Political Sophistication and Economic Voting in the American Electorate: A Theory of Heterogeneous Attribution

Abstract

This article presents a theory of heterogeneous attribution, in which an individual's ability to make causal associations depends on his level of political sophistication. Specifically, we maintain that political sophistication dramatically influences the relative importance of personal and national economic judgments in shaping presidential candidate preference. Pocketbook voting, we argue, should be common only among relatively voters, who are able to make the associative linkage between changes in their personal financial status and governmental policy. Additionally, we hypothesize that evaluations should be most influential for less voters. Low sophisticates, we argue, will tend to assume that the national economy is entirely in the President's hands (and vote accordingly), whereas more voters understand that the economy is affected by many actors and conditions that are largely beyond the President's control. Testing our theory using data from 1992, 1996, and 1998, we find strong support for our hypotheses. T here is little doubt that changing economic conditions profoundly influence voters' electoral decisions. Since the work of Kramer (1971), a large body of research has accumulated indicating that electoral results at both the state and national levels are driven at least partly by economic fluctuations (Tufte 1975; Hibbs, Rivers, and Vasilatos 1982; Chappell and Keech 1988). There is, however, less consensus on how individual voters, many of whom only the fuzziest of notions about many aspects of politics and government, (Markus 1988, 137), evaluate the economic information they receive and connect it to the political world. Do voters assess political candidates based on their personal economic well being (the pocketbook voting hypothesis), or on the perceived health of the national economy (the sociotropic voting hypothesis)? The answer depends centrally on how voters attribute causal responsibility for changes in both personal and national economic conditions, a process that, we argue, is strongly influenced by the voter's level of political sophistication. In this research, we explicitly incorporate political sophistication into the model of economic voting. For years, researchers have debated the degree of sophistication that voters bring to bear when making politically relevant evaluations (Suzuki 1991; Mackuen, Erikson, and Stimson 1992). At the macro-level, the debate has hinged largely on the difference between voter evaluations based on retrospective assessments and those guided by more sophisticated prospective economic forecasts. This dichotomy is typified by Mackuen, Erikson, and Stimson's (1992) distinction between retrospective voters, or cognitive peasants, and prospective voters, or cognitive bankers. Micro-level theorists have also noted the importance of