The Evolution of the Intergenerational Mobility of Education in Chile by Cohorts: Facts and Possible Causes
Documento de Trabajo IE-PUC, N° 348, 2009.
Abstract: In this paper we estimate the evolution of the intergenerational mobility of education in Chile. Even though there have been estimates of the intergenerational mobility of education before, these estimates were for one single year. Here we estimate its evolution for synthetic cohorts born between 1930 and 1978. We do so for three different data sets and obtain practically identical results for all three. We find a sharp and continuous fall from 0.67 for the cohort born in 1930 to 0.41 for the cohort born in 1956, followed by a stagnation of the mobility from then on. To find an explanation for this we study the evolution of educational attainment and examine how it has evolved for children of parents with different educational levels. We conclude that one key factor in stopping the increase in educational mobility has been the difficulty in accessing college by children of parents with low educational attainment. The second part of the paper attempts to explain this evolution. We test three possible explanations. The first is that when mobility increased, mobility was driven by laws that made further education mandatory. We find this hypothesis has practically no explanatory power. The second tests the idea that mobility stopped increasing when increased mobility implied middle class children had to go into college and they faced a financial restriction. We test this hypothesis against the alternative hypothesis that what matters are financial restrictions around the date of birth (that, through a production function “a la Heckman” where human capital formation at all ages are strong complements implies very low rates of return to college). We find that the latter restriction is more important. Finally, our third hypothesis tests whether the increase in single parent households explains the stagnation in mobility. We find that when we use that series by itself to explain mobility it has substantial explanatory power. However, when we include this variable together with the financial restrictions at college entry age and at birth, these latter restrictions leave family composition with no explanatory power. Hence we find that the reason behind the stagnation lies in Chile’s economic stagnation in the years following the Second World War. Hence it could be expected that cohorts born beyond 1978 (cohorts we are not able to study) and in particular after the economic boom that started in the mid eighties may resume the increase in mobility followed up to the cohort born in 1956.