No causal effect of unemployment on smoking? A German panel study.

OBJECTIVES

This study analyses the effects of different unemployment durations on smoking behaviour in Germany by investigating smoking take-up, relapse, quitting and smoking intensity.

METHODS

Longitudinal data from the German Socio-Economic Panel from the years 1998, 2001, 2002, 2004, 2006, and 2008 were used to examine the effect of unemployment (52,940 observations from 17,028 respondents, aged 17-65 years). Unemployment duration was measured at 1-6, 7-12, 13-24, and 24+ months. Effects were estimated using zero-inflated negative binomial regressions and fixed effects logistic panel regressions.

RESULTS

The zero-inflated negative binomial regression models suggest that the likelihood of smoking increases with unemployment, while smoking intensity is not affected. However, fixed effects logistic regression models demonstrate that unemployment is neither a significant predictor for taking up smoking, relapsing, nor quitting.

CONCLUSIONS

The results indicate that in Germany, there is no direct causal effect of unemployment on smoking behaviour. The observed relationship between smoking and unemployment appears to be driven by stable, unobserved differences between employed and unemployed respondents.