The Draft and the Ballot Box The Effect of Conscription Risk on American Political Behavior: 1969--1972

Under what circumstances does direct exposure to government policies induce political participation on the part of ordinary citizens? A number of scholarly studies have reached opposing conclusions about the extent to which an individual's personal proximity to an issue, also called "self interest," corresponds to his or her attitudes about particular policies (Sears and Funk 1991). Despite exceptions, for some time, scholars have maintained a general conclusion that one's individual stake in a policy matter little with respect to attitude formation (Taber 2003). Recent reconsideration of this literature with a renewed focus on causal inference and identification strategy has begun to call into question the conclusion that self interest has no bearing on an individual's political attitudes.;Outside of the debate on attitude formation, there is very little known about the extent to which one's being affected by a policy affects his or her political engagement and participation. In recent and past studies alike, behavioral outcomes have scarcely been addressed, and few studies assess the extent to which policy exposure or "self interest" affects political participation. Responsive participation on behalf of one's interests and preferences is the cornerstone of representative government, and key theories of political science rest on the assumption that citizens vote, not only to influence the course of prospective policymaking, but to render judgment on existing policies. Despite the centrality of the question, there are very few empirical studies of behavioral responses to government policies.;This research examines the effects of exposure to a particularly strong form of government policy, military conscription, on the political behavior of American draft-age men and their families during the final stages of the Vietnam War and the military draft in America. I exploit the nearly-randomized natural experiment of the Vietnam draft lotteries of 1969, 1970, and 1971 to measure the effect of draft risk as assigned by birth date, the priority of which was determined through random drawings, on the 1972 voter registration and turnout of men and their parents.;I measure the direct effects of draft policy exposure on political behavior as well as the moderating effects of social context in order to test three basic hypotheses. First, the Responsive Engagement Hypothesis states that policies have the capacity to engage new segments of the population by creating new incentives for engagement motivated by material self interest. Second, the Policy Exposure Hypothesis states that responsive participation will be greater among individuals affected by policy implementation relative to unaffected members of common issue public. The third and final Contextual Moderator Hypothesis advances that, among members of an affected subconstituency, rates of responsive participation will be highest in contexts within which issue is made salient by related events.;In measuring the ellects of draft risk on the political engagement of draft eligible men, I identify several complications presented by the data. Despite the limitations, I find modest support for the first two hypotheses. Lottery numbers are inversely related to registration and turnout in 1972 meaning that higher levels of risk caused higher rates of subsequent engagement. Among parents, I find support for the second two hypotheses. Parents of men assigned to low numbers, who therefore faced a higher rate of induction, voted at rates higher than parents of men who were relatively safe from induction. I find the effects of draft risk on parents' turnout are particularly strong in the context of prior local casualties.;Chapter 1 introduces the project. Chapter 2 provides an overview of military draft policy in the United States and describes the social and political context of the draft lotteries of 1969, 1970, and 1971. In Chapter 3, presents the theoretical framework for the research and sets the analysis within the context of extant political science literature. Chapter 4 details the identification strategy, offers a thorough qualification of the draft lotteries as a randomized natural experiment, and describes the original data sets. Chapter 5, presents analysis effects of a son's draft risk on the political engagement and participation of his parents in the 1972 election and considers the role of contextual attributes, primarily prior war casualties, in producing heterogeneity of treatment effects across different information environments. Chapter 6 presents exploratory analysis of the effects of draft risk on draft eligible men and identifies several complications with observing the effects of policies which may cause nonrandom attrition. Chapter 7 concludes the dissertation and offers suggestions for future research.