Instrumental variables in structural equation modelling: an application on the impact of labour factors on health and standard of livings

Employment is considered a significant determinant of living standards around the globe. For most people, work is the primary source of income, and job-related events and shocks are common causes for individuals and households to escape from poverty or fall into it. Along with employment, it is well known that the effects of health status on a worker's productivity are critical as healthy workers are more productive when they work, and they can earn more, also enhancing their standard of living. This study exploits the exogenous variations created by the employment support programmes of 2008 and 2011 implemented in Turkey, targeting women and young men. To investigate the impact of those programmes, we exploit the exogenous variations created, and we evaluate their effect on health and standard of living (SoL) through various labour factors. We apply the structural equation modelling (SEM) and estimate the additional health-related costs. Moreover, we propose an instrumental variables (IV) approach within the SEM framework to deal with the endogeneity coming from the reverse causality between labour factors, health and SoL.