Behavioural Decision Theory and the Gains Debate in International Politics
| Published date | 01 June 2002 |
| Author | Jeffery Berejikian,Matthew Mulford |
| Date | 01 June 2002 |
| DOI | 10.1111/1467-9248.00367 |
| Subject Matter | Original Article |
Behavioural Decision Theory and the
Gains Debate in International Politics
Matthew Mulford
London School of Economics
Jeffery Berejikian
University of Georgia
The temptation to free ride and the fear of exploitation are traditionally offered as factors explain-
ing individuals’ choices under conditions of social dilemma. In two experimental studies of pris-
oner’s dilemma behaviour we demonstrate that the ratio-differences of payoffs are as important as
the absolute differences in affecting the likelihood of cooperation. The implications of this addi-
tional structural determinant of dilemma choice are developed and applied to the stalled gains
debate in international relations theory. By incorporating ratio-differences into the analysis ‘realist’
and ‘liberal’ theories of state behaviour may be reconciled.
In the 1997 Presidential Address to the American Political Science Association,
Elinor Ostrom called for the creation of a ‘behavioural theory of collective action’
(1998). According to Ostrom, such theories should be, ‘based on models of the
individual consistent with empirical evidence about how individuals make deci-
sions in social–dilemma situations’ (Ostrom 1998, p. 1). Rather than develop theory
based upon researchers’ perceptions of the salient issues for decision makers, we
agree with Ostrom that the actual behaviour of social actors is the best place to
start. We offer here the results from two decision-making experiments that con-
tribute to a behavioural theory of collective action and also show how the find-
ings can contribute to contemporary debates in international relations theory.
These connections are, by necessity, somewhat speculative. Yet, theoretical induc-
tion based on empirical observation is a step forward from theory building based
on unobserved, untested and descriptively dubious psychological claims.
International Relations and the Prisoner’s Dilemma
Modelling state strategic interactions with the prisoner’s dilemma was one of first
applications of game theory in the study of international relations (e.g., Snyder,
1971; Stein, 1982; Oye, 1985). Scholars continue to find the game relevant to
understanding state choice for issues as diverse as security, trade, and environ-
mental policy. The strengths of the approach lie in an explicit emphasis on anar-
chic interactions, deductive theorising, and the ability to unify the study of conflict
and cooperation.1 Specifically, the ‘dilemma’ pictures each actor having a domi-
nant incentive to defect while universal defection produces socially sub-optimal
results. One obvious solution to the dilemma is the enforcement of contractual
© Political Studies Association, 2002.
Published by Blackwell Publishers, 108 Cowley Road, Oxford OX4 1JF, UK and 350 Main Street, Malden, MA 02148, USA
210
M AT T H E W M U L F O R D , J E F F E RY B E R E J I K I A N
agreements by some external power. Businesses, for example, faced with dilemma
choices are able to make non-simultaneous exchanges because of binding contracts
enforced by legal statutes. Because the international system is anarchic, it lacks just
such enforcement mechanisms to guarantee state cooperation. Two central ques-
tions for international relations scholars are if states, absent enforcement, will co-
operate in situations with the structure of a dilemma? And if so, when? Two general
approaches to the problem exist.
Liberalism, Realism and the Gains Debate
For the sake of simplicity and clarity we will classify international relations theo-
rists into two broad categories: liberals and the realists.2 Liberal scholars are gen-
erally more optimistic than realists about the prospects for meaningful cooperation
between states under conditions of social dilemma. This difference in outlook is
primarily due to a difference in the two approaches’ assumptions about states’
utility functions. Liberals see states’ utility functions as traditional rational choice
theories depict individuals’. That is, states are defined as rational egoists and are con-
sequently modelled as absolute gains maximisers, pursing wealth and power while
remaining largely indifferent to the circumstances of other states. Liberals argue
that because states are generally interested only in their own welfare, it is pos-
sible (if sometimes difficult) for states to carve out stable cooperative relationships
absent enforcement mechanisms. The key to promoting uninforced international
cooperation is, say the liberals, the construction of properly crafted regulatory
regimes.
