Prime Minister replacements, incumbency, and coalition formation

Despite the existence of several, constitutionally provided models of cabinet decision-making in cabinets, Prime ministers have become the publicly most visible and thus key actors in modern governments. Parties’ candidates for the office of Prime Minister structure the election campaign and have a significant influence on the policy profile of their parties and, when winning control over the office as Prime Minister, governmental decision-making. While theories of coalition governance focus on the Prime Minister, her party affiliation and policy preferences, most theories of coalition formation do not take into account the pivotal role the Prime Minister can have for the outcome of the government formation process. We aim to contribute to this gap in the literature on coalition formation and argue that a personal change in the office of the Prime Minister in the outgoing legislative period will have an indirect impact on the chances that an incumbent coalition will re-form the government after an election. Because Prime Ministers elected during an ongoing legislative period try to boost their career chances by taking a specific public profile that is different from their predecessors, they should be more likely to switch coalition partners in the next government formation process. We evaluate this hypothesis by means of original data on government formation processes in the German federal states from 1990 to 2023. The findings support our expectation: integrating a more fine-grained version of the incumbency variable that covers information on the changes in the office of the Prime Minister during the legislative periods increases the predictive power of the regression models when analyzing government formation.