The design of government portfolios – that is, the distribution of competencies among government ministries and office holders – has been largely ignored in the study of executive and coalition politics. This article argues that portfolio design is a substantively and theoretically relevant phenomenon that has major implications for the study of institutional design and coalition politics. The authors use comparative data on portfolio design reforms in nine Western European countries since the 1970s to demonstrate how the design of government portfolios changes over time. Specifically, they show that portfolios are changed frequently (on average about once a year) and that such shifts are more likely after changes in the prime ministership or the party composition of the government. These findings suggest a political logic behind these reforms based on the preferences and power of political parties and politicians. They have major implications for the study of institutional design and coalition politics.

Public spending arguably increases with the number of parties in government as each party seeks to secure benefits to its target groups. In this study, two factors that affect the budgetary consequences of multiparty government are identified. The first is the distribution of a priori voting power. An uneven distribution of voting power implies that all government parties are not expected to be equally successful in budgetary negotiations. The second is the degree of impartiality of the public sector. If the public sector is characterized by corruption and other forms of partiality, distributive issues can be expected to gain importance in representative politics. An analysis of data from 30 European countries suggests that changes in the number of government parties are associated with changes in public spending in cases where equally powerful parties are in government and the public sector is relatively partial.