) Competitve strategy. Free Press, New York

Portes A (1998) Social capital: its origins and applications in modern sociology. Annu Rev Sociol 24:1–24

Posner RA (1998) Economic analysis of law, 5th edn. Aspen Law & Business, New York

Potzl J (2000) Issues in direct channe ¨ l distribution: a comparison of selling via the internet in the airline business and the fast-moving consumer goods industry. Electron Market 10(3):153–157

Poulin R, Boily M-C, Masse BR (2000) Dynamical systems to defi ˆ ne centrality in social networks. Soc Network 22(3):187–220

Powell WW (1990) Neither market nor hierarchy: network forms of organization. Res Organ Behav 12:295–336

Powell WW, Smith-Doerr L (1994) Networks and economic life. In: Smelser NJ, Swedberg R (eds) The handbook of economic sociology. Princeton University Press, Chichester, pp 368–402

Prinz A (2008) Quantitative Analysen zum deutschen und internationalen Luftfrachtmarkt, Lang: Frankfurt-on-Main (also Doctoral Thesis at Universitat Mainz) ¨

Proctor CH, Loomis CP (1951) Analysis of sciometric data. In: Jahoda M, Deutsch M, Cook SW (eds) Research methods in social relations part two: selected techniques. Dryden Press, New York, pp 561–585

Puffer CE (1949) Air transportation. International Textbook Company, Scranton

Pumpe D (2000) Ein Referenzmodell zur Planung und Steuerung der Ablaufe in Seehafen- ¨ Containerterminals, Mensch-und-Buch-Verlag: Berlin (also Doctoral Thesis at Technische Universit at Berlin) ¨

Puttfarken (1997) Seehandelsrecht, Verlag Recht und Wirtschaft: Heidelberg

Quintero R (1985) Zur okonomischen Effektivit ¨ at eines kubanischen Betriebes der internationalen ¨ Spedition: ein Beitrag zur Weiterentwicklung der speditionelle Tatigkeiten in der Republik ¨ Kuba, Doctoral Thesis at Universitat Rostock ¨

Rabe D (2000) Die Probleme bei einer multimodalen Bef orderung unter Einfluß einer Seestrecke – ¨ Sind Losungen in Sicht? Transportrecht No.5/2000, 189–196 ¨

Rackwitz HH (1969) Der Container-Verkehr zur See – Eine Analyse organisatorischer und wirtschaftlicher Probleme unter besonderer Ber¨ucksichtigung der Bundesrepublik Deutschland, Magica-Verlag: Meilen - Z¨urich (also Doctoral Thesis at Universitat Freiburg, Fribourg) ¨

Radcliffe-Brown AR (1940) On social structure. J Roy Anthropol Soc Great Brit Ireland 70:1–12 Radcliffe-Brown AR (1957) A natural science of society. University of Chicago, Chicago

Rahman N (1989) An examination of the exporter and freight forwarder relationships: a strategic alliance perspective. UMI, Ann Arbor (also PhD Thesis at Ohio State University, Columbus)

Ramakrishnan RT, Thakor AV (1984) Information reliability and a theory of financial intermediation. Rev Econ Stud 51(3):415–432

Ramberg J (1998) Unification of the law of international freight forwarding. Uniform Law Review 3(1):5–13

Ramberg J (2000) The FIATA model rules on freight forwarding services – an opinion of Jan Ramberg. http://www.forwarderlaw.com/feature/ramberg3.htm

Ramcke RH (1973) Der Aufbau von Container-Transportsystemen der G¨uterverkehrsbetriebe, Doctoral Thesis at Universitat Hamburg ¨

Rangan S (2000a) The problem of search and deliberation in economic action: when social networks really matter. Acad Manag Rev 25(4):813–828

Rangan S (2000b) Search and deliberation in international exchange: microfoundations to some macro patterns. J Int Bus Stud 31(2):205–222

Rao K, Young RR, Novick JA (1993) Third party services in the logistics of global firms. Logist

Transport Rev 29(4):363–370 Rapoport A (1963) Mathematical models of social interaction. In: Luce RD, Bush RR, Galanter E (eds) Handbook of mathematical psychology vol I. Wiley, New York, pp 493–579

Ratcliff MSW, Bischoff EE (1998) Allowing for weight considerations in container loading. OR Spektrum 20(1):65–71

Rath E (1973) Container systems. Wiley, New York

Rauch JE, Watson J (2002) Entrepreneurship in international trade. NBER Working Paper No.8708. National Bureau of Economic Research, Cambridge

Razzaque MA, Sheng CC (1998) Outsourcing of logistics functions: a literature survey. Int J Phys Distrib Logist Manag 28(2):89–107

Rees R (1985a) The theory of principal and agent: part 1. Bull Econ Res 37(1):3–26

Rees R (1985b) The theory of principal and agent: part 2. Bull Econ Res 37(2):75–95

Reim¨uller H (1919) Das moderne Speditionsgeschaft, Doctoral Thesis at Universit ¨ at Frankfurt am ¨ Reinheimer S (1998) Marktorientierte elektronische Koordination zwischenbetrieblicher Geschaftsprozesse in der Luftfracht, dissertation.de: Berlin (also Doctoral Thesis at Universit ¨ at¨ Erlangen-N¨urnberg)

Reinheimer S, Bodendorf F (1999) A framework for electronic coordination in the air cargo market. Inform Soc 15(1):51–55

Reisman A, Xu X (1996) Libraries. In: Gass SI, Harris CM (eds) Encyclopedia of operations research and management science. Kluwer, Boston, pp 340–343

Reise S (2005) Offshore-Containerterminals als Transhipment-Hub – dargestellt am Beispiel der Deutschen Bucht, Doctoral Thesis at Technische Universitat Dresden ¨

Reitzes JD, Sheran KL (2002) Rolling Seas in Liner Shipping. Review of Industrial Organization 20(1):51–59

Rendez H (1992) Konzeption integrierter Logistik-Dienstleistungen, Huss Verlag: Munich (also Doctoral Thesis Technische Universitat Berlin) ¨

Rennings K (1992) Zur Relevanz der Transaktionskostentheorie f¨ur die Verkehrswirtschaft. In: Rennings K, Fonger M, Meyer H (eds) Transaktionskostentheorie als Entscheidungshilfe f¨ur die Verkehrswirtschaft, Vandenhoeck und Rupprecht: Gottingen, 7–49 ¨

Reve T (1990) The firm as a nexus of internal and external contracts. In: Aoki M, Gustafson B, Williams OE (eds) The firm as a nexus of treaties. Sage, London, pp 133–161

ReVelle C (1996) Location analysis. In: Gass SI, Harris CM (eds) Encyclopedia of operations research and management science. Kluwer, Boston, pp 349–354

Richards W, Seary A (2000) Eigen analysis of networks. J Soc Struct 1, no pages, see: http://www. cmu.edu/joss/content/articles/volume1/RichardsSeary.html

Richardson GB (1960) Information and investment. Oxford University Press, Oxford

Richardson GB (1972) The organisation of industry. Econ J 82(327):883–896

Richter H (1985) Organisationsformen multimodaler seewartiger Containertransportsysteme, Doc- ¨ toral Thesis at Universitat Rostock ¨

Richter H (1989) Thirty years of airline operations research. Interfaces 19(4):3–9

Richter R (1991) Institutionenokonomische Aspekte der Theorie der Unternehmung. In: ¨ Ordelheide D, Rudolph B, B¨u sselmann E (eds) Betriebswirtschaftslehre und okonomische ¨ Theorie, Poeschel Verlag: Stuttgart, 395–429

Richter S (1997) Die Bewertung produktionssynchroner Beschaffungen unter besonderer Ber¨ucksichtigung des Gebietspediteurs-Konzeptes, Cuvillier: Gottingen (also Doctoral Thesis ¨ at Universitat G ¨ ottingen) ¨

Richter, R (2005) The New Institutional Economics: Its Start, its Meaning, its Prospects. European Business Organization Law Review 6(2):161–200

Richter R, Furubotn E (2003) Neue Institutionenokonomik: Eine Einf¨ ¨ uhrung und kritische W¨urdigung, 3rd edn. Mohr Siebeck: T¨ubingen

Ricker Kelly ER, Lecraw D (1985) Trading houses to spur Canadian exports. Bus Q 50(1):36–42 Ricks DA (1993) Blunders in international business. Blackwell, Cambridge

Rieck J (2008) Tourenplanung mittelsta ndischer Speditionsunternehmen: Modelle und Methoden, ¨ Gabler: Wiesbaden (also Doctoral Thesis at Technische Universitat Clausthal) ¨

Rindfleisch A, Heide JB (1997) Transaction cost analysis: past, present, and future applications. J Market 61(4):30–54

Ringhausen H-H (1971) Einsatz der elektronischen Datenverarbeitung in der Kraftwagenspedtion unter besonderer Ber¨ucksichtigung des Sammelgutverkehrs, Doctoral Thesis at Universitat¨ Innsbruck

Robinson R (2002) Ports as elements in value-driven chain systems: the new paradigm. Marit Pol Manag 29(3):241–255

Roethlisberger FJ, Dickson WJ (1939) Management and the worker. Cambridge University Press, Cambridge

Rogers DI (1974) Sociometric analysis of interorganizational relations: application of theory and measurement. Rural Sociol 39(4):487–503

Rogers EM, R Agarwala-Rogers (1976) Communication in organizations. Free Press, New York Rogerson WP (1992) Contractual solutions to the hold-up problem. Rev Econ Stud 59(4):777–794

Romano RE (1994) Double moral hazard and resale price maintenance. Rand J Econ 25(3): 455–466

Ronen D (1983) Cargo ships routing and scheduling: survey of models and problems. Eur J Oper Res 12(2):119–126

