Bargaining with a biased autocrat
| Author | John Slinkman,Colin Krainin |
| Published date | 01 April 2017 |
| Date | 01 April 2017 |
| DOI | 10.1177/0951629816630438 |
| Subject Matter | Articles |
Article
Bargaining with a biased
autocrat
Journal of Theoretical Politics
2017, Vol. 29(2) 273–298
©The Author(s) 2017
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI:10.1177/0951629816630438
journals.sagepub.com/home/jtp
Colin Krainin
Princeton University, USA
John Slinkman
Novetta, USA
Abstract
This paper presents a specific dynamic extension to the model outlined by Jackson and Morelli
(Political bias and war. The American Economic Review 97(4): 1353–1373). In particular, we focus on
the implications of a long-lived leader who possesses bias towards or against war that potentially
differs from the country (s)he represents. Bargaining and war outcomes are characterized in this
environment. We emphasize three results: (1) war sometimes occurs only after substantial delay;
(2) if it is possible to select a leader, a country will always select a leader who is biased towards
war; (3) when the distribution of power is uneven, a country might select an aggressive leader
who eventually leads the country into war. If applied in the context of a civil war model, the
third result implies that particularly disadvantaged rebel groups might select aggressive leaders to
bargain with the central government and, potentially, incite conflict.
Keywords
Bargaining; conflict; leaders; war
1. Introduction
In 1933, Adolf Hitler came to power in an enfeebled Germany. Over the next few years,
his aggressive diplomacy resulted in significant bargaining gains for Germany in the
international arena, culminating in the Munich Agreement of 1938. Yet,by the next year,
Hitler’s tactics pushed his country into a conflict that eventually escalated into WorldWar
II and ended in almost total disaster for Germany.
Corresponding author:
Colin Krainin, 024 Corwin Hall, Princeton, NJ 08544-1012, USA.
Email: ckrainin@princeton.edu
274 Journal of Theoretical Politics 29(2)
That aggressive autocratic leaders such as Hitler can instigate war is intuitively obvi-
ous. Yet, barring uncertainty over the degree of aggressiveness, why are such leaders
often allowed to renegotiate the international bargain in their favor instead of being
attacked at their weakest point? Second, how is it that the host country tolerates, or even
elects, such a leader?
To investigate these questions, we build a focused, dynamic extension of the model
in Jackson and Morelli (2007) in which one country in a two-country bargaining game is
endowed with or selects a potentially biased leader.1As in Jackson and Morelli (2007),
bias determines how a permanent leader values war versus peaceful outcomes. Unlike
Jackson and Morelli (2007), war may result evenwhen only one country chooses a leader
with bias.2Using this model, we then characterize three results that address the empirical
pattern discussed above: (1) war sometimes occurs only after substantial delay; (2) if it
is possible to select a leader, a country will always select a leader who is biased towards
war; (3) when the distribution of power is uneven, a country might select an aggressive
leader who eventually leads the country into war. If applied in the context of a civil war
model, the third result implies that particularly disadvantaged rebel groups might select
aggressive leaders to bargain with the central government and, potentially, incite conflict.
Intuitively, our model can feature a period of delay before war because leaders make
demands relative to their countries’ strengths. If a country’s leader is biased enough
towards conflict, the leader will demand a relatively large amount of resources. This
causes the country to accrue more and more resources over time. This, in turn, increases
the demands of the leader, who now values war even more. These demands may eventu-
ally become so large that conflict is inevitable. That the other country in the model does
not preempt this chain of demands with immediate war is a somewhat subtle result, but
it hinges on the willingness of that country to exchange a longer period of consumption
without war for worse prospects in an inevitablefuture war. While altering the model may
reduce the starkness of this ‘no preemption’ result, the basic logic of exchanging current
peaceful consumption in exchange for a worse future war applies quite generally.3
If given the option, a country’s population always chooses a leader biased towards war
because this type of leader can capture more of the bargaining surplus through biased in-
transigence. Strikingly,it is easy to compute examples in which a country selects a leader
who is so biased that war eventually occurs. This type of result holds when countries are
initially quite weak and are willing to face the eventual inefficiency of war in exchange
for higher current consumption. Therefore, our model predicts that weak countries are
particularly susceptible to selecting leaders whose long-run interests prove to be counter
to their own.
Previous explanations for delays in war with warlike leaders have focused on incom-
plete information concerning the nature of the leader in question. However,Powell (2006)
argues that the reason that Britain and France came to declare war on Germany in
1939 was not because they were uncertain as to Hitler’s true intentions. Actually, these
countries became more resolved towards war as these intentions were revealed. Hence,
appeasement was at least partially the result of incomplete information, whereas actual
war requires a complete information explanation. We do not include incomplete infor-
mation in our model, so as to better demonstrate the dynamic effects of biased leaders in
isolation from complicating factors. Nonetheless, we view our results as complementary
to incomplete-information arguments and believe that an extension of our model to this
setting would be valuable.
To continue reading
Request your trial