This study applies the recently developed bootstrap panel causality test proposed
by Ko´nya (Econ Model 23:978–992, 2006) to investigate the causal link between
happiness and smoking using per capita cigarette consumption and happiness index for five
countries (i.e. Japan, France, Germany, the UK, and the US) over the period of 1961–2003.
A key feature of the bootstrap panel causality is that it is more robust than other methods
due to the generation of country-specific critical values from the bootstrapping method.
Empirical results show a feedback for both Japan and France and independence for the
other three countries. These results indicate smoking make people happy. However, in both
Japan and France people smoke less if they feel happy. To reduce the omitted variable bias,
we also added per capita real GDP as a control variable in our study over the period of
1961–2003. When doing this the empirical results show a feedback for France, a one-way
Granger causality running from happiness to cigarette consumption for both Japan and the
UK, and independence for the other two countries, Germany and the US. These results
indicate smoking make people happy in France. However, in Japan, France and the UK
people smoke less if they feel happy.