DRAFT: PLEASE DO NOT QUOTE WITHOUT PERMISSION OF AUTHORS COMMENTS SOLICITED AND VERY WELCOME Testing the Convergence Hypothesis in Immigrant Academic Achievement: A Longitudinal Analysis

Based on a model of human capital accumulation, we study convergence in academic achievement of native-born and immigrant students. The model captures the idea that human capital accumulation is a dynamic process depending on cognitive achievement accumulated at previous stages of a student’s academic life. This theoretical formalization provides a framework for the empirical analysis of dynamic trajectories of academic performance. We find that foreign-born students start at lower levels of performance in reading and higher levels of performance in math. Foreign-born students catch up in reading and converge to a higher long run level of performance, however. The nativity gap does not fade in time. The theoretical model allows us to predict that foreign-born will converge to higher achievement levels across all ethnicities. This process is slow and convergence takes 14-15 years. Once we add student fixed effects the results are unchanged, the convergence speed reduces however to 1.5-2 years.