Endogenous networks and international cooperation
| Published date | 01 May 2016 |
| Date | 01 May 2016 |
| Author | Max B Gallop |
| DOI | 10.1177/0022343316631033 |
| Subject Matter | Research Articles |
Endogenous networks and international
cooperation
Max B Gallop
Department of Government and Public Policy, University of Strathclyde
Abstract
The rise of social network analyses in the social sciences has allowed empirical work to better account for inter-
dependencies among actors and among their actions. However, this work has been, to a large extent, descriptive: it
has treated these actions as exogenous and immutable. In many cases these networks describe actions like alliance
formation or trade – phenomena that are the outcome variables for programs of social scientific research. In this
article, I attempt to account for both interdependencies and the endogenous nature of networks by incorporating
formal theory, helping answer the question of how these networks arise by looking at the incentives of actors to form
links with each other. I discuss the appropriate solution concept for a network formation game, and present an
algorithm for finding the equilibrium of these networks computationally as well as ways to compare the theoretical
networks to observed networks in order to evaluate the fit of the theory. I apply these methods to the study of
international cooperation – a subject where both the interdependencies and purposive nature of actors must be
accounted for. The theoretical network is able to reproduce a number of important observed characteristics. Still,
there are more factors that must be accounted for if we want to understand how the network of international
cooperation is formed.
Keywords
cooperation, formal modeling, networks
Introduction
In recent years the field of international relations, and
political science more generally, has increasingly
employed social network analysis. Such analyses allow
scholars to model and accountforinterdependencies
rather than assuming our variables are purely atomistic.
What has received less attention is the fact that many of
the networks we are interested in are, themselves, the
products of strategic processes, and these processes are
in fact the sorts of behavior that social scientists study. In
this article, I propose to apply formal and game theoretic
logics to the formation of networks, and in particular
networks of interstate cooperation.
I proceed by discussing some of the history of network
analysis, and its relationship to formal theory. I then
discuss the solution concepts used to determine a theo-
retically stable network and introduce an algorithm for
finding this equilibrium, as well as a simple theory of
interstate cooperation. Finally, I discuss the theoretical
equilibrium that is found when applying these tech-
niques and theories to the 50 richest countries in the
world and discuss how it compares to a similar empirical
network.
Network theory began with graph theory, which goes
back as far as Euler’s work on the bridges of Ko
¨nigsberg
in the early 18th century. In 1959, Erd}os & Re
´nyi intro-
duced random graphs, allowing the first probabilistic
interpretation of graphs. From this idea grew the notion
and discipline of network science. In network analysis,
the basic concern is to capture the interdependencies
among the elements, or nodes, through the study of the
linkages among them. Through the study of the struc-
ture of relationships, sociology, political science, and
economics have begun to use networks to make impor-
tant advances. Granovetter (1973) discovered the
Corresponding author:
max.gallop@duke.edu
Journal of Peace Research
2016, Vol. 53(3) 310–324
ªThe Author(s) 2016
Reprints and permission:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0022343316631033
jpr.sagepub.com
To continue reading
Request your trial