Published online by Cambridge University Press: 03 August 2015
Scholars argue that states undertake foreign military interventions for economic reasons, yet few have investigated whether intervention produces economic benefits. This article answers this question in the context of US foreign-imposed regime changes (FIRCs) in Latin America. Because FIRCs install leaders who are sympathetic to the intervener’s interests, economic arguments maintain that these interventions should increase bilateral trade between the targets and imposing countries. Yet security-based arguments assert that FIRCs should have little economic effect, as regime changes target threats rather than generate economic benefits. A third perspective argues that FIRCs reduce trade by generating political instability, which causes foreign firms to cut back on their involvement and domestic firms to experience difficulty getting goods to market. To test these competing arguments, this study employs a novel dataset on bilateral trade (1873–2007) compiled through archival research in Washington, DC. Using a gravity model and synthetic controls, it finds that FIRC produces an average decrease of 45 per cent in the dollar value of bilateral trade. Further analysis of archival sector-level data and case studies cast doubt on alternate explanations.
Department of Political Science, University of California, San Diego (email: pzachary@ucsd.edu); Department of Public Policy, University at Albany, SUNY (email: kdeloughery@albany.edu); Department of Political Science, The George Washington University (email: downes@gwu.edu). Previous versions of this article were presented at the 2013 meeting of the International Studies Association and seminars at UC San Diego and George Washington University. The authors thank Lawrence Broz, Chris Fariss, Mike Findley, James Fowler, Julia Gray, Emilie Hafner-Burton, Kyle Joyce, Miles Kahler, Stephen Kaplan, David Lake, Brad LeVeck, William Reed, Molly Roberts, Phil Roeder, Rachel Stein and Caitlin Talmadge and the anonymous reviewers for helpful comments and suggestions. We also thank the librarians at the Department of Commerce for their invaluable guidance as well as Corey Hill, Michael Joseph, Julia Macdonald and Michael Weaver for excellent research assistance. Zachary acknowledges support from the National Science Foundation and the San Diego Supercomputer Center. All remaining errors are the authors’ own. Data replication sets are available at http://dataverse.harvard.edu/dataverse/BJPolS, and online appendices are available at http://dx.doi.org/doi:10.1017/S0007123415000332.