Working Paper: NBER ID: w13287
Authors: Dale T. Mortensen
Abstract: A synthesis of the Lucas-Prescott island model and the Mortensen- Pissarides matching model of unemployment is studied. By assumption, all unmatched workers and jobs are randomly assigned to islands at the beginning of each period and the number of matches that form on a particular island is the minimum of the two realizations. When calibrated to the recently observed averages of U.S. unemployment and vacancy rates, the model fits the observed vacancy-unemployment Beveridge relationship very well and implies an implicit log linear relationship between the job finding rate and the vacancy-unemployment relationship with an elasticity near 0.5. The constrained efficient solution to the model is decentralized by a equilibrium outcome in which wages on each island are determined by a modified auction. Although the efficient solution explains only about 25% of the observed volatility in the U.S. vacancy-unemployment ratio, an equilibrium outcome in which wages are determined as the solution to a strategic bargaining game explains almost all of it.
Keywords: No keywords provided
JEL Codes: E24; E32; J64
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| job finding rate (J68) | vacancy-unemployment ratio (J69) |
| vacancy-unemployment ratio (J69) | job finding rate (J68) |
| constrained efficient solution (D61) | observed volatility in vacancy-unemployment ratio (J69) |
| strategic bargaining game (C79) | observed volatility in vacancy-unemployment ratio (J69) |
| unmatched workers and jobs (J69) | matching process (C78) |