You are here
Unemployment insurance and frequent users: heterogeneous effects of a longer duration of entitlement
The main purpose of the paper is to assess the impact of a change in maximum duration of UI, using a sharp RDD approach. We focus on the January 2001 reform in Italy, that raised the maximum duration from 6 to 9 months for workers aged 50 years or more at the time of dismissal. We match administrative data from different sources to reconstruct working histories of a large sample from the whole population of UI beneficiaries, 1999-2003. According to administrative information and sequences of UI spells, we identify different types of unemployment histories: every year, about 50% of the total amount of UI monetary transfers is granted to workers who repeatedly use them, often with the same employer, and most of them are seasonal workers. The average treatment effect on the duration of UI is around one month, but it is strongly heterogeneous. There is no significant impact for seasonal beneficiaries (an obvious, but noteworthy consequence of their seasonal sequence of employment-unemployment episodes, coupled with the eligibility rule of half a year of full-time job in each of the previous two years), while the impact is roughly two months for standard beneficiaries: that is, the elasticity with respect to the extension of duration from 6 to 9 months is definitely high, around 65%. Finally, the extension of duration does not have any appreciable effect on transitions to employment. Thus, the active component of the UI scheme is not working properly, or not at all. Provisions about services to be provided by local Labour Exchanges are poorly implemented, while the enforcement of conditionality for UI beneficiaries is lacking. Thus, the extension of duration results only in an increase of the total cost of the policy, with no effect on re-employment probabilities. Overall, the Italian UI scheme appears to lag well behind an efficient design, and a substantial fraction of total expenditure goes to non-standard beneficiaries. The functioning of the UI scheme is clearly distorted vis à vis its purposes. Authors of this paper are Roberto Leombruni, Adriano Paggiaro and Ugo Trivellato.
Adriano Paggiaro is Assistant Professor in Statistics at the Department of Economic Statistics, University of Padua.
His main research interests are policy impact evaluation, problems of measurement and analysis of the labor market, longitudinal data statistical models and integration of economic data from different sources.
IRVAPP/Fondazione Bruno Kessler - Via Santa Croce 77 - Trento