Abstract:
There is a widespread feeling that welfare states are distorting incentives to a larger extent than what is socially optimal. In the present paper, this idea is formally analyzed with a model in which a benevolent government seeks to maximize the utility of risk-averse-opportunistic agents. The government can redistribute output from the "lucky" to the "unlucky", providing insurance. Under commitment, the government optimizes before agents choose the level of effort. Aware of incentive problems, the government will typically provide incomplete insurance, in order to induce some effort above the minimum. Under discretion, the government reoptimizes after agents chose effort. Thus, it faces a pure risk-sharing problem and provides full insurance. Agents anticipate government's actions so they choose the minimum effort. Social welfare might be lower under discretion: there would be overinsurance.