Strategic models of coalition bargaining formation have demonstrated the importance of institutional features for an understanding of cabinet formation in West European democracies. Yet little is know about the empirical regularities of government formation processes. In this article we analyse the duration of formation processes using a semi-parametric estimation procedure on a dataset of 304 government formations in thirteen multi-party democracies. The results are consistent with a bargaining approach under incomplete information.A naive model of coalition bargaining would suggest that one of the actors who participates in the negotiations, say a party leader, makes a proposal, then another negotiator accepts the proposal, rejects it or makes a counterproposal, which then is accepted, rejected, etc. until an agreement is reached. The formation time until a new government is installed then reflects this back and forth between offers and counter-offers. This view, however, is incompatible with rational bargaining, as recent game-theoretic models have indicatedprovided negotiations are conducted under complete information. 1 In equilibrium the first offer is made immediately (there is no delay) and is always accepted (there is no rejection). 'Bargaining' in the ordinary sense of the word as the back and forth of making and rejecting offers never occurs. 2

In this article we will review the literature on cabinet durability and cabinet termination. 1 The fact that many cabinets in Western multi-party democracies do not serve out their full potential legal term in office has given rise to an important and growing body of research in political science. Cabinet durability is one of the three main features of cabinets, the others being cabinet party composition and allocation of portfolios. 2 Each is of theoretical interest in itself, , respectively. This research is sponsored by the Netherlands Organization for Scientific Research (NWO) under grant PG 50-370. It was begun while Van Roozendaal was a visiting researcher at the School of Social Sciences, University of California, Irvine. We are indebted to the staff of the Word Processing Center, School of Social Sciences, UCI, for technical assistance and to Dorothy Green and Chau Tran for bibliographic help. The authors' names are in alphabetical order. 1 Our initial aim was to update the excellent review of cabinet durability done over a decade ago by Lawrence C. Dodd, 'The Study of Cabinet Durability: Introduction and Commentary ', Comparative Political Studies, 17 (1984), 155-62, which became badly out of date because of the considerable volume of recent empirical and theoretical work. Subsequent to the initial submission of this article, we discovered that Paul Warwick, Government Survival in Parliamentary Democracies (New York: Cambridge University Press, 1994), had provided a substantial review of this literature in the course of elaborating his own model of government survival. We have incorporated references to that work here. However, while we agree with much of what Warwick has written, our perspective on the extent of progress in understanding the determinants of cabinet termination and on what remains to be done and how best to proceed is rather different from his. 2 There is a voluminous literature attempting to predict the party composition of cabinets from the party breakdown in the legislature and other factors, such as the ideological propinquity of the various parties. An excellent recent review is Michael Laver and Norman Schofield, Multiparty Government: The Politics of Coalition in Western Europe (Oxford: Oxford University Press, 1990), which also contains important new theoretical contributions of their own. The literature on portfolio allocations had been relatively dormant since the work of Eric Browne and his colleagues, e.g., Eric C. Browne and Mark Franklin, '

No abstract

Abstract. The centre is an interesting concept for formal models of, for instance, political coalition formation, but also for European party systems. In this article, the focus will be on centre parties and their impact on coalition cabinet formation. Using game theory, we describe one concept of the centre, the central player, and we introduce a concept of a strong centre, the dominant central player. With these concepts of the centre, we will predict what kind of coalitions will be formed. These prediction principles will be confronted with data of cabinet formations in France, the Federal Republic of Germany, Italy, Ireland and The Netherlands (1945–1988). We show that these centre‐oriented prediction principles of coalition formation perform well in our data set.

Abstract. We propose a rational choice model of premature cabinet termination involving considerations of expected gain in terms of electoral payoffs, policy payoffs, or portfolio payoffs. This approach, which distinguishes contextual variables that will generally affect the nature of cost‐benefit calculations made by political actors from the factors that are most likely to have a direct impact on a particular decision to precipitate a cabinet crisis, leads us to several testable hypotheses. We provide a first illustrative test of our predictions with data from the Netherlands.