Dynamic Approaches to Studying Parliamentary Coalitions

Abstract

Democratic governments come in two pure forms-presidential and parliamentary systems. main distinction lies in the relationship between the executive and the legislature; a separation of powers characterizes a presidential system, while in a parliamentary system, there is a fusion of the executive and the legislature. majority of democracies in the world use some variation of a parliamentary system (Lijphart 1999, 118). In these nations, where multiple parties compete for power, the electorate determines the membership of the legislature, which in turn chooses the executive (composed of the prime minister and the cabinet). legislature also typically maintains the power to dismantle the government via a vote of no confidence. executive thus must garner the support of enough of the legislators so that the legislature does not vote (usually by a majority) to decompose the cabinet in favor of another government. In most cases, this requires a coalition between various parties that control, or at least are tolerated by, a majority of the legislators. In many ways, research on coalition behavior in parliamentary democracies exemplifies progress: generating path-breaking theoretical models (e.g., Riker 1962; Baron and Ferejohn 1989), documenting near-perfect predictive relationships (e.g., Browne and Franklin 1973; Warwick and Druckman 2006), and melding theoretical and empirical work (e.g., Diermeier and Stevenson 2000; Martin and Stevenson 2001; Skjaeveland, Serritzlew, and BlomHansen 2007). Yet, a well recognized but largely unresolved problem plagues the bulk of extant studies: they are static. typical research study explores a single coalition process at one point in time, with limited attention to dynamics external to coalition politics. alternative is to have dynamic studies that explore interactive processes (e.g., with feedback) over time. Even the earliest scholars of coalition politics recognized the importance of developing dynamic studies (e.g., Leiserson 1970, 271; Bueno de Mesquita 1975; Browne and Dreijmanis 1982, 340; Laver 1986, 33-34; Laver and Hunt 1992, 74-75). Laver (1998, 22) explains, The absence of these [dynamic] features from government-formation models is not because theorists regard them as unimportant. reason is more prosaic-it is very difficult to incorporate them in a rigorous manner. Incorporating dynamic elements into coalition theory would represent substantial progress. Indeed, the presumed goal of the research is to understand which coalition forms, how long it lasts, and how it shapes governing decisions and policy. It seems fairly obvious that processes of coalition formation, governance, and duration relate to one another and also to other political and economic dynamics. Understanding how coalition politics works and isolating causal mechanisms then require a consideration of these dynamic relationships-something that is still lacking. This symposium brings together three of the most promising research programs aimed at incorporating dynamic elements into coalition models. articles, each of which constitutes one example from the authors' larger research programs, include a study by Daniel Diermeier, Roderick Swaab, Victoria Medvec, and Mary Kern that explores the relationship between coalition formation and future bargaining expectations; an article by Lanny Martin and Georg Vanberg that looks at the methods governing parties use to communicate with voters who will later decide the fates of these parties; and an article by William Bernhard and David Leblang that investigates the endogenous relationship between cabinet stability and the economy. Each article not only offers critical insights into the particulars of coalition politics but also provides methodological guidance on how one can go about creating dynamic models of coalitions (e.g., the articles use experiments, content analyses, and instrumental variable models). symposium concludes with commentaries from Michael Laver and Kaare Strom. …