Labor market modeling recognizing latent job attributes and opportunity constraints : an empirical analysis of labor market behavior of Eritrean women

Abstract:\nThis paper analyzes labor market behavior of urban Eritrean women with particular reference to the\nimpact of education, earnings and labor market opportunities. Unlike traditional models of labor\nsupply, which assume that work can be supplied freely in the labor market, we develop a framework\nthat explicitly takes into account the notion of job opportunities and observable sets of feasible jobs.\nThe framework is formulated within a random utility setting in which unobservable jobs can\nconveniently be treated as latent alternatives. The framework can also readily take into account\nobserved restrictions on the sets of feasible jobs.\nThe empirical estimation of the model is based on data from the labor force module of the\n1996/97 Eritrean Household Income and Expenditure Survey for urban areas. We estimate structural\nchoice probabilities of being in the states “Not employed”, “Working in the wage sector”, and\n“Working as self-employed”, where it is taken into account that some women are constrained in their\nlabor market choices.\nWe find that the effect on wages of changes in education level is high; improving the education\nlevels of women greatly improves their wages, which again contributes to bringing more women into\nthe labor force. However, our data do not support the assumption that basic education increases\nwomen’s job opportunities. In order to do so, at least secondary education is required. Keywords: Female labor supply, Random utility model, Returns to education, Labor markets in\ndeveloping countries.