Regimes can support cooperation in many ways.3 First, they codify rules so that
states have shared expectations about what sorts of behaviour will be considered
non-cooperative. If the rules of the game are completely agreed upon and clear,
then the probability of accidental or unintended defection is diminished. Second,
regimes may restructure the interaction by extending the game out into the future,
creating an iterated game and thus making conditional cooperation strategies part
of stable equilibria. Third, they make others’ choices transparent. This last func-
tion is particularly important because it reduces the fear that cooperating states will
be secretly exploited by defecting states. Information about others’ behaviour is
best collected within the context of a formal institution that clearly defines cheat-
ing and observes compliance (Stein, 1983).
Well-crafted international institutions thus mollify the pernicious characteristics of
anarchic systems by reducing transaction costs, extending the number of games
out into the future, and providing accurate information about others’ behaviour.
With the help of institutions, states can offer conditionally cooperative strategies
further supporting robust patterns of mutual cooperation.4 Where no such in-
stitutions exist, or where they are poorly crafted, cooperation is more likely to
collapse as states fall prey to the incentives to defect.5 The blame for failed inter-
national cooperation is therefore often laid at the doorstep of faulty or anaemic
institutions that are unable to adequately co-ordinate behaviour under anarchy.
Design the institutions correctly and you create the conditions for international
cooperation.
B E H AV I O U R A L D E C I S I O N T H E O RY A N D T H E G A I N S D E B AT E
211
Realists, in contrast, are pessimistic about the prospects for international coopera-
tion generally, and even more so under conditions of social dilemma. In place of
the liberal concern over the lack of co-ordinating mechanisms, anarchy’s chief con-
sequence for realists is that it requires states to provide for their own security. This
forces states to perpetually fear for their survival, resulting in a primary concern
about their relative strength vis-à-vis other states. For realists, the fundamental goal
of states is to prevent others from achieving relative advantage (Grieco, 1988).6
Regardless of the viability of international regimes (as defined in liberal terms) real-
ists predict that states will usually defect in a prisoner’s dilemma. Unilateral defec-
tion can produce both absolute and relative gains for the defecting state while at
the same time shielding it from relative decline resulting from the predatory prac-
tices of rivals. That is, ‘Realist theory argues that international anarchy inhibits the
willingness of states to work together even when they have common interests, and that
international institutions are largely unable to mitigate the constraining effects of
anarchy on international cooperation’ (Grieco, 1990, p. 27, italics added).
Throughout much of the 1990s the debate between the competing perspectives of
liberalism and realism dominated much of the research on international coopera-
tion. This debate revolved around the absolute vs. relative gains assumptions
concerning state utility functions. This polarisation into competing camps produced
theories that would have otherwise contributed more to our understanding of
interstate politics (Powell, 1994, p. 313). Each side in the debate independently
developed its own logic so that in the end realism and liberalism appeared theo-
retically irreconcilable.7 An empirical record that failed to provide a convincing case
for either side exacerbated the theoretical impasse.8 Attempts have been made to
unify the competing schools. But, to date, no particular reconciliation has become
generally accepted. In fact, the disagreement is now simply accepted as part of the
contested theoretical terrain (Waltz, 1998).9
The enduring legacy of the gains debate is that we have two very different views
describing the hurdles to cooperation between states, and these opposing positions
are each confronted by a world in which states appear to exhibit behaviour, in dif-
ferent situations, consistent with both.10 One reasonable conclusion is that both
realism and liberalism may be right, but that some crucial contextual variable is
missing from current models. In this light, we offer an addition to realist and liberal
explanations of cooperation that, rather than relying upon scholarly deductions
about the implications of anarchy, is grounded in systematic observation of
dilemma decision making. To anticipate, the results from our decision-making
experiments suggest that a more accurate model of how decision makers perceive
the relative size of potential gains and losses in social dilemmas can help to resolve
the gains debate.
...
To continue reading
Request your trial