Skip navigation.

Policy-bundling and Religion: An Experimental Approach

TitlePolicy-bundling and Religion: An Experimental Approach
Publication TypeConference Paper
Year of Publication2014
AuthorsGallego, Aina, and Jonathan Rodden
Conference Name3rd European Conference on Comparative Electoral Research
Keywordsexperiments, issue bundling, vote choice

This paper studies vote choices in two-dimensional policy spaces. In particular, it focuses on policy disagreements related to economic redistribution and to religion, two dimensions that have continued to provide structure to political conflict in many societies around the world. One of the most basic claims in comparative political science is that multi-dimensional politics plays out very differently in countries with multiple political parties than in countries with only two parties. In a multi-party system, parties can more fully occupy a two-dimensional issue space, providing alternatives for cross-pressured voters— those with right-of-center preferences on one issue but left-of-center preferences on the other—to find parties with proximate views on both dimensions. By contrast, a strict two-party system in which economic and moral issues are bundled together by the parties offers no such alternatives.
We ask: what are the implications of the policy-bundling that forces such choices upon cross-pressured voters in two-party systems? There are a number of claims in the literature, but most are extremely difficult to substantiate with traditional survey data within or across countries. This paper uses a survey experiment to examine the implications of policy bundling for vote choice.
We present results of survey experiments conducted in the US and in Spain that vary the information provided about candidate. We contrast conditions with and without issue bundling, and discover that in a hypothetical two-party system with policy bundling, the economically left-wing party receives far fewer votes than in the other conditions. Our survey experiments allows for a direct comparison of the limited menu of choices available in a two party system with the fuller menu of candidates that is typical of a system of proportional representation. This approach reveals a rather striking asymmetry that is consistent with the classic Marxist story: policy-bundling favors the candidates of the economic right, in large part because religious voters are less willing to suppress their moral values preferences than are secular voters.