Ronen D (1984) Review of cargo ship routing and scheduling. J Bus Logist 5(1):112–125

Ronen D (1993) Ship scheduling: the last decade. Eur J Oper Res 71(3):325–333

Rose AK (2000) One money, one market: the effect of common currencies on trade. Econ Pol 15(3):9–45

Rosen S (1985) Implicit contracts: a survey. J Econ Lit 23(3):1144–1175

Rosenbloom B (1991) Marketing channels: a management view, 4th edn. Dryden, Chicago

Rosenbloom B (1995) Channels of distribution. In: Baker MJ (ed) Companion encyclopedia of marketing. Routledge, London, pp 137–154

Rosenbloom B (1999) Channel management. In: Baker MJ (ed) The IEBM encyclopedia of marketing. ITP, London, pp 100–114

Rosenthal MS (1950) Techniques of international trade. McGraw-Hill, New York

Rossger E (1960) Luftverkehr und Sp ¨ edition, Forschungsberichte des Landes Nordrhein-Westfalen Nr.882, Westdeutscher Verlag: Cologne - Opladen

Ross SA (1973) The economic theory of agency: the principals' problem. Am Econ Rev 63(2): 134–139

Rubinstein A (1998) Modelling bounded rationality. MIT-Press, Cambridge

Rubinstein A, Wolinsky A (1987) Middlemen. Q J Econ 102(3):581–593

Rudolph W (2002) Supply chain management: Der 4PL im Praxistest. Distribution 10:10–11

Ruhnau B (2000) Eigenvector-centrality – a node-centrality? Soc Network 22(4):357–365

Rumpf C (1997) Qualitatsmanagement speditioneller Dienstleistungen: eine informationsorien- ¨ tierte Analyse der Planung und Vermarktung der Qualitat des Dienstleistungsangebots von ¨ Speditionen auf der Grundlage eines Geschaftstypenansatzes, Deutscher Verkehrs-Verlag: ¨ Hamburg (also Doctoral Thesis at Universitat Gießen) ¨

Rushmeier RA, Hoffman KL, Padberg M (1995) Recent Advances in Exact Optimization of Airline Scheduling Problems, Working Paper, Department of Operations Research & Operations Engineering, George Mason University: Fairfax

Rushmeier RA, Kontogiorgis SA (1997) Advances in the optimization of airline fleet assignment. Transport Sci 31(2):159–169

Rust J, Hall G (2003) Middlemen versus marketmakers: a theory of competitive exchange. J Polit Econ 111(2):353–403

Rutherford M (1983) J.R. Common's institutional economics. J Econ Issues 17(3):721–744

Rutherford M (2000) Institutionalism between the wars. J Econ Issues 34(2):291–303

Rutherford M (2001) Institutional economics: then and now. J Econ Perspect 15(3):173–194

Ryan DM (1996) Airline industry. In: Gass SI, Harris CM (eds) Encyclopedia of operations research and management science. Kluwer, Boston, pp 10–12

Saam NJ (2002) Prinzipale, Agenten und Macht: eine machttheoretische Erweiterung der Agenturtheorie und ihre Anwendung auf Interaktionsstrukturen in der Organisationsberatung, Mohr Siebeck: T¨ubingen

Sabidussi G (1966) The centrality index of a graph. Psychometrika 31(6):581–603

Salanie B (1990) S ´ election adverse et aversion pour le risque, Annales d' ´ Economie et de Statistique ´ 18:131–149

Salzen Hv. (2000) Seehafeninformationssysteme. In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, Munich

Samuels WJ (1991) Institutional economics. In: Eatwell J, Milgate M, Newman P (eds) The new palgrave: a dictionary of economics, vol 2 (Reprint). Macmillan, New York, pp 864–866

Sappington DEM (1991) Incentives in principal-agent relationships. J Econ Perspect 5(2):45–66 Sarathy R (1985) Japanese trading companies: can they be copied? J Int Bus Stud 28(1):101–119

Sawyer S, Crowston K, Wigand RT, Allbritton M (2003) The social embeddedness of transactions: evidence from the residential real-estate industry. Inform Soc 19(2):135–154

Schachinger M (1989) Kooperationen der osterreichischen Speditionen mit ausl ¨ andischen Partnern ¨ - Auswirkungen der bevorstehenden Liberalisierung der Verkehrsmarkte. In: DVWG (ed) Neue ¨ Aufgaben der Spedition im Rahmen verkehrslogistischer Entwicklungen, Schriftenreihe der Deutschen Verkehrswissenschaftlichen Gesellschaft e.V. (DVWG) No. B116, 31–41

Schaffer HF (2003) Von K ¨ itty Hawk nach Montreal – Der Weg zur International Civil Aviation Organisation (ICAO), Transportrecht 10:377–383

Scharl KJ (1963) Betriebswirtschaftliche Probleme des Speditionsgewerbes – insbesondere Kostenrechnung, Preisbildung und Besteuerung, Doctoral Thesis at Universitat Erlangen-N ¨ urnberg ¨

Scharlibbe U (1925) Das Speditionsgeschaft in den Seeh ¨ afen unter besonderer Ber¨ ¨ ucksichtigung Hamburgs, Doctoral Thesis at Universitat Frankfurt am Main ¨

Schauer O (2001) Das deutsche Transportrechtsreformgesetz – Entwicklung, Grundz¨uge, Kommentar zu ausgewahlten Auswirkungen auf die deutsche bzw. oesterreichische Trans- ¨ portwirtschaft, Doctoral Thesis at Universitat Linz ¨

Scheibe H-J (1980) Der Einfluß transnationaler Reedereien auf die Struktur der internationalen Linienschiffahrt, Doctoral Thesis at Universitat Bremen ¨

Schelling TC (1956) An essay on bargaining. Am Econ Rev 56(3):281–306

Schenk H (1930) Finanzierung und Organisation des Luftverkehrs, Unter besonderer Ber¨ucksichtigung der deutschen Verhaltnisse, Dissertation at Universit ¨ at Frankfurt-on-Main ¨

Schenk M (1984) Soziale Netzwerke und Kommunikation, Mohr: T¨ubingen (also Habilitation Thesis at Universitat Augsburg) ¨
Schindler A (1998) Vertragsbedingungen f¨ur den G ¨uterkraftverkehrs- und Logistikunternehmer (VBGL). In: Korf W (ed) Lorenz – Der G¨uterverkehr nach neuem Recht, DVZ: Hamburg, 234–287
Schindler A (2003) Neue Vertragsbedingungen f¨ur den G¨uterverkehrs-, Speditions- und Logistikunternehmer (VBGL), Transportrecht No.5/2003, 194–196
Schinzinger F (1991) German historical school. In: Eatwell J, Milgate M, Newman P (eds) The new palgrave: a dictionary of economics, vol 2 (Reprint). Macmillan, New York, pp 516–518
Schlichting W (1931) Der Seehafenspediteur in der deutschen Verkehrswirtschaft, Spaeth & Linde: Berlin - Vienna
Schmidt RH (1987) Agency Costs are not a "Flop"!. In: Bamberg G, Spremann K (eds) Agency theory, information, and incentives. Springer, Berlin, pp 495–509
Schmidt C (1996) Neue Trends in der Logistik und ihre Auswirkungen auf die Spedition, in: Jahrbuch der G¨u terverkehrswirtschaft 1996, Servicegesellschaft f¨ur Spedition und Logistik: Frankfurt-on-Main, 70–74
Schmidt L (2007) EU-Kartellrecht nunmehr pur im internationalen Seeverkehr, Transportrecht No.1/2007, 18–22
Schmidt RH, Hackethal A, Tyrell M (1998) Disintermediation als Bedrohung f¨ur Banken? Kreditwesen No.24/1998, 1371–1376
Schmidt G, Wilhelm WE (2000) Strategic, tactical and operational decisions in multi-national logistics networks: a review and discussion of modelling issues. Int J Prod Res 38(7): 1501–1523
Schmidthofer A (2004) Taktische Bestimmung von Tourgebieten f¨ ¨ ur die speditionelle Tourenplanung, Shaker: Aachen (also Doctoral Thesis at Universitat Augsburg) ¨
Schmitt WM (1981) Die Problematik von Leistung, Kapazitat, Kosten und Erl ¨ osen in der ¨ Containerschiffahrt, Lang: Frankfurt-on-Main - Bern (also Doctoral Thesis at Technische Universit at Berlin) ¨
Schmitz H-P (1957) Die Zusammenarbeit zwischen Eisenbahn und Spedition auf vertraglicher Basis in der Bundesrepublik Deutschland, Fischer: D¨usseldorf (also Doctoral Thesis at Universita t zu K ¨ oln) ¨
Schmitz PW (2001) The hold-up problem and incomplete contracts: a survey of recent topics in contract theory. Bull Econ Res 53(1):1–17
Schmitz M (2005) Modellierung und Simulation von Gebietsspediteurnetzen in der Automobilindustrie, Verlag Praxiswissen: Dortmund (also Doctoral Thesis at Universitat Dortmund) ¨
Schneider L (1925) Der Sammelladeverkehr der Spediteure auf den deutschen Eisenbahnen, Doctoral Thesis at Universitat Frankfurt am Main ¨
Schneider D (1987) Agency costs and transaction costs: flops in the principal-agent-theory of financial markets. In: Bamberg G, Spremann K (eds) Agency theory, information, and incentives. Springer, Berlin, pp 481–494
Schneider D (1993) Wettbewerbsvorteile integrierter Systemanbieter im Luftfrachtmarkt, Lang: Frankfurt-on-Main (also Doctoral Thesis at Wissenschaftliche Hochschule f¨ur Unternehmensf uhrung Koblenz) ¨
Schonknecht A (2007) Entwicklung eines Modells zur Kosten- und Leistungsbewertung von ¨ Containerschiffen in intermodalen Transportketten, Doctoral Thesis at Technische Universitat¨ Hamburg-Harburg
Schoppe SG (1995) Moderne Theorie der Unternehmung, Oldenbourg: Munich, Vienna
Schott K (1971) Deckungsbeitragsrechnung in der Spedition, DVZ: Hamburg (also Doctoral Thesis at Universitat Frankfurt am Main) ¨
Schott R (1989) Stauplanung f¨ur Containerschiffe, Vandenhoeck & Ruprecht: Gottingen (also ¨ Doctoral Thesis at Universitat Hamburg) ¨

