Economic voting in volatile times

This research note examines the effect of employment status and economic assessments on voter turnout in presidential elections over the past two decades-a time period in which unemployment underwent considerable fluctuation. Pooled, cross-sectional data from the 1984-2000 American National Election Studies is used to test the three hypotheses, "preoccupation," "no effect," and "mobilization" that may contribute to an understanding of how individuals respond electorally to their economic situation. The results of the logit analyses suggest that unemployed individuals were less likely to vote in the 1984 and 1988 elections, but this variation vanished in 1992, 1996, and 2000. A negative assessment of the state of national economy had a mixed impact on turnout - depressing turnout in 1988 and 2000, having no effect in 1984 and 1996, and a mobilizing effect in 1992. The effect of a worsening personal financial situation had a depressing effect on turnout in all election years except 1992. Social scientists have often been intrigued by the effect of the state of the economy on voter turnout-primarily in those decades when rising inflation and unemployment were salient political issues. Yet little research is done on the effect on turnout when economic conditions, or perceptions of such, have improved. Do the unemployed react politically even when their plight is not shared by as many as in previous year? This research provides an updated analysis of this linkage by analyzing the 1984-2000 period when unemployment was initially high but then lessened considerably (Southwell, 1996).