Empirical models of government choice

In this chapter, we discuss the evolution of the quantitative methods used to study why we get the governments we do. Government choice can be thought of as a multinomial discrete choice model in which one potential government is selected out of the set of all potential governments in a formation opportunity. Because of its ability to provide additional substantive insights, as well as handle unobserved heterogeneity and relax the independence of irrelevant alternatives assumption, we recommend that scholars employ the random coefficients mixed logit model for studying government choice. We demonstrate how a simple reconceptualization of potential governments allows us to jointly model the choice of government and prime ministerial party, thereby narrowing the gap between theory and empirics. We also highlight some of the challenges that arise when trying to make substantive inferences about real-world implications from models of government choice.