Representation at the Margins: Partisan Politics and Public Money in Canada

This doctoral dissertation is a study of how political parties influence budgetary policy in Canada. It includes three empirical chapters that look at different aspects of partisan representation. Taken together, these chapters shed light on the role of parties and institutions in policy.;There is a fertile literature in political science that argues that parties do not matter as much as they should in Canada. This dissertation provides an in-depth analysis of partisan representation that contradicts this position. It shows that parties play a significant role in budgetary policy, if only at the margins.;In the first manuscript, we provide a comprehensive classification of ridings that identifies electoral strongholds and battlegrounds. We show how party support stability can be explained by the presence of certain socio-economic characteristics. This first piece fills a gap in the literature on electoral politics by focusing exclusively on meso-level representation of voters' interests. Building on what is found in this first manuscript and the literature on the role of parties in budgetary politics, the second manuscript tests empirically for the existence of partisan effects in budgetary policy between governing parties and across party systems at the federal level. The results suggest that governing parties show partisan differences in spending that are consistent with their electoral incentives. This second piece also makes an important methodological contribution to the discipline, offering a robust and versatile model specification that is exportable to other parliamentary systems. It is impossible to fully understand the impact of federalism on public spending if one looks only at the federal government since part of the budgetary story occurs at the provincial level. The third manuscript complements the findings of the second by looking at provincial budgetary policy. We first measure the magnitude of policy convergence between provinces in taxing and spending. Then we test for the presence of a partisan effect despite policy convergence. We find evidence in support of both effects.