Schotter A (1986) The evolution of rules. In: Langlois RN (ed) Economics as a process: essays in the new institutional economics. Cambridge University Press, Cambridge, pp 117–133

Schramm H-J, Niedermaier A (2008a) Make, buy or co-operate? Expansion of european transport and logistics companies revisited, NOFOMA 2008 – 20th Annual Conference for Nordic Researchers in Logistics, 05–06.06.2008, Helsinki
Schramm H-J, Niedermaier A (2008b) Patterns of joint venture, merger and acquisition activitcy in the transport and logistics industry, Logistics Research Network Annual Conference 2008, 10–12.09.2008, Liverpool
Schrijver A (2002) On the history of the transportation and maximum flow problems. Math Program 91(3):437–445
Schuh G (1971) Die Aufgabenteilung im binnenl andischen Seehafenverkehr mit Containern, ¨ Handelsblatt: D¨usseldorf (also Doctoral Thesis at Universitat zu K ¨ oln as Schuh, G. (1970) ¨ Leistungsbezogene Funktionsteilung im binnenlandischen Seehafenverkehr mit Containern) ¨
Sch¨uller M (2003) Strategieentwicklung airlinegef¨uhrter Supply Chains: spezifische Erfolgsfaktoren des Supply-Chain-Managements in der Luftfracht und Handlungsempfehlungen f¨ur Luftfracht-Carrier, Kovac: Hamburg (also: Doctoral Thesis at Universitat Bremen 2002) ¨
Schulz-Hanßen E (1965) Die Verkehrswirtschaft des Seehafens: eine Begriffs- und Wesensbestimmung, Duncker & Humblot: Berlin (also Doctoral Thesis at Universitat Hamburg) ¨
Schumacher W (1987a) Die Entwicklung der betriebswirtschaftlichen Logistik und ihr Einfluß auf das zuk¨unftige Leistungsbild eines deutschen Speditions- und Lagereigewerbes, M¨uller Botermann: Cologne (also Doctoral Thesis Universitat zu K ¨ oln 1986) ¨
Schumacher W (1987b) Reaktion der Speditionen auf ver anderte Marktstrukturen im Expreß- ¨ und Paketbereich, in: Transportmarkte: Kurier,- Expreß- und Paketdienste, Spiegel-Verlagsreihe ¨ Markte im Wandel, Band 13, 82–91 ¨
Schumpeter JA (1908a) Das Wesen und der Hauptinhalt der theoretischen Nationalokonomie, ¨ Duncker & Humblot: Leipzig
Schumpeter JA (1908b) On the concept of social value. Q J Econ 23(2):213–232
Schumpeter JA (1931) Theorie der wirtschaftlichen Entwicklung eine Untersuchung ¨uber Unternehmergewinn, Kapital, Kredit, Zins und den Konjunkturzyklus, 3rd edn. Duncker & Humblot: Leipzig [English translation as: Schumpeter JA (1934) The theory of economic development. Harvard University Press, Cambridge]
Schwartz SH (1992) Universials in the content and structure of values: theoretical advances and empirical tests in 20 countries. In: Zanna MP (ed) Advances in experimental social psychology. Academic Press, San Diego, pp 1–65
Schwartz SH (1994) Beyond individualism/collectivism new dimensions of values. In: Kim U, Triandis HC, Kagitcibasi C, Choi S-C, Yoon G (eds) Individualism and collectivism: theory, methods and applications. Sage, London, pp 85–119
Schwartz A (1998) Incomplete contracts. In: Newman P (ed) The new palgrave dictionary of economics and the law, vol 2. Macmillan, London, pp 277–283
Schwartz SH (1999) A theory of cultural values and some implications for work. Appl Psychol Int Rev 48(1):23–47
Schwartz SH, Sagiv L (1995) Identifying culture specifics in the content and structure of values. J Cross Cult Psychol 26(1):92–116
Scott J (1991) Social network analysis: a handbook. Sage, London
Securius T, Boning E (1941) G¨ ¨ uterverkehr und Spedition im Großdeutschen Reich, Teil III: Spedition, Gehlen: Leipzig - Berlin
Seeley JR (1949) The net of reciprocal influence a problem in treating sociometric data. Can J Psychol 3(4):234–240
Segendorff B (1998) Delegation and threat in bargaining. Game Econ Behav 23(2):266–283
Seidelmann C (1969) Kombinierter Verkehr mit Containern, Schriftenreihe des Verbandes der Automobilindustrie No.1, VDA: Frankfurt-on-Main
Seidelmann C (1971) Kombinierter Verkehr im Luftfrachtdienst, Doctoral Thesis at Universitat¨ Marburg

Seidelmann C (2000a) Container. In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen: Munich

Seidelmann C (2000b) Kombinierter Verkehr. In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, Munich
Seidenfus HS (1973) Heutige Stellung und K¨unftige Bedeutung der Spedition und gewerblichen Lagerei innerhalb unserer Verkehrswirtschaft. In: DVWG (ed) Der Spediteur in der Verkehrswirtschaft, Schriftenreihe der Deutschen Verkehrswissenschaftlichen Gesellschaft e.V. (DVWG) No. B17, 1–12
Seiffert B (1986) Zahlungsmittelbewirtschaftung im Verkehrsbetrieb, Duncker und Humblot: Berlin (also Doctoral Thesis at Technische Universitat Berlin) ¨
Seiler U (1981) Die Funktion des Spediteurs in der schweizerischen Verkehrswirtschaft. Doctoral Thesis at Universitat Bern ¨
Selviaridis K, Spring M (2007) Third party logistics: a literature review and research agenda. Int J Logist Manag 18(1):125–150
Semeijn J, Vellenga DB (1995) International logistics and one-stop shopping. Int J Phys Distrib Logist Manag 25(10):26–44
Severin H (1992) Entwicklung eines Distributionsplanungs- und -steuerungssystems f¨ur speditionelle Dienstleistungen, Huss: Munich (also Doctoral Thesis at Technische Universit at Berlin) ¨
Shalaby MIM (1988) Analysis of container control systems adapted for Egyptian container terminals, Doctoral Thesis at Universitat Rostock ¨
Shapiro SS, Wilk MB (1965) An analysis of variance tests for normallity (Complete samples). Biometrika 52(3-4):591–611
Shavell S (1979a) Risk sharing and incentives in the prinicpal and agent relationship. Bell J Econ 10(1):55–73
Shavell S (1979b) On moral hazard and insurance. Q J Econ 93(4):541–562
Shaw ME (1954) Group structure and the behavior of individuals in small groups. J Psychol 38:139–149
Shaw S (1999) Airline marketing and management, 4th edn. Ashgate, Aldershot
Shaw S (2004) Airline marketing and management, 5th edn. Ashgate, Aldershot
Sheffi Y (1990) Third Party Logistics present and future prospects. Journal of Business Logistics 11(2):27-39
Shelanski HA, Klein PG (1995) Empirical research in transaction cost economics: a review and assessment. J Law Econ Organ 11(2):335–361
Shenkar O (2001) Cultural distance revisited: towards a more rigorous conceptualization and measurement of cultural differences. J Int Bus Stud 32(3):519–535
Sheppard EJ, Seidman D (2001) Ocean shipping alliances: the wave of the future? Int J Marit Econ 3(4):351–367
Sheppard BH, Tuchinsky M (1996) Interfirm relationships: a grammar of pairs. Res Organ Behav 18:331–373
Sherwood CS, Burns R (1992) Solving international transportation problems. Rev Bus 14(1):25–31 Shimbel A (1953) Structural parameters of communication networks. Bull Math Biophys 15(4):501–507
Shleifer A, Vishny RW (1993) Corruption. Q J Econ 108(3):599–617
Shubik M (1991) Fiat money. In: Eatwell J, Milgate M, Newman P (eds) The new palgrave: a dictionary of economics, vol 2 (Reprint). Macmillan, New York, pp 316–317
Shulman N (1976) Network analysis: a new addition to an old bag of tricks. Acta Sociologica 19(4):307–323
Siebert H (1991) Okonomische Analyse von Unternehmensnetzwerken, Managementforschung ¨ 1:291–311
Siegel S (1956) Nonparametric statistics for the behavioral sciences. MacGraw-Hill, New York Siegmund H (2007) Macht durch B¨undelung, Logistik inside No.11/2007, 32–34
Silberstein F (1914) Arbeitertarifvertrage im Speditionsgewerbe, Jacob: Erlangen (also Doctoral ¨ Thesis at Universitat Erlangen) ¨

Silverman BS, Nickerson JA, Freeman J (1997) Profitability, transactional alignment, and organizational martality in the U.S. trucking industry. Strat Manag J 18(Summer Special Issue):31–52 Simmel G (1908) Soziologie, Duncker and Humblot: Berlin

