Abstract. This article examines the determinants of the salaries that Spanish university graduates earn on the labor market. Different earnings equations are estimated that allow us to measure the economic returns to investment in human capital at the university level, demonstrating that: on the one hand, considering schooling to be an exogenous variable gives a downward bias to the estimations of the private rates of return to an university education; on the other hand, not taking into account the aspects of the demand-side of the labor market in the traditional Mincerian earnings function, even though schooling is considered as an endogenous variable, the rates of return estimated for an university education would be given an upward bias. The problem concerning the endogeneity of schooling has been corrected in this article by using the instrumental variables technique.
Keywords: demand for schooling, earnings Differential, human capital, rate of return
Education and earnings are positively linked. "The universality of this positive association between education and earnings is one of the most striking findings of modern social science." (Blaug 1972, p. 54). Ever since the precursory works of Becker (1964) and Mincer (1974), numerous contributions to the field of Economics of Education have shown, for Different countries and time periods, that an individual's academic qualifications play an important role in establishing the salary he or she receives. In addition, particularly following Card and Krueger's (1992) influential publication demonstrating that there is a positive significant association between the quality of schools--measured using indices such as the teacher/pupil ratio or teachers' income--and the incomes students earn on the labor market, a great number of research papers in recent years have concentrated on analyzing the influence the quality of the schooling received has on an individual's earnings (James et al. 1989; Jones and Jackson 1990; Betts 1995, 1996a, b; Brewer et al. 1996; Grogger 1996; Heckman et al. 1996; Johnson and Stafford 1996; Card and Krueger 1998). (2)
The fact is that, today, there is a wealth of literature regarding the estimation of education's rates of return. The empirical tool used in most of this research has been the traditional Mincerian earnings function (Psacharopoulos 1994; Cohn and Addison 1998; Asplund and Pereira 1999; Harmon et al. 2001, 2003). This function, proposed by Mincer (1974), is a semi-logarithmic function, such that an individual's earnings vary linearly with the time dedicated to education and quadratically with experience. (3)
However the estimations of the rates of return to education have been controversial right from the outset. The criticisms concern the main econometric problems we come across when using the Mincerian function, as mentioned by Griliches (1977). These problems arise when: (i) certain variables are not included in the human capital earnings function--for example, an individual's ability or aptitude, which is supposedly correlated to education; (ii) the quantity of schooling is not measured correctly; (iii) schooling is treated as an exogenous variable. For these reasons, a large proportion of the most recent literature on education returns has concentrated on...