Estimating the Formateur Bonus across Executive Formats and Party Systems

The literature on cabinet proportionality—i.e. the fairness of cabinet payoffs to legislative parties participating in coalition governments—has historically been focused on parliamentary systems. In pure parliamentarism, scholars have generally have found strong support for “Gamson’s Law” qualified by occasional overrepresentation of small parties. As the literature broadened to include semipresidentialism, scholars found a larger “formateur bonus” for coalition-proposing parties holding the presidency, conditional on the degree of constitutional powers afforded to the president. In pure presidential systems, the emerging literature on coalitional presidentialism has found a consistently large and disproportionate payoff for the party of the unipersonal executive, who is always the formateur in such systems. However, these three literatures on cabinet payoffs have largely operated in isolation, making it difficult to estimate, in comparative terms, the size of the formateur bonus across constitutional formats. In this paper, we provide the first large-N, cross-national estimation of the size of the formateur bonus across multiple executive formats. We do so by expanding the new WhoGov dataset (Nyrup and Bramwell, APSR 2020) to include legislative seats shares of all cabinet-represented parties, thus allowing us to measure cabinet coalescence and the size of the formateur bonus in over 3500 country-years of competitive politics in 117 countries in the period from 1966 to 2018. To this we add party-system covariates such as the predominant form of party linkages (V-Dem and V-Party). Our hypothesis is that the formateur bonus should be the largest where executive discretion is high and party programmaticity is low, that is, in unipersonal executives embedded in predominantly clientelistic party systems.