Identifying causal effects with panel data: the case of friendship and happiness

Abstract

Previous attempts at estimating causal relationships between the extent of social
contact with friends and happiness rely heavily on standard cross-section econometric
techniques, and therefore do not control for unobserved inborn heterogeneity and
time-varying endogeneity. Using a rich panel dataset from the British Household
Panel Survey, this paper estimates micro-econometric life satisfaction equations and
exploit both single and simultaneous equations panel data estimators to address both
sources of endogeneity: unobserved heterogeneity and time-variant endogeneity.
Conditioning on these unobservables, I find a strong causal effect of friendship on
happiness, which has been greatly understated at the cross-section.