Simon HA (1945) Administrative behavior a study of decision-making processes in administrative organization. Free Press, New York
Simon HA (1957) Models of man social and rational. Wiley, New York
Simon HA (1991a) Bounded rationality. In: Eatwell J, Milgate M, Newman P (eds) The new palgrave: a dictionary of economics, vol 1 (Reprint). Macmillan, New York, pp 266–268
Simon HA (1991b) Satisficing. In: Eatwell J, Milgate M, Newman P (eds) The new palgrave: a dictionary of economics, vol 4 (Reprint). Macmillan, New York, pp 243–245
Simon R-P (1993) Euro-Logistik-Netzwerke: Entwicklung eines wettbewerbsstrategischen Integrationskonzeptes f¨ur die Sammelgut-Logistik im europaischen Markt, Eul: Bergisch Gladbach, ¨ Cologne (also Doctoral Thesis at Universitat Wuppertal 1992) ¨
Sink HL, Langley CJ Jr. and Gibson BJ (1996) Buyer observations of the US third-party logistics market. Int J Phys Distrib Logist Manag 28(3):38–46
Sinnig E (1989) Krise und Entwicklungstendenzen der Linienschiffahrtskartelle unter dem Einfluss der Containerisierung, Doctoral Thesis at Universitat Rostock ¨
Sjogren M (2004) Antonovs Mrija wieder im Gespr ¨ a ch, Deutsche Verkehrszeitung No.95/2004 ¨ (02.08.2004), 7
Sjostrom W (2004) Ocean shipping cartels: a survey. Rev Netw Econ 3(2):107–134
Skjøtt-Larsen T (1999a) Supply chain management: a new challenge for researchers and managers in logistics. Int J Logist Manag 10(2):41–53
Skjøtt-Larsen T (1999b) Interorganisational relations from a supply chain management point of view. Logistik Management 1(2):96–108
Skjøtt-Larsen T (2000) Third party logistics from an interorganizational point of view. International Journal of Physical Distribution & Logistics Management 30(2):112–127 [remark: "oe" is the danish "o" - see Skjoett-Larsen T (1999a,b)]
Sklivas SS (1987) The strategic choice of managerial incentives. Rand J Econ 18(3):452–458
Slack B (1998) Intermodal transportation. In: Hoyle B, Knowles R (eds) Modern transport geography. Wiley, Chichester
Slack B, Comtois C, McCalla R (2002) Strategic alliances in the container shipping industry: a global perspective. Marit Pol Manag 29(1):65–76
Slack B, Comtois C, Sletmo G (1996) Shipping lines as agents of change in the port industry. Marit Pol Manag 23(3):289–300
Slack B, Fremont A (2005) Transformation of port terminal operations: from the local to the global. ´ Transport Rev 25(1):117–130
Smelser NJ, Swedberg R (1994) The sociological perspective on the economy. In: Smelser NJ, Swedberg R (eds) The handbook of economic sociology. Princeton University Press, Chichester, pp 3–26
Smirnov NV (1939a) Sur les ecarts de la courbe de distribution empirique, Matematicheskij ´ Sbornik / Recueil Mathe matique 6(1):3–26 [in Russian with French summary] ´
Smirnov NV (1939b) On the estimation of the discrepancy between empirical curves of distribution for two independent samples, Bulletin de l'Universite d' ´ Etat ´ a Moscou: S ` erie internationale, ´ Section A 2:3–16
Smith P (1974) Air freight operations, marketing and economics. Faber and Faber, London
Smith PB (2002) Culture's consequences: something old and something new. Hum Relat 55(1):119–135
Smith BC, Leimkuhler JF, Darrow RM (1992) Yield management at American Airlines. Interfaces 22(1):8–31
Snijders TAB (1981) The degree variance: an index of graph heterogeneity. Soc Network 3(3): 163–174
Snoek G (1967) Wettbewerbs- und Wachstumsprobleme im internationalen Luftfrachtverkehr, Mohr (Paul Siebeck): T¨ubingen (also Doctoral Thesis at Universitat zu Kiel 1966) ¨

Snow CC, Miles RE, Coleman Jr. HJ (1992) Managing the 21st century network organizations. Organ Dynam 20(3):5–20

Snowdon JL, Paleologo G (2008) Airline optimization. In: Ravindran R (ed) Operations research and management science handbook. CRC Press, Boca Raton
Søderberg AM, Holden N (2002) Rethinking cross cultural management in a globalizing business world. Int J Cross Cult Manag 2(1):103–121
Solandt O (1955) Observation, experiment, and measurement in operations research. J Oper Res Soc Am 3(1):1–19
Søndergaard M (1994) Research note: Hofstede's consequences: a study of reviews, citations and replications. Organ Stud 15(3):447–456
Spence M (1973) Job market signaling. Q J Econ 87(3):355–374
Spence M (1974) Competitive and optimal response to signals: analysis of efficiency and distribution. J Econ Theor 7(3):296–332
Spence M, Zeckhauser R (1971) Insurance, information, and individual action. Am Econ Rev 61(2):380–387
Spens KM, Kovacs G (2006) A content analysis of research approaches in logistics research. Int J ´ Phys Distrib Logist Manag 36(5):374–390
Spera K (2002) Handel und Transport Handbuch f¨ur die G¨uterbeforderung in den Außen- ¨ wirtschaftsbeziehungen, logotrans: Wien
Spera K (2011) Handel und Transport Handbuch f¨ur die G¨uterbeforderung in den Außen- ¨ wirtschaftsbeziehungen (2nd ed.), logotrans: Wien
Spremann K (1987) Agent and principal. In: Bamberg G, Spremann K (eds) Agency theory, information, and incentives. Springer, Berlin u.a., pp 3–37
Spremann K (1988) Reputation, Garantie, Information, Zeitschrift f¨ur Betriebswirtschaft 58(5-6):613–629
Spremann K (1990) Asymmetrische Information, Zeitschrift f¨ur Betriebswirtschaft 60(5-6): 561–586
Spulber DF (1985) Risk sharing and inventories. J Econ Behav Organ 6(1):55–68
Spulber DF (1996a) Market making by price-setting firms. Rev Econ Stud 63(217):559–580
Spulber DF (1996b) Market microstructure and intermediation. J Econ Perspect 10(3):135–152
Spulber DF (1999) Market microstructure: intermediaries and the theory of the firm. Cambridge University Press, Cambridge
Spulber DF (2003) The intermediation theory of the firm: integrating economic and management approaches to strategy. Managerial and decision economics 24(4):253–265
Stabell CB, Fjeldstad OD (1998) Configuring value for competitive advantage: on chains, shops and networks. Strat Manag J 19(5):413–437
Stahl DO II (1988) Bertrand competition for inputs and Walrasian outcomes. Am Econ Rev 78(1):189–201
Stahl D (1995a) Internationale Speditionsnetzwerke: eine theoretische und empirische Analyse im Lichte der Transaktionskostentheorie, Vanderhoeck und Ruprecht: Gottingen (also Doctoral ¨ Thesis at Universitat M¨ ¨ unster 1994)
Stahl D (1995b) Die organisatorische Gestaltung internationaler Speditionsnetzwerke, Internationales Verkehrswesen 47(7-8):437–444
Stahlberg M (1931) Organisation und betriebswirtschaftliche Grundprobleme der deutschen Handelsluftfahrt, Doctoral Thesis at Universitat Frankfurt am Main ¨
Stamm T (1908) Die Pflichten des Spediteurs. R. Noske, Borna-Leipzig (also Doctoral Thesis at Universitat Leipzig) ¨
Stank TP, Maltz AB (1996) Some propositions on third party choice: domestic vs. international logistics providers. J Market Theor Pract 4(2):45–54
Stapleton A, Hanna JB (2002) Technological innovation adoption: an empirical investigation of steamship line sales force integration. Transport J 41(4):5–22
Steenken D, Voß S, Stahlbock R (2004) Container terminal operation and operations research a classification and literature review. OR Spectrum 26(1):3–49

Stefansson G (2004) Collaborative logistics management, Chalmers Reproservice and Vasastadens Bokbinderi: Go teborg (also PhD Thesis at Chalmers University of Technology) ¨

