Optimal Unemployment Insurance with Variable Skill Levels

I study the consequences of heterogeneity of skills for efficient unemployment insurance design, using a principal-agent setup. Agents' wages depend on individual productivity and the quality of the worker-firm match. Productivity declines during unemployment and increases on the job. For plausible parameter constellations, more productive agents experience shorter unemployment durations and higher productivity growth under the optimal contract. Benefits do not only depend on employment histories, but also on the skills reported by the agent. Agents are rewarded for accepting jobs that are in line with the insurer's expectations, meaning that agents typically consume more if they find better jobs.