What explains labor reallocation out of agriculture? We propose an accounting framework that leverages observable variation across birth cohorts to study the role of human capital accumulation. We model a dynamic overlapping generations economy where heterogeneous individuals choose whether to stay in or move out of agriculture, subject to mobility frictions. The model shows analytically that labor reallocation within- and across-cohorts pins down the relative role of human capital vs. sectoral prices/productivities in labor reallocation. We apply the framework to micro data from 52 countries. We document novel empirical patterns on labor reallocation by cohort and use them, through the lens of our model, to discipline the size of mobility frictions and show two results: (i) human capital explains one third of labor reallocation, on average; but (ii) it has a minor role in explaining why some countries have faster reallocation than others. Furthermore, we use years of schooling as a direct measure of human capital to validate our main approach and we exploit a large-scale school construction program in Indonesia as a natural experiment to study the effects of an exogenous increase in human capital. We show that the program led to labor reallocation out of agriculture.
