A Dynamic Generalization of Becker's Assortative Matching Result

This paper considers a dynamic partnership model in which agents' future productivity depends in part on their current match. Efficient rankings over human capital distributions are characterized in several common stochastic orders. These rankings are used to characterize the returns to human capital for individual agents in the associated market equilibrium. Assortative matching theorems are developed using the planner's characterization results. While positive assortative matching is efficient given sufficient complementarity in static production and dynamic transitions, curvature assumptions on static production are necessary for a robust sorting theory in this dynamic environment