Stefansson G (2006) Collaborative logistics management and the role of third-party providers. Int J Phys Distrib Logist Manag 36(2):76–92
Steffen S (1996) Wettbewerbsstrategien f¨ur deutsche Speditionen: am Beispiel des Marktes f¨ur K¨uhlgut-Logistik, Gabler: Wiesbaden (also Doctoral Thesis at Universitat Bielefeld 1994) ¨
Steiger D (2006) Vorgebaute Einheiten verandern den Markt, Deutsche Verkehrszeitung ¨ No.105/2005 (02.09.2006), 5
Steininger S (2000) Lead Logistic Provider, Jahrbuch der Logistik 2000, Verlagsgruppe Handelsblatt: D¨usseldorf, 112–114
Stephenson K, Zelen M (1989) Rethinking centrality: methods and examples. Soc Network 11(1):1–37
Stern H (1938) Die Grundlagen des Wettbewerbes und der Geschaftspolitik in der internationalen ¨ Spedition, Eigenverlag: Vienna (also Doctoral Thesis at Hochschule f¨ur Welthandel Wien)
Stern LW, El-Ansary AI (1992) Marketing channels, 4th edn. Prentice Hall, Englewood Cliffs
Stigler GJ (1966) The theory of price, 3rd edn. Macmillan, New York
Stiglitz JE (1974) Inventives and risk sharing in sharecropping. Rev Econ Stud 41(126):219–255
Stiglitz JE (1975a) Inventives, risk, and information: Notes towards a theory of hierarchy. Bell J Econ 6(2):552–579
Stiglitz JE (1975b) The theory of "screening", education, and the distribution of income. Am Econ Rev 65(3):283–300
Stiglitz JE, Weiss A (1981) Credit rationing in markets with imperfect information. Am Econ Rev 71(3):393–410
Stock JR (1988/1989) A compendium of doctoral research in logistics: 1970–1986. J Bus Logist 8(2):123–202 and 9(1):125–233
Stock JR (1997) Applying theories from other disciplines to logistics. Int J Phys Distrib Logist Manag 27(9):515–539
Stock JR (2001) Doctoral research in logistics and logistics-related areas: 1992–1998. J Bus Logist 22(1):125–256
Stock JR, Broadus CJ (2006) Doctoral research in supply chain management and/or logisticsrelated areas: 1999–2004. J Bus Logist 27(1):139–496
Stock JR, Luhrsen DA (1993) Doctoral research in logistics-related areas: 1987–1991. J Bus Logist 14(2):197–373
Stoll HR (1978) The supply of dealer services in securities markets. J Finance 33(4):1133–1151
Stoll HR (1985) Alternative views of market making. In: Amihud Y, Ho SY (eds) Market Making and the Changing Structure of the Securities Industry, DC Health, Lexington, pp. 67-92
Stoll HR (2003) Market microstructure. In: Constantinides G, Harris M, Stulz RM (eds) Handbook of the economics of finance vol 1: Corporate finance. Elsevier, Amsterdam, pp 553–604
Stolte JF, Emerson RM (1977) Structural equality: position and power in network structures. In: Hamblin R, Kunkel J (eds) Behavioral theory in sociology. Transaction, New Brunswick, pp 117–138
Stone M (1998) The expansion of UK-based logistics service providers into continental Europe, PhD-Thesis at Heriot-Watt University
Stone MA (2001), European expansion of UK third-party logistics service providers. Int J Logist Res Appl 4(1):97–115
Stopford M (2000) Maritime economics, 2nd edn. Routledge, London
Stopford M (2002) E-commerce-implications, opportunities, and threats for the shipping business. Int J Transport Manag 1(1):55–67
Stottinger B, Schlegelmilch BB (1998) Explaining export development through psychic distance: ¨ enlightening or elusive? Int Market Rev 15(5):357–372
Strandenes SP (2000) The shipbroking function and market efficiency. Int J Marit Econ 2(2):17–26 Straube M (2003) Kommentar zum Handelsgesetzbuch (3rd ed.), Manz:Wien
Strausz R (1997) Delegation of monitoring in a principal-agent relationship. Rev Econ Stud 64(3):337–357

Suelflow JE, Hille J (1970) The transportation company: an economic argument for intermodal ownership. Land Econ 46(3):275–286

Suhl L (1995) Computer-aided Scheduling: An Airline Perspective, DUV: Wiesbaden (also Habilitation Thesis at Technische Universitat Berlin 1993) ¨
Summers CW (1969) Collective agreements and the law of contract. Yale Law J 78(4):525–575
Sustrate V (1986) Determinaten der Entwicklung des Luftfrachtverkehrs und Quantifizierung k¨unftiger Luftfrachtpotetiale – aufgezeigt am Beispiel der Bundesrepublik Deutschland, Institut f¨ur Verkehrswesen, Eisenbahnbau und -betrieb: Hannover (also Doctoral Thesis at Universitat¨ Hannover)
Svendsen AS (1958) Seeverkehr und Schiffahrtswirtschaft Sea Transport and Shipping Economics, Institut f¨ur Schiffahrtsforschung: Bremen
Swedberg R (1991) Major traditions of economic sociology. Annu Rev Sociol 17:251–276
Swedberg R (1997) New economic sociology: what has been accomplished, what is ahead? Acta Sociologica 40(2):161–182
Sydow J (1992) Strategische Netzwerke und Transaktionskosten, Managementforschung 2:239–311
Sydow J (1993) Strategische Netzwerke: Evolution und Organisation (Reprint), Gabler: Wiesbaden Sydow J (1998) Franchise systems as strategic networks: studying network leadership in the service sector. Asia Pac J Market Logist 10(2):108–120
Sydow J, van Well B, Windeler A (1998) Networked networks. Int Stud Manag Organ 27(4):47–75 Sydow J, Windeler A, Krebs M, Loose A, van Weil B (1995) Organisation von Netzwerken: Strukturtheoretische Analysen der Vermittlungspraxis in Versicherungsnetzwerken, Westdeutscher Verlag: Opladen
Taube VG (2002) Zur Messung des Sozialkapitals von Akteuren mit Einfluß in empirischen ¨ Netzwerken, Lang: Bern (also Doctoral Thesis at Universitat Essen) ¨
Taube VG (2004) Measuring the social capital of brokerage roles. Connections 26(1):29–52 ¨
Talley WK (2000) Ocean container shipping: Impacts of a technological improvement. J Econ Issues 34(4):933–948
Tansel BC, Francis RL, Lowe TJ (1983a) Location on networks: A survey. Part I: The p-center and p-median problems. Manag Sci 29(4):482–497
Tansel BC, Francis RL, Lowe TJ (1983b) Location on networks: A survey. Part II: Exploiting tree network structure. Manag Sci 29(4):498–511
Taylor M (1969) Influence structures. Sociometry 32:490–502
Taylor CR (1995) The economics of breakdowns, checkups, and cures. J Polit Econ 103(11):53–74 Taylor SS, Fisher D, Dufresne RL (2002) The aesthetics of management storytelling: A key to organisational learning. Manag Learn 33(3):313–330
Taylor JC, Jackson GC (2000) Conflict, power, and evolution in the intermodal transportation industry's channel of distribution. Transport J 39(3):5–17
Telser LG (1980) A theory of self-enforcing agreements. J Bus 53(1):27–44
Templin C (2007) Bodenabfertigungsdienste an Flughafen in Europa, K ¨ olner Wissenschaftsverlag: ¨ Koln (also Doctoral Thesis at Universit ¨ at zu K ¨ oln 2006) ¨
Teodorovic D (1986) Transport networks A qua ´ ntitative treatment. Gordon and Breach, New York
Teodorovic D (1988) Airline operations research. Gordon and Breach, New York ´
Terhorst R (1992) Das Modell eines Frachtflughafens und sein Beitrag zur logistischen Optimierung der Luftfrachttransportkette, Doctoral Thesis at Universitat zu K ¨ oln ¨
Tetley W (1992) Legal effects of development in technology. In: Dolman AJ, van Ettinger J (eds) Ports as nodal points in a global transport system. Pergamon Press, London, pp 177–186
Teutsch C (2001) Spezielle Probleme des multimodalen Transports unter besonderer Ber¨ucksichtigung der Binnenschiffahrt. In: Riedel E (ed) Multimodal Transport und Binnenschiffahrt. Nomos, Baden Baden, pp 53–63
Thaler SP (1990) Betriebswirtschaftliche Konsequenzen des EG-Binnenmarktes und der EG-G¨u terverkehrsliberalisierung f¨ur europaische Speditionsunternehmungen, Haupt: Stuttgart (also ¨ Doctoral Thesis at Universitat Basel) ¨

Theilen B (1996) Optimal contract and organizational design under moral hazard and adverse selection. Eul, Bergisch Gladbach

Thibaut JW, Kelley HH (1959) The social psychology of groups. Wiley, New York

Thomas CJ (1996) Military operations research. In: Gass SI, Harris CM (eds) Encyclopedia of operations research and management science. Kluwer, Boston, pp 394–398

Thomas C, Campbell K, Hines G, Racer M (1998) Airbus packing at federal express. Interfaces 28(4):21–30

Thomchick EA, Rosenbaum L (1984) The role of U.S. export trading companies in international logistics. J Bus Logist 5(2):85–105

Thomchick E, Young RR, Ruamsook K (2004) Importing and exporting: A longitudinal study of process component importance and expertise. Transport J 43(3):34–47

Thorby C (1997) US customs are changing. Containerization Int 2/1997:36–39

Thorby C (2001a) A burdensome act. Containerization Int 3/2001:73–77

Thorby C (2001b) Value-added carriers. Containerization Int 4/2001:52–57

Thorby C (2001c) Close encounters of the fourth kind. Containerization Int 5/2001:42–47

Thorelli HB (1986) Networks: Between markets and hierarchies. Strat Manag J 7(1):37–51

Thost HW (1921) Konzentrationsformen des Speditionsgewerbes, Doctoral Thesis at Universitat¨ Hamburg

Tiebel J (1925) Die Einwirkungen des Krieges auf das deutsche Speditionsgewerbe, Doctoral Thesis at Universitat Jena ¨

Tietzel M (1981) Die Okonomie der Property Rights: Ein ¨ Uberblick. Z Wirtschaftspolitik 30: ¨ 207–243

Tinic SM (1972) The economics of liquidity services. Quart J Econ 86(1):79–93

Tirole J (1988) The theory of industrial organization. MIT-Press, Cambridge

Tirole J (1999) Incomplete contracts: Where do we stand? Econometrica 67(4):741–781

Tolsto˘ı AN (1930) Metody nakhozhdeniya naimen'shego summovogo kilometrazha pri planirovanii perevozok v prostranstve, Planirovanie perevozok, vol 1. Transpechat' NKPS, Moscow, pp 23–55 [in Russian]

Tolsto˘ı AN (1939) Metody ustraneniya neratsional'nykh perevozok pri planirovanii. Sotsialistichesk ˘ı Transport 9:28–51 [in Russian]

Townsend RM (1979) Optimal contracts and competitive markets with costly state verification. J Econ Theor 21(2): 265–293

Trabold H (2002) Export intermediation: An empirical test of Peng and Ilinitch. J Int Bus Stud 33(2):327–344

Trappe J (1995) Der Schiffsagent – einige rechtliche Aspekte. Transportrecht 9/1995:321–325

