Working Paper: CEPR ID: DP966
Authors: Melvyn G. Coles; Eric Smith
Abstract: This paper analyses how the levels of unemployment and vacancies affect the rate at which unemployed workers find employment -- the worker-firm `matching function'. In particular we test the robustness of previous empirical work by checking whether we obtain the same estimated function using cross-section data rather than aggregate time-series data. We find strong evidence of constant returns to scale similar to previous work. We also find larger cities have higher wages. This provides indirect support for increasing returns, but where those returns are taken in the form of better-quality matches.
Keywords: matching; returns to scale; cross-section estimation
JEL Codes: J64
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| unemployment (J64) | matching function (C78) |
| vacancies (J63) | matching function (C78) |
| city size (R12) | wages (J31) |
| better quality of matches (L15) | higher wages (J39) |
| younger population (J11) | matching rates (C52) |
| more educated workforce (J24) | matching rates (C52) |