| Political parties bargain over the allocation of cabinet portfolios when forming coalition governments. Non-cooperative theories of legislative bargaining typically predict the formation of only minimal winning coalitions, and that the "formateur" of the government coalition enjoys a disproportionate share of government ministry positions. However, the empirical evidence seems to contradict these theoretical predictions for at least two reasons. First, only one third of all governments are minimal winning coalition (the rest are either minority or super-majority coalitions) second, coalition parties receive shares of cabinet portfolios that are almost perfectly proportional to their nominal voting weight, including the formateur party. This paper examines government formation as a process in which the role of government formateur is determined endogenously, or within a coalition. Furthermore, parties are assumed to have different preferences over cabinet positions and policy outcomes. In equilibrium, if parties have similar preferences over office benefits and policy outcomes, the share of seats the parties are allocated will be proportional to the parties' sizes. However, when parties are relatively more concerned about policy than they are about office benefits, minority governments can be obtained as an equilibrium outcome. In this case, the government coalition gets the external support of those parties whose policy concerns are sufficiently high to make them willing to sacrifice office perks entirely in order to obtain a government policy that is closer to their ideal policy. |
