Increasing Returns to Schooling by Ability ? A Comparison Between US and Sweden

Using US survey data (NLSY) and Swedish register data the relationship between returns to schooling and ability is estimated separately for each country. A significant and positive relationship is found for Sweden, but not for US. The purpose of this study is to propose an explanation to why such differences might occur. While many studies have focused on whether credit constraints results in inefficiencies in the schooling market our study answers the opposite question; if weak credit constraints lead to inefficiencies, i.e. in too much use of the schooling system by the wrong people. We argue that the US schooling system is more effectively sorting out less productive investments in education than the Swedish schooling system, and therefore it is an imperfect allocation of individuals going to higher education in Sweden that make a relationship between returns to schooling and ability to be observed in Sweden but not in the US. Since the relationship between returns to schooling and ability is the same when the schooling systems of the two countries are more similar, i.e., at lower levels of education, is indicative of that our hypothesis can be correct. The empirical findings in this study are of course not convincing evidence on its own. But the results, and the common differences in IV and OLS returns to schooling estimates, suggests and agrees with such an explanation. JEL classification: J24, I21