Preference Bundles in the Explanation of Americans' Policy Preferences

• In Event: Political Knowledge and Information Processing

Abstract

One way to explain the formation of citizens' policy preferences is in terms of the concept of a "preference bundle." A preference bundle is a group of two or more policy preferences that the informed public perceives as constituting a discrete political identity. For example, the bundle {–abortion, +execution, +gun rights, –redistribution} is perceived by the informed public as a typical "conservative" bundle ("+" = support; "-" = oppose). By contrast, the bundle {+abortion, –execution, +gun rights, –redistribution} is perceived by the informed public as a typical "libertarian" bundle.

This paper describes several psychological mechanisms by which preference bundles affect the political preferences of ordinary citizens. A preference bundle can be understood as a belief about "which preferences go together," and people's beliefs about which preferences go together can affect the preferences that they actually come to hold. For instance, a citizen may become attracted to a policy preference because it is part of a bundle whose other preferences she already holds.

The paper also describes different ways in which a preference bundle can be formed in the first place. Briefly, a preference bundle can arise through historical contingency (as when a political party bundles together two unrelated preferences in order to win votes from two different groups) or through psychological affinity (as when, for psychological reasons, the types of voters who hold one preference also hold another preference). In general, political scientists have shown much interest in the first process (historical contingency) but little interest in the second (psychological affinity).

Clarifying the concept of a preference bundle can help to sharpen analysis in political psychology. For example, it is not useful, generally, to index the meaning of the terms "liberal" ("conservative") to the set of policy preferences that the Democratic (Republican) Party happens to espouse at a given time. Some of the policy preferences that a political party espouses may have no psychological connection to each other. For example, there is no reason to assume that the preference {+environmental protection} is an intrinsically liberal (as opposed to intrinsically conservative) preference. An important challenge for political psychology is to develop tools for distinguishing between (1) policy preferences that are constitutive of a given political identity and (2) policy preferences that are associated with a political identity due mainly to historical accident.

Author

• Philip Petrov