The Contingent Effects of Ballot Initiatives and Candidate Races on Turnout

Abstract

This study analyzes the potential of campaigns for ballot measures and elected offices to draw to the polls citizens who otherwise would not vote. The salience of initiatives and popular referenda in each U.S. state is measured for general elections from 1972 to 1996. Using a pooled time-series cross-sectional data set, a model of state-level turnout is estimated. Distinguishing between midterm and presidential years, the analysis tests for the effects of campaigns across those two kinds of contexts. Contrary to previous studies finding no effects at all for ballot initiatives, the results indicate that these measures-like gubernatorial and senatorial campaigns-provide a boost to turnout during midterm years but not presidential years. Highsalience initiatives and popular referenda are estimated to increase turnout by about four points during midterm elections, as compared to effects for closely contested gubernatorial and senatorial campaigns of four points and three points, respectively. he long-standing debate over the desirability of letting voters decide upon initiatives and popular referenda has intensified in recent years. Supporters claim that ballot measures strengthen democracy and create a more engaged citizenry, ultimately boosting election-day turnout (Schmidt 1989; Zimmerman 1986). A variety of activists, political consultants, journalists, and elected officials make similar assertions (Barabak 1998; National Journal's Congress Daily 1998). Yet it is likely that initiatives and referenda vary in their ability to increase voter participation. Some propositions are marked by extensive media coverage, vigorous campaigns, and high awareness among voters, while others rank low on each of these dimensions. The rational choice approach to turnout provides a theoretical framework for expecting the former kinds of measures to exert a stronger effect on turnout than the latter ones. Riker and Ordeshook (1968), extending the work of Downs (1957), propose that an individual's decision to vote depends positively upon the benefits-the probability of affecting the outcome times the utility gained if the preferred candidate wins-and negatively upon the costs. As for the benefits, it is the most salient initiatives and referenda for which citizens will perceive the greatest differences from voting one way over the other. After all, when an individual knows and cares little about a ballot proposition, his or her perceived utility won't change much regardless of the outcome. Thus, the initiatives and referenda with the highest public salience should show the strongest relationship with turnout because of their impacts upon the benefits of voting. Examining the cost side of the ledger also leads to the expectation that the most salient ballot measures produce the largest increase in turnout. The costs of voting, which include gathering information, fall as voters can acquire it without exerting much effort. Highly salient measuresthose attracting voluminous media attention and lively campaigns-have lower costs than low-visibility measures for which citizens must seek out facts and arguments. By holding down the costs, the presence of salient