Short versus Long Coalitions: Electoral Accountability and the Size of the Public Sector


  • We are indebted to participants of various conferences and seminars for helpful feedback on earlier versions this article, including political economy panels at the American Political Science Association meetings, the Harvard Research Group on Political Institutions and Economic Policy (cosponsored by the Weatherhead Center for International Affairs and CBRSS), the Yale Workshop on Electoral Microfoundations of Economic Policy Making, and seminars at Rice University, the University of Colorado, the University of California at Davis, and Washington University. Gary Cox, Rob Franseze, Jeff Frieden, Justin Fox, Torben Iversen, Jeff Lewis, John Huber, Ken Kollman, Mariano Tomassi, and Alan Wiseman offered helpful comments on earlier drafts. Frances Rosenbluth would like to thank the Yale Provost Office and the Yale Leitner Program in International and Comparative Political Economy for funding. We gratefully acknowledge the able research assistance of Abbie Erler, Mathias Hounpke, Amelia Hoover, and Tom Pepinsky. Jamie Druckman and Paul Warwick generously provided government coalition data; Lanny Martin and Randy Stevenson, data on ideology; and Torben Iversen and Tasos Kalandrakis, data for government expenditure and economic conditions.

Kathleen Bawn is associate professor of political science, University of California, Los Angeles, Los Angeles CA 90405 ( Frances Rosenbluth is professor of political science, Yale University, New Haven, CT 06520-8301 (


This article examines the policy consequences of the number of parties in government. We argue that parties externalize costs not borne by their support groups. Larger parties thus internalize more costs than small parties because they represent more groups. This argument implies that the public sector should be larger the more parties there are in the government coalition. We test this prediction using yearly time-series cross-sectional data from 1970 to 1998 in 17 European countries. We find that increasing the number of parties in government increases the fraction of GDP accounted for by government spending by close to half a percentage point, or more than one billion current dollars in the typical year. We find little support for the alternative claim that the number of legislative parties affects the size of the public sector, except via the number of parties in government.