Collaborative Research: On the Electoral Link Between Presidents and Assemblies: Presidential Coattails in Comparative Perspective

This project assesses the institutional sources of presidential coattail effects in the world's presidential democracies. Although coattail effects are studied extensively in the United States, virtually no comparative work exists on the subject. Scholars of American politics agree that presidential coattails can affect the balance of seats in the legislature and, consequently, can influence executive-legislative relations more generally. We propose to explore this dynamic in comparative perspective.

Prominent comparativists, such as Arend Lijphart, Juan Linz and Arturo Valenzuela, argue that presidentialism is inferior to parliamentarism because the necessity of inter-branch negotiations opens the possibility of executive-legislative deadlock. Deadlock figures prominently in recent crises in countries such as Brazil, Peru and Russia and lends credence to these views. In its mildest form, scholars argue that deadlock can prevent governments from responding to immediate crises, such as needed fiscal adjustment following international shocks. In its more moderate form, deadlock can prevent governments from implementing deeper political changes, such as social security reform or privatization, and, in its most extreme form, executive-legislative deadlock even can lead to democratic breakdown. Although scholars have explored some of the institutional sources of deadlock, and also have argued that executive elections can affect legislative elections, none have gathered the requisite data and developed hypotheses to explain the absence or presence of coattail effects in presidential democracies.

National Science Foundation support is requested to gather constituency-level data on election results, electoral laws and political parties in 31 presidential democracies. A database is constructed, and the degree of presidential coattails -- operationalized as the degree of straight-ticket voting -- is determined. Statistical analysis is employed to test hypotheses about the salient institutional variables associated with stronger or weaker presidential coattails. The following institutional variables are hypothesized to affect the strength of presidential coattails: ballot structure, district magnitude, the timing of elections for different offices, the presence and timing of subnational executive elections, the degree of party institutionalization, the extent of control by the central party over the nomination procedure for legislative office, and the level of central party control over campaign finance and media access.

This research is vital for students of presidentialism and democratic governance generally, because in contrast to existing work, it identifies the institutional configurations that foster or impede strong electoral links between the incumbent president and members of the legislature, as opposed to purely institutional links such as decree and veto powers. This, in turn, informs future work about the conditions under which executive-legislative deadlock emerges and when democratic presidential systems are more or less likely to function smoothly, break down, or consolidate. In short, this research provides a substantial addition to the literature on democratic governance in presidential systems.

The database that emerges from this research project also represents a unique resource for students of elections, electoral systems, and legislative behavior. As Gary Cox notes in Making Votes Count, the future of electoral systems research lies in the analysis of district-level data, data which unfortunately have not been collected in a systematic manner for more than a handful of countries. The book that results from this project, as well as the future work by scholars using our data, will lead to a substantial improvement in scholarly understanding of the consequences of political institutions for democratic governance.