Economic Retrospective Voting in American National Elections: A Micro-Analysis

A number of recent studies examine the traditional hypothesis that the electoral fortunes of the incumbent president's party rise and fall in direct relation to fluctua- tions in the state of the national economy. Typically these studies employ a longitudi- nal design in which a party's aggregate congressional vote serves as the dependent variable, and various economic indicators serve as independent variables. On balance. the election returns appear to bear some relation to economic conditions, although various disagreements exist. Using data from the 1956 to 1974 SRC election studies this paper attempts to uncover an individual-level basis for the macro-relationship found by earlier studies. Specifically, do citizens vote for or against the incumbent president's party as a function of their personal economic condition? The survey data permit us to conclude that a citizen's personal economic condition affects his presi- dential vote. For congressional voting, however, the findings are positive until 1960 and negative thereafter. And contrary to some previous research, we find no system- atic relationship between a citizen's personal economic condition and his decision to vote or abstain.