Yuleng Zeng

Date of Award

Spring 2020

Document Type

Open Access Dissertation


Political Science

First Advisor

Timothy Peterson


This dissertation studies whether and how economic interdependence promotes peace, focusing on three intertwined strategic questions originating from states’ incentives to hurt adversaries and disincentives to hurt themselves during international crisis bargaining. States that can exert substantial economic pains on potential rivalries also have strong incentives to bluff (i.e., to make claims on things they are not willing to fight for). It is therefore puzzling (1) whether bluffing makes conflict either more likely given there will be more incidents of crises or less likely given potential targets are more willing to concede; (2) once in conflict, whether states should cut losses in time or hold out longer in case the opponent has been bluffing and will quit soon; (3) as trade shifts states’ power and bargaining leverage unevenly, declining states can be incentivized to preempt (or, conversely, rising states can be incentivized to procrastinate), resulting in either more or fewer conflicts even when information asymmetry and concerns over bluffing are alleviated over time.

To address these puzzles, I use three game theoretical models to examine the respective strategic interaction and then apply statistical analysis to test their empirical implications. First, I tackle the theoretical foundation of commercial peace and reexamine the two competing theories in this field: costly signaling and opportunity costs. Utilizing a crisis bargaining model, I theorize and demonstrate that the bargaining environment allows the two mechanisms to operate simultaneously: the bluffing incentive does not increase the likelihood of war because bluffers exploiting information asymmetry are simultaneously restrained by the coercive channel. Building on this understanding of two parallel mechanisms at work, the second chapter investigates how coercion and information interact beyond the stage of conflict onset. Using a war of attrition model, I show that states can intentionally choose to endure economic losses to demonstrate resolve. Whether this strategy is attractive or not depends on how much they value the disputed good, i.e., issue salience. As such, the effects of economic ties on conflict termination are conditional. When the issue salience is high enough, the signaling mechanism kicks in and incentivizes states to hold out longer. Otherwise, states in conflict should opt to cut their losses in time as the coercive mechanism dominates. Finally, I investigate how bilateral trade can affect the shift of power over time. Using a stochastic game, I demonstrate a scope condition under which trade can stoke, rather than restrain, conflicts. Specifically, when states’ relative efficiency of translating trade gains into military power is at the extremes, they have smaller incentives to initiate conflicts. However, when their efficiency moves closer to parity (bounded by existing military balance), increasing bilateral trade can exacerbate commitment problems, leading to a higher likelihood of costly conflict.

To test the above theories, I use relevant statistical tools, including structural estimation, survival analysis, and network analysis, to examine interstate trade and conflict data and find supporting results. I conclude that while economic interdependence promotes peace in a one-shot crisis bargaining situation, the strategic interactions become more involved as we account for the impact of time. Under certain conditions, states can be incentivized to hold out longer in conflict despite strong economic ties. As trade shifts power balance over time, countries can also willingly endure short-term economic pains for the potential benefits of long-term settlements. These findings further the existing theories in trade-conflict studies. They also hold important implications for economic coercion, power competition, and contemporary policy issues including the recent China-U.S. trade war.


© 2020, Yuleng Zeng