Networks in economics [Elektronische Ressource] / by Jochen Möbert

Moebert@Inv41464

Sujets

• Mathematics

Informations

• Publié par: technischen_universitat_darmstadt
• Publié le: 01 janvier 2008
• Nombre de lectures: 20
• Langue: English
• Poids de l'ouvrage: 2 Mo

Extrait

Networks in Economics
by Jochen Möbert
Technische Universität Darmstadt - Fachbereich Rechts- und Wirtschafswissenschaften - Dissertation zur Erlangung des
akademischen Grades Doctor rerum politicarum (Dr. rer. pol.) vorgelegt von Diplom-Volkswirt Jochen Möbert (geb. in
Weinheim/Bergstraße) - Erstreferent und Betreuer: Prof. Dr. Horst Entorf - Koreferent: Prof. Dr. Carsten Helm
Tag der Einreichung: 17. April 2008 - Tag der mündlichen Prüfung: t1b7a. Juni 2008 - Darmstadt 2008, D17To Anika, Christine and SebastianTo My ParentsPreface
No large project can be accomplished without many supporters. Thank you to all
colleagues,Ph.D.studentsandstudentsatDarmstadtUniversityofTechnology,University
of Mannheim and Universitat Aut`onoma de Barcelona for discussing economic topics,
having lunch and proof reading my work.
I am especially grateful to my advisor at University of Frankfurt, Prof. Dr. Horst
Entorf, who supported me throughout the last years. Moreover, I am also obliged to
my co-supervisor Prof. Dr. Carsten Helm for comments and ameliorations.
Special thanks go to my family for their support during the last years. In particular, my
children have strongly facilitated this work by totally disregarding it.
Jochen M¨obert
Darmstadt, 17 April 2008Content
Chapter Title Page
1 Introduction 1
1.1 Scientific Motivation for Network Research 1
1.2 Dissertation Progress 2
2 Jefficiency vs. Efficiency in Social Network Models 5
2.1 Introduction 5
2.2 Notation 6
2.3 Efficiency vs. Jefficiency 9
2.4 Jefficiency in the Symmetric Connection Model 17
2.5 Conclusion 20
2.6 Appendix 22
3 Power and Ownership Structures among German Companies 26
3.1 Introduction 26
3.2 Company Network Literature 29
3.3 The Data Base 31
3.3.1 Descriptive Network Statistics 35
3.3.2 MAN-Analysis 38
3.3.3 Sub-Networks 42
3.4 Important Companies 44
3.4.1 Central Nodes in the Global Network 44
3.4.2 Analysing the Centrality Concept 48
3.5 Conclusion 57
3.6 Appendix 59Chapter Title Page
4 DAX-Executive Remuneration and the
Supervisory Board Network of Executives 83
4.1 Introduction 83
4.2 Executive Remuneration Literature 84
4.3 The Data Base 86
4.4 Results 91
4.5 Conclusion 96
4.6 Appendix 97
5 The Coauthor Network of the Deterrence Puzzle 99
5.1 Introduction 99
5.2 Literature on Coauthor Networks 101
5.3 Hypotheses 105
5.4 Descriptive Analysis of the Meta Data Base 107
5.5 Inductive Analysis 123
5.6 Conclusion 129
5.7 Appendix 130
6 Supplement 147
6.1 Glossary 147
6.2 MAN Classification Scheme 150
References 153Chapter 1
Introduction
1.1 Scientific Motivation for Network Research
The topic of this dissertation “Networks in Economics” is special. Classical economic themes
in microeconomic as well as macroeconomics are either concerned with individuals interacting
through anonymous markets or with the results of the interaction. In recent decades as well as
centuries, given the lack of data and computing facilities this was a very rational and successful
approach. Often these classical models are implicitly solved by the introduction of an abstract
concept. The Walrasian auctioneer who clears the market by matching supply and demand is
certainly the most popular example. An abstract concept allows to focus on final outcomes such
as prices and quantities but turn a blind eye on social processes. Hence, we can regard these
models as benchmark cases. In some circumstances the benchmark describes actual outcomes
fairly well whereas in others more information is necessary to get a deeper understanding of
observed processes and outcomes.
Economists are aware of these shortcomings and have considered much more elaborate models
in recent years. For instance, the microfoundation of macroeconomics, agent-based models,
models taking into account the heterogeneity of agents, social interaction models, among others
broaden our view and may enhance our understanding. All modern researchers are always eager
to improve standard models through the incorporation of additional pieces of information. The
question comes up, when the process of more detailed models will come to a stop? Implicitly,
this question asks “What is at the centre of all economic research?” For many economists a
natural answer is that the atomistic subject of economics is either an individual, an institution
11or the interplay between both. Economic network analysis exactly investigates the relationship
of these atomistic subjects to each other and the importance of these relationships for economic
outcomes. Henceforth,economicnetworkanalysismaynotbeontheagendaoftoday’seconomists
but might pave the way for a new way of thinking among economists.
1.2 Dissertation Progress
The interest in economic network analysis arose while I was writing my master’s thesis
“Development of Bubbles: Theory and a case study of the ‘Neuer Markt’ ”. Information on
future stock market developments often contains no information which is based on any form of
economic reasoning. Unfounded speculation can be self-perpetuating if naive investors or other
financial market participants react on each other, i.e. take long positions if a peer group of each
investorpurchasesshares. Thisprocessescanbeverydynamicand, therefore, noreliablemodels
forecasting the burst of stock market bubbles exist.
The idea arises that a successful forecast model might catch the social networks of important
financial market participants. Hence, during the first years of working on my dissertation, I
plannedtorevealandinvestigatethesocialnetworksoffinancialanalysts. Intheaftermathofthe
burst of the bubble the disappointed small investors often adjudged them guilty. Beyond doubt,
their forecasts were overconfident, however, beforehand bearish financial analysts performed
poorly when the stock market indices continuously hit new all-time-highs. They lost their jobs
and overconfident analysts were encouraged to continue their misinterpretations of the market
situation.
2Attheoutsetofmydissertation,Icontactedthe‘SocietyofInvestmentProfessionalsinGermany’
an organisation representing “more than 1,100 individual members representing over 400
3investment firms, banks, asset managers, consultants and counselling businesses” and proposed
to survey the behaviour and social networks of its members. In a first step, the research idea
was presented to the managing director who supported my research scheme. Unfortunately, the
DVFA-board of directors turned down the project. I continued this line of research by directly
contacting banks. However, the analysts always denied to answer questions on the structure of
their social networks. Consequently, the project was closed.
1In very recent years, economists have also started to explore the influence of neuroscientific processes on
decision-making. These researchers might probably define the atomistic subject of economics quite differently.
2‘Deutsche Vereinigung fu¨r Finanzanalyse und Asset Management’ (DVFA) located in Frankfurt am Main.
3http : //www.dvfa.de/home engl/dok/35307.php downloaded on May 25, 2007.
2Subsequently,theideaarosetosurveygroupsforwhichtheinformationgatheringislessdifficult.
Iplannedtoevaluatesocialnetworksinschoolclassesandtheirinfluenceonmarksortestscores.
Managing such a project needs a large amount of resources and administrative work. Pupils
as well as parents, teachers, and administrative staff have to be contacted to organize such a
project. Accordingly,thisapproachseemednotfeasibleduetothelargeworkload. Toreducethe
burden of work I was looking for existing data sets containing social network information. After
reading several journ