Democracy, Interdependence, and the Sources of the Liberal Peace

Published date01 January 2008
Date01 January 2008
DOI10.1177/0022343307084921
AuthorJoseph M. Grieco,Christopher F. Gelpi
Subject MatterArticles
17
Introduction
Can the world be made more peaceful
through commerce? For more than a century,
and particularly during the past decade, this
question has constituted an important focal
point of scholarly debate in the field of inter-
national relations.1We suggest that, by virtue
of the domestic political institutions in
which their leaders operate, democratic states
react to greater trade integration with a
reduced propensity to initiate militarized
disputes with their partners. On the other
hand, while some autocratic leaders might
react to growing trade integration with a
determination to avoid conf lict, such leaders
do not face institutional constraints to do so.
Thus, for autocratic states, we expect no sys-
tematic relationship between their level of
trade integration with commercial partners
and their propensity to initiate military con-
f licts against those partners. We find robust
support for these expectations in our analysis
of trade integration and international con-
flict initiation from 1950 to 1992.
Our substantive interest is to determine if
regime type influences the pacifying effects
© 2008 Journal of Peace Research,
vol. 45, no. 1, 2008, pp. 17–36
Sage Publications (Los Angeles, London, New Delhi
and Singapore) http://jpr.sagepub.com
DOI 10.1177/0022343307084921
Democracy, Interdependence, and the Sources
of the Liberal Peace*
CHRISTOPHER F. GELPI & JOSEPH M. GRIECO
Department of Political Science, Duke University
Can the world be made more peaceful through commerce? Empirical studies of the impact of trade on
military conf lict have yielded conflicting results depending on the specif ic measures and empirical
domains that scholars select for their studies. The article suggests that these varying results may be due
to inadequate specification of the conditions under which trade will prevent conflict. In particular, pre-
vious research suggests that democratic leaders rely on public policy successes, such as economic growth,
to maintain their political viability to a greater degree than do autocratic leaders. Since trade can help
promote growth, the authors argue that democratic leaders should be more averse than autocratic leaders
to initiating military conf licts with trading partners, for such conflicts might damage commercial ties
and hamper politically important economic growth. The authors find support for this expectation in
their analysis of trade integration and international conf lict initiation by democratic and autocratic
states between 1950 and 1992. The results are robust across different data sources on trade and con-
f lict, suggesting that the conditional impact of trade on conflict may explain the variance in previous
results. However, these results have sobering implications for the view that trade dependence by itself
can be a mechanism for preventing autocratic states from using military force.
*The dataset used in this ar ticle can be found at
http://www.prio.no/jpr/datasets and http://www.duke.
edu/~gelpi. The authors would like to thank Bruce Russett,
John Oneal, Katherine Barbieri, Robert Keohane, Peter
Feaver, Brian Pollins, Edward Mansfield, Håvard Hegre,
and two anonymous reviewers for their comments on the
manuscript. Any remaining errors are, of course, the
responsibility of the authors. Comments or questions can
be directed to Christopher Gelpi at gelpi@duke.edu and
Joseph Grieco at grieco@duke.edu.
1On the development of this literature, see Mansfield &
Pollins (2003).
84921_JPR_17-36.qxd 12/14/2007 2:46 PM Page 17
of economic interdependence. However, we
also address two methodological problems
regarding the literature on trade and conflict.
First, authors often present directional prop-
ositions about trade dependence and military
conflict, but typically test those arguments
with non-directional data. That is, authors
often argue that State A’s dependence on
State B will prevent A from using force
against B, but only test whether pairs of
states that trade with one another are more
likely to engage in conflict. Such designs do
not pay sufficient attention to which state in
the dyad is the more dependent on trade and
which state did or did not initiate a military
dispute in a given year. Second, much of the
quantitative research on trade and conflict
has been plagued by the substantial amount
of missing trade data. We utilize new methods
for coping with this problem.
Our discussion proceeds in two main
stages. First, we present the analytical ground-
ing for our hypothesis that the relationship
between economic interdependence and the
initiation of military conflict is contingent
upon the domestic political regime of the
country in question. Second, we develop and
execute a statistical test of this argument.
The Interaction of Trade,
Democracy, and Conflict
We proceed with two assumptions. First, mili-
tary conflict is typically the result of deliber-
ate decisions made by national leaders. Second,
national leaders wish to retain office.
Given these assumptions, we ask, how will
conflict-induced disruptions of trade inf lu-
ence the ability of democratic and autocratic
leaders to retain office? In developing our
response, we build on the work by Olson
(1993, 2000) and by Bueno de Mesquita et al.
(1999a,b, 2003). The latter assume that leaders
wish to retain office and that countries differ
in terms of the range of citizens who select
leaders (the ‘selectorate’) and the minimum
size of coalitions whose support a leader must
gain in order to win and retain office (the
‘minimum winning coalition’), or what Olson
(1993, 2000) called encompassing interests.
Leaders, Bueno de Mesquita et al. (1999a)
suggest, try to stay in office either by provid-
ing ‘private goods’ exclusively to coalition
members, or by generating policy successes
that are enjoyed by the entire citizenry. They
argue that the relatively wide breadth of
selectorates and winning coalitions in de-
mocracies makes it difficult for democratic
leaders to retain office through payments of
private goods. Thus, democratic leaders gen-
erally need to provide broader policy suc-
cesses in order to remain politically viable.
Bueno de Mesquita et al. suggest that one
such type of policy success is economic
growth (2003: 101–102, 149–161), and, in
support of that view, they present evidence
that the tenure of democratic leaders is more
sensitive to economic growth than that of
autocratic leaders (2003: 306–308).
As a result, we expect trade to be of greater
interest to democratic than to autocratic
leaders. Although not without controversy, a
key expectation in the field of economics is
that trade generally promotes the growth of
per capita national income.2By consequence,
we may expect democratic leaders to be more
concerned than autocratic leaders about the
deleterious economic effects of a breakdown
of foreign trade as the result of a militarized
dispute, for it is in democracies where the
delivery of economic performance is more
uniformly important to national leaders.3
journal of PEACE RESEARCH volume 45 / number 1 / january 2008
18
2See, for example, Krueger (1997), Edwards (1993, 1998),
and Lawrence & Weinstein (1999). For an important cri-
tique of such arguments, see Rodrik (1999).
3Papayoanou (1996) puts forward this argument for the
onset of World War I in Western Europe. Our analysis
suggests that democracies resolve disputes among them-
selves out of a mutual fear of losing the benefits of inte-
gration; by constrast, Morrow (1999), Li & Boehmer
(2001), and Crescenzi (2003) suggest that relatively more
dependent democratic states, insofar as their leaders sustain
large political losses if trade or financial ties with other
countries are broken, send a stronger signal of resolve in
diplomatic disputes and, thus, can bully both authoritarian
and democratic partners. Future analysis should adjudicate
between these two different mechanisms.
84921_JPR_17-36.qxd 12/14/2007 2:46 PM Page 18

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT