Empirical research on the relationship between illicit drug use and labour
market success has been found to have mixed results in the literature. Relevant
sources of variability are the methods used to account for the potential
endogeneity of drug use. Using clinical data of drug users, this paper utilizes a
recursive simultaneous-equations approach as an alternative for estimating the
effect of consumption on labour participation and control the endogeneity
problem. Our results confirm that drug use is endogenously determined, and
provides evidence to support that frequent use of dependency drugs greatly
decreases the likelihood of be in work. The results derive here show the
potentially labour market-related costs of drug use and abuse, mainly in terms
of production loss.