Trettin H (1977) Vergleichende Untersuchungen von Transporttechnologien mit der Ableitung von Aussagen¨uber Moglichkeiten und Grenzen des Containertransports in der DDR sowie ¨ Vorschlagen zu seiner effektivsten Entwicklung, Doctoral Thesis at Hochschule f¨ ¨ ur Verkehrswesen Dresden

Triebel V, Hodgson S, Kellenter W, M¨uller G (1995) Englisches Handels- und Wirtschaftsrecht, 2nd rev. edn. Verlag Recht und Wirtschaft, Heidelberg

Trommer A (2000a) IATA-Agent. In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, Munich

Trommer A (2000b) Luftfrachtmakler. In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, Munich

Trommer A (2000c) Luftfrachtspediteur. In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, Munich

Trompenaars F (1993) Handbuch Globales Managen – Wie man Kulturelle Unterschiede im Geschaftsleben versteht. Econ Verlag, D¨ ¨ usseldorf – Vienna - New York - Moskau, [German ed. of F. Trompenaars (1993) Riding the Waves of Culture: Understanding Cultural Diversity in Business, Economist Books, London]

Trompenaars F, Hampden-Turner C (2002) Riding the waves of culture: Understanding cultural diversity in business, 2nd edn. Nicholas Brealey, London

Trost DG (1999) Vernetzung im G¨u terverkehr: okonomische Analyse von Zielen, Ansatzpunkten ¨ und Maßnahmen zur Implementierung integrierter Verkehrssysteme unter Ber¨u cksichtigung logistischer Anspr¨uche verschiedener Marktsegmente, Deutscher Verkehrs Verlag: Hamburg (also Doctoral Thesis at Universitat Gießen) ¨

Tschudi J-P (1975) Die Vertrage des Speditionsgesch ¨ afts unter Ber¨ ¨ ucksichtigung der Allgemeinen Bedingungen des Schweizer Spediteur-Verbandes, Buch + Offsetdruck Stadelmann: Z¨urich (also Doctoral Thesis at Universitat Z¨ ¨ urich 1974)
Turnbull TW, Valla J (1986) Strategies for international industrial marketing the management of customer relationships in European industrial markets. Croom Helm, London
Turner RH (1985) Unanswered questions in the convergence between structuralist in interactionist role theories. In: Helle HJ, Eisenstadt SN (eds) Perspectives on sociological theory, vol 2: Microsociological theory, Sage, Beverly Hills, pp 22–36
Uhlig T (1996) Logistik-outsourcing: Eine lukrative Tatigkeit f¨ ¨ ur Verlader und Dienstleister. Internationales Verkehrswesen 48(4):39–43
Uhlmann S-J (2004) Okologieorientierte Die ¨ nstleistungspolitik von Speditionen in logistischen Allianzen, Doctoral Thesis at Universitat Bremen ¨
Ulderup G (1924) Spedition und Lufttransportwesen. In: Fischer v. Poturzyn and Jurinek JM (eds) Jahrbuch f¨ur Luftverkehr 1924. Richard Pflaum, Munich, pp 199–204
Ulderup J (1935) Der Stand des Weltluftverkehrs und seine Probleme insbesondere in Deutschland, Junker und D¨unnhaupt: Berlin (also Doctoral Thesis at Technische Hochschule Berlin)
Ullmann EM (1950) Entwicklung und Funktionen der internationalen Spedition, Doctoral Thesis at Hochschule f¨ur Welthandel Wien
UN/CEFACT (2000) BPAWG Model of The International Supply Chain Model (Interim Report UN/CEFACT/BPAWG/BP105.Rev.2)
UN/CEFACT (2001) UN/CEFACT's Modeling methodology(CEFACT/TMWG/N090R10 November 2001) Techniques and Methodologies Working Group
UN/ECE (2001) Terminology of Combined Transport, United Nations: New York Geneva
Uzzi B (1996) The sources and consequences of embeddedness for the economic performance of organizations: The network effect. Am Soc Rev 61(4):674–698
Uzzi B (1997) Social structure and competition in interfirm networks: The paradox of embeddedness. Admin Sci Quart 42(1):35–67
Vadnai E (1921) Die Stellung des Spediteurs in der arbeitsteiligen Verkehrswirtschaft, Doctoral Thesis at Universitat Hamburg ¨
Vahrenkamp (2002) Die Rolle der Luftfracht in der internationalen Logistik eine Ubersicht, ¨ Logist Manag 4(4):13–22
Vaile R, Grether ET, Cox R (1952) Marketing in the american economy. Ronald Press, New York Valder H (2002) Allgemeine Deutsche Spediteursbedingungen (ADSp) 2003 – Inhalte und Auswirkungen im U berblick. In: Hector B, Brauch M, Knorre J, Salzmann A, Thonfeld ¨ H-D, Valder H, Waldner K (eds) Allgemeine Deutsche Spediteursbedingungen (ADSp) 2003. Hamburg, DVZ, pp 9–20
Vanberg VJ (1998a) Austrian school of economics and the evolution of institutions. In: Newman P (ed) The new palgrave dictionary of economics and the law, vol 1. Macmillan, London - Basingstoke, pp 134–140
Vanberg VJ (1998b) Freiburg school of law and economics. In: Newman P (ed) The new palgrave dictionary of economics and the law, vol 2. Macmillan, London - Basingstoke, pp 172–179
Vanelslander T (2005) The economics behind co-operation and competition in seaport container handling, Ph.D-Thesis at Universiteit Antwerpen, Antwerp (UMI 3220057)
Veblen TB (1898) Why is economics not an evolutionary science? Quart J Econ 12(2):373–397
Veblen TB (1899) The theory of the leisure class: An economic study in the evolution of institutions. Macmillan, New York
Veblen TB (1900) The preconceptions of economic science. Quart J Econ 14(2):240–269
Veblen TB (1909) The limitations of marginal utility. J Polit Econ 17(9):620–636
Veblen TB (1919) The place of science in modern civilization and other essays. Huebsch, New York

Vertinsky I (1998) Operations research. In: Eatwell J, Milgate M, Newman P (eds) The new palgrave: A dictionary of economics, vol 3 (Reprint), Macmillan, New York, pp 713–715

Vickers W (1984) Delegation and the theory of the firm. Econ J 95/Supplement:8–37
Victor-Granzer B (2005) Haftungsrahmen und Anspruchsverjahrung nach franz ¨ ossichem Fracht- ¨ und Speditionsrecht. Transportrecht 9/2005:354–358
Virum H (1993) Third party logistics development in Europe. Logist Transport Rev 29(4):355–36 Vis IFA, de Koster R (2003) Transhipment of containers at a container terminal: An overview. Eur J Oper Res 147(1):1–16
Vogel W (1947) Die Organisation im modernen Speditionsbetrieb, Doctoral Thesis at Hochschule f¨ur Welthandel Wien
Vogler W (1924) Der Spediteur im Landverkehr unter Zugrundelegung der Geschaftsorganisation ¨ einer bestimmten Frankfurter Speditionsfirma, Doctoral Thesis at Universitat Frankfurt am ¨ Main
Vonnegut H (1921) Die wirtschaftliche Bedeutung des Sammelgutverkehrs der Spediteure, Doctoral Thesis at Universitat Heidelberg ¨
Voss H (1996) Virtual organizations: The future is now. Strat Leader 8/1996:12–16
Wagner J (1952) Ermittlung des moglichen Luftg¨ ¨ uterexportaufkommens f¨ur Norddeutschland und das Ausfuhrgeschaft¨ ¨ uber den Flughafen Hamburg, Doctoral Thesis at Universitat Hamburg ¨ 1953
Waldenberger P (1997) Analyse, Kritik und Evaluation des Ansatzes von Miles-Snow anhand des Beispiels eines internationalen Speditionskonzernes, Doctoral Thesis at Universitat Innsbruck ¨ 1998
Walker G, Kogut B, Shan W (1997) Social capital, structural holes and the formation of an industry network. Organ Sci 8(2):109–125
Walter K (1971) Die volkswirtschaftliche Bedeutung des Containerverkehrs, Doctoral Thesis at Universitat zu K ¨ oln ¨
Wasserman S, Faust K (1994) Social network analysis. Cambridge University Press, Cambridge Wathne KH, Heide JB (2004) Relationship governance in a supply chain network. J Market 68(1):73–89
Wecker G (1969) Kostenrechnung im Speditionsbetrieb, Doctoral Thesis at Universitat M¨ ¨ unchen Weddewer M (2008) Verrechnungspreissysteme f ¨ur horizontale Speditionsnetzwerke: Simulationsgest¨utzte Gestaltung und Bewertung, DUV: Wiesbaden (also Doctoral Thesis at Universitat¨ Paderborn 2007)
Wegdell S (1996) Die nationalen skandinavischen Strasseng¨utertransportrechte und die Haftung nach schwedischem, norwegischem, finnischem und danischem Recht, Lit: Hamburg (also ¨ Doctoral Thesis at Universitat Hamburg, 1995–1996) ¨
Weimann G (1989) Social networks and communication. In: Asante MK, Gudykunst WB (eds) Handbook of international and intercultural communication. Sage, Newbury Park, pp 186–203
Weir JD, Thomas MU (2008) Military applications. In: Ravindran R (ed) Operations research and management science handbook. CRC Press, Boca Raton
Weisel H (1934) Das N¨urnberger Speditionsgewerbe im Rahmen des deutschen Speditionsgewerbes, Osterchrist: Nuremberg (also Doctoral Thesis at Handelshochschule N¨urnberg)
Weiss SE (1994a) Negotiating with "Romans" Part 1. Sloan Manag Rev 35(2):51–61
Weiss SE (1994b) Negotiating with "Romans" Part 2. Sloan Manag Rev 35(3):85–99
Weisskopf G (1984) Der Luftfrachtmarkt, unter besonderer Ber¨ucksichtigung der Beziehungen zwischen Luftverkehrsgesellschaften und Luftfrachtspediteuren, Juris: Z¨urich (also Doctoral Thesis at Hochschule St. Gallen)
Welch DE, Welch LS, Marschan-Piekkari R (2001) The persistent impact of language on global operations. Prometheus 19(3):193–209
Welch C, Wilkinson I (2004) The political embeddedness of international business networks. Int Market Rev 21(2):216–231
Welker CB (1988) Anforderungen an logistische Dienstleistungen der Speditionsbetriebe unter Ber¨ucksichtigung eines ge anderten Bestellverhaltens industrieller Unternehmungen, Verkehrs- ¨ Verlag Fischer: D¨usseldorf (also Doctoral Thesis at Universitat zu K ¨ oln) ¨

