Work in Progress

Previous studies have not reached consensus on whether there exists a causal relationship of smoking on wages. This study aims at filling this gap by providing new empirical evidence from a rich survey panel of German individuals. On average, smokers earn 15% less. This raw gap shrinks to 1.5 – 9.5% after controlling for a large set of observable characteristics, comparing siblings, or using past smoking behavior. To further deal with endogeneity, I propose a novel instrument, which provides exogenous variation in smoking behavior: Smoking bans at schools, introduced by the federal states at different years. While OLS estimates appear to be negatively biased, instrumental variable estimates indicate no causal overall effect of smoking on wages. However, significantly different effects in different directions appear for men (positive) and women (negative).