The Causal Effect of Election Delay on Union Win Rates: Instrumental Variable Estimates from Two Natural Experiments

Abstract

The role of election delay in union organizing campaigns has become a key policy issue in several countries. Previous studies have documented a negative correlation between delay and union success. However, elections are generally not randomly delayed; in particular, election delay is correlated with various “hard‐to‐observe” union and employer strategies. This article exploits several exogenous sources of variation in election delay to estimate a causal model. The results indicate that two‐stage least squares estimates of the causal effect of election delay on union success are twice the magnitude of standard OLS estimates.