Wellman B (1988) Structural analysis: From method and metaphor to theory and substance. In: Wellman B, Berkowitz SD (eds) Social structures: A network approach. Cambridge University Press, Cambridge, pp 19–61

Wells AT (1999) Air transportation a management perspective. Wadsworth, Belmont
Wendorff J (1984) Auswirkungen der Energiepreisentwicklung auf den Luftfrachtverkehr, Doctoral Thesis at Universitat K ¨ oln ¨
Wendorff J (1986) Auswirkungen der Energiepreisentwicklung auf den Luftfrachtverkehr, Doctoral Thesis at Universitat zu K ¨ oln ¨
Wernerfelt B (1988) Umbrella branding as a signal of new product quality: An example of signalling by posting a bid. Rand J Econ 19(3):458–466
Westenberg J (1948) Significance test for median and interquartile range in samples from continuous populations of any form. Proceedings Koninklijke Nederlandse Akademie Van Wetenschappen, vol 51. pp 252–261
Westfall FW (1987) An Analysis of Freight Forwarder Operations in an International Distribution Channel, UMI: Ann Arbor (also PhD Thesis at Ohio State University, Columbus)
Wettbewerbsverein (2004) White Paper 2004: State and partly-state owned companies distort competition in the European freight forwarding and logistics market, Verein zur Forderung des ¨ Wettbewerbs und lauteren Verhaltens im Speditions-, Logistik- und Transportgewerbe e.V. K oln and H.A.L.T.E., Honorable association de transporteurs et logisticiens europ ¨ eens, Paris ´
Weyer J (2000) Zum stand der netzwerkforschung in den sozialwissenschaften. In: Weyer J (ed) Soziale netzwerke: Konzepte und methoden der sozialwissenschaftlichen netzwerkforschung. Oldenbourg, Vienna, pp 1–35
White HC (1992) Agency as control in formal networks. In: Nohria N, Eccles RG (eds) Networks and organizations: Structure, form, and action. Harvard Business School Press, Harvard, pp 92–117
White HC, Boorman SA, Breitger RL (1976) Social structure from multiple networks. I. Blockmodels of roles and positions. Am J Sociol 81(4):730–780
White DR, Owen-Smith J, Moody J, Powell WW (2004) Networks, fields and organizations: Micro-dynamics, scale and cohesive embeddings. Comput Math Decis Theor 10(1):95–117
Wiener H (1947) Structural determination of paraffin boiling points. J Am Chem Soc 69(1):17–20 Wieske T (2002) Zukunft der ADSp – wie lange ist ein Bedingungswerk noch zeitgemaß? ¨ Versicherungsrecht No.34/2002, 1489–1491
Wieske T (2003) Transportrecht schnell erfasst. Springer, Berlin
Wieske-Hartz HC (2004) Airline Operations: Praxishandbuch f¨ur Personal und Fluggesellschaften und Flughafen. Books-on-Demand, Norderstedt ¨
WIFI (1993) Erhebung von Strukturdaten f¨ur das Speditionsgewerbe 1992, Wirtschaftsforder- ¨ ungsinstitut Osterreich: Wien ¨
Wilcoxon F (1945) Individual comparisons by ranking methods. Biometrics Bull 1(6):80–83
Willer D (1992) Predicting power in exchange networks: A brief history and introduction to the issues. Soc Network 14(3-4):187–211
Willer D (1999) Network exchange theory. Praeger, New York
Willer D (2003) Power-at-a-distance, Soc Forces 81(4):1295–1334
Williams EW (1973) International goods movement: New problems and opportunities. Columbia J World Bus 8(3):107–112
Williamson OE (1975) Markets and hierarchies: Analysis and antitrust implications. Free Press, New York
Williamson OE (1976) Franchise bidding for natural monopolies in general and with respect to CATV. Bell J Econ 7(1):73–104
Williamson OE (1979) Transaction-cost economics: The governance of contractual relations. J Law Econ 22(2):233–261
Williamson OE (1981) The economics of organization: The transaction cost approach. Am J Socio 87(3):548–577
Williamson OE (1983) Credible commitments: Using hostages to support exchange. Am Econ Rev 73(4):519–540

Williamson OE (1985) The economic institutions of capitalism: Firms, markets, relational contracting. Free Press, New York

Williamson D (1986) Costly monitoring, financial intermediation, and equilibrium credit rationing. J Monetary Econ 18(2):159–179
Williamson OE (1991) Comparative economic organization: The analysis of discrete structural alternatives. Admin Sci Q 36(2):269–296
Williamson OE (1996) The mechanisms of governance. Oxford University Press, New York
Williamson OE (1998a) The institutions of governance. Am Econ Rev 88(2):75–79
Williamson OE (1998b) Opportunistic behavior in contracts. In: Newman P (ed) The new palgrave dictionary of economics and the law, vol 2. Macmillan, Basingstoke, pp 703–710
Williamson OE (1999) The mechanisms of governance. Oxford University Press, New York [paperback ed. of Williamson (1996)]
Williamson OE (2000) The new institutional economics: Taking stock, looking ahead. J Econ Lit 38(3):595–613
Williamson D (2002) Forward from a critique to Hofstede's model of national culture. Hum Relat 55(11): 1373–1395
Williamson NC, Bello DC (1984) Export management companies: The implications for developing the American export trading company. Int Market Rev 1(4):24–39
Wilson C (1977) A model of insurance markets with incomplete information. J Econ Theory 16(2):167–207
Wilson C (1980) The nature of equilibrium in markets with adverse selection. B J Econ 11(1): 108–130
Win A (2008) The value a 4PL provider can contribute to an organisation. Int J Phys Distrib Logist Manag 38(9):674–684
Windisch E (1996) Marktabgrenzung von Luftfrachtleistungen: Analysen Konzepte Losungsans ¨ atze, Gabler: Wiesbaden (also: Doctoral Thesis at Universit ¨ at Frankfurt-on-Main ¨ 1995)
Witt P (1992) Zur Methodik der Netzwerkanalyse. In: Albach H, Witt P (eds) Transformationsprozesse in ehemals volkseigenen Betrieben. Schaeffer-Poeschel, Stuttgart, pp 17–42
Witt P (1996) Planung betrieblicher Transformationsprozesse. Wiesbaden, Gabler
Wittenbrink P (1995) B¨undelungsstrategien der Speditionen im Bereich der City-Logistik: eine okonomische Analyse, Vandenhoeck und Ruprecht: G ¨ ottingen (also Doctoral Thesis at Univer- ¨ sitat M¨ ¨ unster)
Wlcek H (1998) Gestaltung der G¨u terverkehrsnetze von Sammelgutspeditionen, GVB: Nuremberg (also Doctoral Thesis Universitat Augsburg) ¨
Wolf G (1967) Die Entwicklung des Weltluftverkehrs nach dem Zweiten Weltkrieg, Mohr (Paul Siebeck): T¨ubingen (also Doctoral Thesis at Universitat zu Kiel 1966) ¨
Wolf M (1999) Kundenorientierung von Transportdienstleistern: Konzeption und Anwendung eines Messinstrumentes, Gabler: Wiesbaden (also Doctoral Thesis at Universitat Erlangen-N¨ ¨ u rnberg 1998)
Wolf D (2000a) Combined transport operator (CTO). In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, Munich
Wolf D (2000b) Haus-Haus-Verkehr. In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, Munich
Wolf D (2000c) Multimodal transport operator (MTO). In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, Munich
Wolf D (2000d) Transportketten. In: Bloech J, Ihde GB (eds) Logistik total [CD-Rom]. Vahlen, Munich
Wolfe AW (1978) The rise of network thinking in anthropology. Soc Network 1(1):53–64
Wolff B, Neuburger R (1995) Zur theoretischen Begr ¨undung von Netzwerken aus der Sicht der Neuen Institutioneno konomik. In: Jansen D, Schubert K ( ¨ eds) Netzwerke und politikproduktion. Sch¨uren, Marburg, pp 74–94
Wolinsky A (1993) Competition in a market for informed experts' services. Rand J Econ 24(3):382–398

Wolters H (1973) Das Verhaltnis des Speditionsgewerbes zu den verschiedenen Verkehrstr ¨ agern ¨ und die Bedeutung des Spediteurs als Wegbereiter des internationalen Verkehrs. In: DVWG (ed) Der Spediteur in der Verkehrswirtschaft, Schriftenreihe der Deutschen Verkehrswissenschaftlichen Gesellschaft e.V. (DVWG) No. B17, 26–36

Wood DF, Barone A, Murphy P, Wardlow DL (1995) International logistics. Amacom, New York Wood DF, Johnson JC (1996) Contemporary transportation, 5th edn. Prentice Hall, New Jersey

Woodbridge C (2002) The big four. Containerization International No.3/2002, 86–91

