Chris Sandmann from the London School of Economics and Political Science will present "Probabilistic Assortative Matching and Search: TU and NTU (joint with Nicolas Bonneton)

Abstract: This paper re-visits the impact of search frictions in matching markets. In our model, idiosyncratic shocks generate probabilistic matching patterns; pairs with greater match payoffs are likelier to match. This nuance motivates a novel definition with testable axioms—probabilistic assortative matching (P-PAM). P-PAM demands that match probabilities center around complementary types. We show that they don’t: when match payoffs derive from Nash bargaining over output (TU paradigm), search frictions undermine P-PAM regardless of the strength of complementarity: At the top likeliest partner types always exceed complementary types. This occurs because search frictions disproportionately erode the bargaining power of high-productivity agents. Weaker sorting properties robustly hold, by contrast, even in non-stationary environments. Under exogenous payoffs (NTU paradigm), P-PAM can occur but only so under a restrictive condition: no complementarities between types. Using the theory of FBSDEs, we also prove equilibrium uniqueness when the time horizon is finite and the search pool evolves stochastically.

Slides