A Multiple-State Non-Stationary Model of Welfare Exit - HAL-SHS - Sciences de l'Homme et de la Société Accéder directement au contenu
Communication Dans Un Congrès Année : 2003

A Multiple-State Non-Stationary Model of Welfare Exit

Résumé

We propose a structural discrete-time dynamic model of labor market transitions that incorporates general forms of non-stationarity resulting from a variety of major institutional figures that can affect current and future disposable income, such as unemployment insurance or welfare-to-work incentive programs. We propose a structural discrete-time dynamic model of labor market transitions that incorporates general forms of non-stationarity resulting from a variety of major institutional features that can affect current and future disposable income, such as unemployment insurance or welfare-to-work incentive programs. In the literature, these institutional frameworks are usually oversimplified in order to avoid the complexities that this model handles. Since an analytic solution is generally unavailable, however, we provide a simple algorithm for solving the model numerically. This modeling strategy is applied to the welfare system in France. There are two sources of non-stationarity: individuals can cumulate welfare payments and labor earnings for a fixed time period after leaving welfare; and the length of this period is a function of previous labor market history. The structural model, including preference parameters over part-time and full-time work, is estimated based on survey longitudinal data of welfare recipients. We find that reservation wages for welfare recipients are indeed sensitive to nonstationary elements of the individual's labor market history.
Fichier non déposé

Dates et versions

halshs-00365550 , version 1 (03-03-2009)

Identifiants

  • HAL Id : halshs-00365550 , version 1

Citer

Marc Gurgand, David Margolis. A Multiple-State Non-Stationary Model of Welfare Exit. Econometric Society European Meetings, Aug 2003, Stockholm, Sweden. ⟨halshs-00365550⟩
89 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More