Woolcock M (1998) Social capital and economic development: Toward a theoretical synthesis and policy framework. Theor Soc 27(2):151–208

Wrong DH (1961) The oversocialized conception of man in modern sociology. Am Socio Rev 26(2):183–193

Wu SD (2004) Supply chain intermediation: A bargaining theoretic framework. In: Simchi-Levi D, Wu SD, Shen Z-J (eds) Handbook of quantitative supply chain analysis: Modeling the e-business era. Kluwer, Boston, pp 67–115

Wu Z, Choi TY (2005) Supplier-supplier relationships in the buyer-seller triad: Building theories from eight case studies. J Oper Manag 24(1):27–52

W¨ust A (1927) Die Bedeutung des Luftverkehrs f¨ur das Wirtschaftsleben, Heider: Bergisch Gladbach (also Doctoral Thesis at Universitat zu K ¨ oln) ¨

Wyremba H-J (1970) Beitrag zur Gestaltung optimaler Container-Umschlagssysteme f¨ur Seehafen, ¨ Doctoral Thesis at Hochschule f¨ur Verkehrswesen Dresden

Yamagishi T, Gillmore MR, Cook KS (1988) Network connections and the distribution of power in exchange networks. Am J Sociol 93(4):833–851

Yamaguchi K (1996) Power in networks of substitutable and complementary exchange relations: A rational choice model and an analysis of power centralization. Am Socio Rev 61(2):308–332

Yanelle M-O (1989) The strategic analysis of intermediation. Eur Econ Rev 33(2)–3:294–301

Yanelle M-O (1997) Banking competition and market efficiency. Rev Econ Stud 64(2):215–239

Yavas A (1991) Market-makers and match-makers: Certainty case. University of Iowa: mimeo

Yavas A (1992) Marketmakers versus matchmakers. J Financ Intermediation 2(1):33–58

Yavas A (1994a) Economics of brokerage: An overview. J R Estate Lit 2(1):169–195

Yavas A (1994b) Middlemen in bilateral search markets. J Lab Econ 12(3):406–429

Yavas A (1995) Seller-broker relationships as a double moral hazard problem. J Hous Econ 4(3):244–263

Yavas A (1996a) Matching buyers and sellers by brokers: A comparison of alternative commission structures. R Estate Econ 24(1):97–112

Yavas A (1996b) Search and trading in intermediated markets. J Econ Manag Strat 5(2):195–216

Yavas A (2001) The immediacy service of the specialist as a coordination mechanism. Int Rev Econ Finance 10(3):205–221

Yavas A, Miceli TJ, Sirmans CF (2001) An experimental analysis of the impact of intermediaries on the outcome of bargaining games. R Estate Econ 29(2):251–276

Yeh R-S, Lawrence JJ (1995) Individualism and confucian dynamism: A note on Hofstede's cultural root to economic growth. J Int Bus Stud 26(3):655–669

Yinger J (1981) A search model of real estate broker behavior. Am Econ Rev 71(4):591–604

Yonay YP (1998) The struggle over the soul of economics – institutionalists and neoclassical economists in america between the wars. Princeton University Press, Princeton

Yu G (1998) Operations research in the airline industry. Kluwer, Boston

Zadek H (2001) Strategische Neuausrichtung von Logistikdienstleistern. Ind Manag 17(5):28–31 Zeleny LD (1940a) Measurement of social status. Am J Sociol 45(4):576–582

Zeleny LD (1940b) Status: It's measurement and control in education. Sociometry 4(2):193–204

Zeleny LD (1941) Measurement of sociation. Am Socio Rev 6(2):173–188

Zenios S (2004) Decentralized decision making in dynamic technological systems. In: Simchi-Levi D, Wu SD, Shen Z-J (eds) Handbook of quantitative supply chain analysis: Modelling the e-business era, Kluwer, Boston - Dordrecht - London, pp 117–140

Zettl W (1979) Effizienz der Luftfracht-Transportkette und Massnahmen zu ihrer Verbesserung, tuduv-Verlagsgesellschaft: Munich (also Doctoral Thesis Universitat M ¨ ¨ unchen)

Ziegler J (1938) Die Bedeutung des Luft-Linienverkehrs im neuzeitlichen Transportsystem, Dittert: Dresden (also Doctoral Thesis at Universitat Leipzig) ¨

Zimmermann G (1992) Logistische Gestaltungsaspekte des gewerblichen Luftfrachtverkehrs: Eine Analyse technisch-wissenschaftlicher Gegebenheiten im zeitlichen Wandel aus deutscher Sicht, Doctoral Thesis at Technische Universitat Berlin ¨
Zinnecker K-H (1972) Technisch-wirtschaftliche Prognose interkontinentaler G¨uterverkehrssysteme unter besonderer Ber¨ucksichtigung genormter Ladeeinheiten, Doctoral Thesis at Universitat Karlsruhe ¨
Zollner WA (1990) Strategische Absatzmarktplanung Kunden und Wettbewerbsanalyse f¨ ¨ ur Logistikunternehmen, Springer, Berlin - Heidelberg - New York (also Doctoral Thesis at Technische Hochschule Darmstadt 1989/1990)
Zsidisin GA, Ellram LM (2003) An agency theory investigation of supply risk management. J Supply Chain Manag 38(3):15–27
Zukin S, DiMaggio P (1990) Introduction. In: Zukin S, DiMaggio P (eds) Structures of capital: The social organization of the economy. JAI Press, Cambridge, pp 1–36

Appendix

For reasons of environmental protection, the appendix is provided via Springer eBook Collection.

schrammfreightforwarderintermediaryrole2012.md 1.1 MB История Исходник

status: converted

Freight Forwarder's Intermediary Role in Multimodal Transport Chains

Hans-Joachim Schramm

Freight Forwarder's Intermediary Role in Multimodal Transport Chains

© Springer-Verlag Berlin Heidelberg 2012

Foreword

Contents

List of Figures

List of Tables

List of Symbols (in order of appearance)

List of Abbreviations

Chapter 1 Introduction

1.1 Freight Forwarding as a Research Topic So Far

1.2 Purpose and Research Objectives

1.3 Research Methodology and Outline

Chapter 2 Legal and Functional Perspective of Freight Forwarding in Multimodal Transport Chains

2.1 The Freight Forwarder from a Legal and Functional Point of View

2.1.1 Actual Legal Status of Forwarding Business in Germany and Austria

"§ 453: Forwarding contract

2.1.2 Comparisons with Legal Status of Forwarding Business in Other Countries

2.1.3 Freight Forwarders from a Functional Point of View

2.1.4 Fields of Activity and Specialization in Forwarding Business

2.2 Freight Forwarding in Multimodal Transport Chains

2.2.1 Introductory Notes and Basic Definitions

2.2.2 The Ocean Shipping Industry in the General Cargo Segment

2.2.2.1 Technological Development and Functional Aspects

2.2.2.2 Institutional Environment and Market Structure

2.2.3 The Cargo Segment in the Airline Industry

2.2.3.1 Institutional Environment and Market Structure

2.2.3.2 Technological Development and Functional Aspects

2.2.4 Mapping of Inter-organizational Interaction and Cargo Movement

2.2.4.1 Traditional Sea Transport Operations of General Cargo

2.2.4.2 Contemporary Sea Transport Operations of Containerized Cargo

2.2.4.3 Contemporary Air Cargo Transport Operations

2.3 Summary and Concluding Remarks on the Functional Perspective

Chapter 3 New Institutional Economics, Intermediation Theory, and the Intermediary Perspective

3.1 New Institutional Economics

3.1.1 Ancedents of the New Institutional Economics

3.1.2 Common Assumptions of New Institutional Economics

3.1.2.1 Methodological Individualism and Institutions

3.1.2.2 Property Rights and Other Contractual Relationships

Absolute Property Rights

Relative Property Rights

Other Forms of Contractual Relationships

3.1.2.3 Positive Transaction Costs

3.1.2.4 Imperfect Information Structure and Bounded Rationality

Imperfect Information Structures

Bounded Rationality

3.1.3 Building Blocks of New Institutional Economics

3.1.3.1 Transaction Cost Theory: Organization Failures Framework and Governance of Contractual Relations

Organization Failures Framework

Characterization of Transactions

Governance of Contractual Relations

Some Applications of Transaction Cost Theory

3.1.3.2 Agency Theory: Agency Costs, Adverse Selection, Moral Hazard and Hold-Up

Adverse Selection

Moral Hazard

Hold-up

Some Applications of Agency Theory

3.1.4 Conclusions and Some Comparisons

3.2 Market Microstructure and Intermediation Theory

3.2.1 Market Makers, Specialists, Dealers, Marketers and Merchants

3.2.1.1 Early Contributions on Financial Intermediation

3.2.1.2 Transaction Cost Reduction, Risk Bearing and Immediacy Service

3.2.1.3 The Impact of Asymmetric Information on Market Making

3.2.1.4 Liquidity Service and Consumption Smoothing

3.2.1.5 Market Making on Product Markets

3.2.2 Matchmakers and Brokers

3.2.2.1 Matchmaking on Financial Markets

3.2.2.2 Matchmaking on Other Markets

3.2.2.3 The Impact of Asymmetric Information on Matchmaking

3.2.2.4 Comparative Studies

3.2.3 Information Production, Guarantors and Certificators

3.2.3.1 Information Production and Signalling

3.2.3.2 Information Production and Selling

3.2.3.3 Certification and Guarantee Provision on Product Markets

3.2.4 Delegated Expertise, Bargaining, Contracting and/or Monitoring

3.2.4.1 Delegated Expertise

3.2.4.2 Delegated Bargaining

schrammfreightforwarderintermediaryrole2012.md 1.1 MB

История Исходник