American Political Science Review

Research Article

The Economic Origins of Democracy Reconsidered


a1 University of Minnesota

a2 Georgetown University


The effects of inequality and financial globalization on democratization are central issues in political science. The relationships among economic inequality, capital mobility, and democracy differ in the late twentieth century for financially integrated autocracies vs. closed autocracies. Financial integration enables native elites to create diversified international asset portfolios. Asset diversification decreases both elite stakes in and collective action capacity for opposing democracy. Financial integration also changes the character of capital assets—including land—by altering the uses of capital assets and the nationality of owners. It follows that financially integrated autocracies, especially those with high levels of inequality, are more likely to democratize than unequal financially closed autocracies. We test our argument for a panel of countries in the post–World War II period. We find a quadratic hump relationship between inequality and democracy for financially closed autocracies, but an upward sloping relationship between inequality and democratization for financially integrated autocracies.


c1 John R. Freeman is Johnston Professor, Department of Political Science, University of Minnesota, 1414 Social Science Building, 267 - 19th Avenue South, Minneapolis, MN 55455 (

c2 Dennis P. Quinn is Professor, McDonough School of Business, Georgetown University, Rafik B. Hariri Building, 37 and O Streets, NW, Washington, DC 20057 (


The first version of this article was presented at the 2008 Meeting of the American Political Science Association. We thank Georgetown audiences at the McDonough School of Business and the Mortara Tuesday Political Economy seminars for their comments. We also thank seminar participants at the International Political Economy Society 2008 conference, ETH Zurich, the International Monetary Fund (IMF), Yale University, and the Universities of Iowa, Virginia, and Zurich. We thank Aart Kraay and James Galbraith for discussion about the inequality data sets, Keith Ord and Michael Tomz for advice on the research design, and Pietra Rivoli for comments on our theoretical argument. Robert Bates, Jeff Frieden, Mark Kayser, Irfan Nooruddin, Thomas Sattler, Ken Scheve, and Vineeta Yadav also offered valuable comments. We thank APSR co-editor Ron Rogowski for his exceptionally insightful comments. For research assistance we thank Rebecca Anderson, Naphat Kissamrej, Andrew Lucius, Dafina Nikolova, and Erica Owen. We thank Annalisa Quinn for her editorial work. Of course, we alone are responsible for the contents of the article. Dennis Quinn gratefully acknowledges funding support from the National Science Foundation, the Research and Fiscal Affairs departments at the IMF, and the McDonough School of Business at Georgetown. The replication data will be posted on publication and will be available from either author.