Abstract
This paper investigates the extent to which voters respond to threats to the nation by estimating the electoral effects of one of the most severe international crises in American history, i.e. the Cuban Missile Crisis. To establish causality, I exploit both the timing of the crisis, which happened two weeks before the 1962 midterm elections, and the geographical variation in the range of Soviet missiles: only localities within 1,000 nautical miles from Cuba could be targeted. Difference-in-differences regressions comparing changes in electoral outcomes of counties lying within and outside the range of the missiles show that the crisis increased voter turnout of about 5 percentage points. Voters who had the largest returns from the peaceful resolution of the crisis, i.e. those living within the missile range, rewarded the President’s party: vote shares for the Democrats increased by 3 percentage points in target counties. Historical surveys finally provide first-stage evidence of increased fear and concerns over Cuba among respondents living within the range of Soviet rockets.
Local Organizers: Annalisa Loviglio, Massimiliano Onorato, Alessandro Sforza