Optimal Welfare-to-Work Programs

Abstract

A Welfare-to-Work (WTW) program is a mix of government expenditures on various labor market policies targeted to the unemployed (e.g., unemployment insurance, job search monitoring, social assistance, wage subsidies). This paper provides a dynamic principal-agent framework suitable for analyzing chief features of an optimal WTW program such as the sequence and duration of the different policies, the dynamic pattern of payments along the unemployment spell, and the emergence of taxes/subsidies upon re-employment. The optimal program endogenously generates an absorbing policy of last resort ("social assistance") characterized by a constant lifetime payment and no active participation by the agent. Human capital depreciation is a necessary condition for policy transitions to be part of an optimal WTW program. The typical sequence of policies is quite simple: the program starts with standard unemployment insurance, then switches into monitored search and, finally, into social assistance. The optimal benefits are decreasing during unemployment insurance and constant during both job search monitoring and social assistance. Whereas taxes (subsidies) can be either increasing or decreasing with duration during unemployment insurance, they must decrease (increase) during a phase of job search monitoring. In a calibration exercise, we use our model to analyze quantitatively the features of the optimal program for the U.S. economy. With respect to the existing U.S. system, the optimal WTW scheme delivers sizeable welfare gains to unskilled workers because the incentives to search for a job can be retained even while delivering more insurance, and using costly monitoring less intensively.JEL Classification: D82, H21, J24, J64, J65.

Training and Search During Unemployment

“…The optimal provision of incentives is achieved by decreasing consumption with the length of the unemployment spell, a well-known result established by Shavell and Weiss (1979) and Hopenhayn and Nicolini (1997). When the level of human capital has depreciated too much, the social planner puts the unemployed on social assistance, paying constant bene…ts without encouragement to enter employment again, as shown by Pavoni and Violante (2007) and Pavoni (2009)…”

“…The exogenous depreciation of human capital determines the optimal timing of these three policies, which naturally a¤ects the optimal consumption pro…le as well. Like Pavoni and Violante (2007), I characterize the optimal dynamics of the unemployment policy but I consider training as an alternative policy. In contrast to the previous policies, the introduction of training endogenizes the change in human capital which has important consequences for the optimal consumption pro…le.…”

“…Since human capital depreciates, it may reach a level at which the returns to providing costly incentives for search are too low and unemployment consumption becomes constant. 5 This paper is closest to Pavoni and Violante (2007) the optimal sequencing of di¤erent unemployment policies and the consequences for the optimal consumption path. Their paper introduces social assistance and job monitoring as alternative policies to the standard unemployment policy which induces search through a wedge between employment and unemployment consumption.…”

“…4 Shimer and Werning (2006) analyze the optimal timing of consumption in a McCall search model, assuming that human capital depreciation reduces the arrival rate of job o¤ers or deteriorates the distribution of the wages being paid on the job. Pavoni and Violante (2007) and Pavoni (2009) analyze the optimal unemployment insurance contract in a model with endogenous search, assuming that the depreciation of human capital reduces the output upon re-employment and the probability to …nd employment. Since human capital depreciates, it may reach a level at which the returns to providing costly incentives for search are too low and unemployment consumption becomes constant.…”

“…Both Pavoni and Violante (2005) and Wunsch (2008) consider the introduction of a training technology in the setup of Pavoni and Violante (2007), but rely on numerical simulations to shed light on the optimal use of the training technology in the unemployment policy. The training technology di¤ers from the technology considered in this paper.…”