In this dissertation, I analyze the qualitative and quantitative importance of adverse selection for the design of unemployment insurance.;In the first chapter I construct a repeated agency problem to study the optimal unemployment insurance contract. Agents exert observable effort that affects the probability of employment, and they receive idiosyncratic shocks to their utility cost of effort, which remains private information. For agents with low levels of wealth, the presence of hidden information causes expected consumption to be increasing over the length of an unemployment spell. Unlike standard moral hazard models, in the current environment with observable effort but hidden types, the spreading of continuation values in the provision of incentives occurs for both employed and unemployed agents; therefore, declining consumption over the duration of unemployment no longer represents a necessary consequence of providing incentives to unemployed workers.;The standard moral hazard model of unemployment insurance fails to capture the transition from unemployment (searching) to inactivity (not searching) observed in the data. In Chapter 2, I construct a dynamic contracting model of optimal unemployment insurance with adverse selection and moral hazard that does capture this transition. Adverse selection occurs because agents receive unobservable idiosyncratic taste shocks affecting the marginal rate of substitution between consumption and leisure. My model generates novel qualitative and quantitative implications for the provision of unemployment insurance. Qualitatively, for some agents, incentives in the optimal contract imply expected consumption may increase over the duration of unemployment. Quantitatively, compared to a planner who ignores adverse selection and focuses only on moral hazard, the optimal contract achieves an additional 47% of cost savings. Of the extra savings, around 3.2% arises from improved incentives to exert effort, leading to higher expected output. The remaining portion arises from a more efficient allocation of consumption across taste shocks.
展开▼
