Are temporary jobs a port of entry into permanent employment? Evidence from matched employer-employee data

Updated version. It was "Temporary jobs: Port of entry, Trap, or just Unobserved Heterogeneity?"_x000d_ _x000d_ Are temporary jobs a port of entry into permanent employment? In this paper we argue that the answer crucially depends on the type of temporary contracts being considered, as the different contracts observed in practice are typically characterized by varying combinations of training, tax-incentives and EPL provisions. We base our empirical evidence on a longitudinal sample of labour market entrants in Italy, a country where a large number of temporary contracts coexist with a relatively high employment protection for standard employees. We estimate dynamic multinomial logit models with fixed effects, to allow for non-random sorting of workers into the different types of contracts. We show that the transition to permanent employment is more likely for individuals holding any type of temporary contracts than for the unemployed, thus broadly confirming the existence of port-of-entry e¤ects. Yet, not all temporary contracts are the same: training contracts are the best port of entry, while freelance contracts are the worst. We also show that temporary contracts are generally a port-of-entry into a permanent position within the same employer, but not across firms, implying that little general-purpose training is gained while on temporary jobs. Moreover, the time needed for an internal transformation from a temporary to a permanent position appears rather long, suggesting that firms are likely to use (a sequence of) temporary contracts as a cost-reduction strategy, rather than as a screening device for newly hired workers._x000d_ _x000d_ Jel classification: J41, J63_x000d_ _x000d_ Key words: temporary jobs, port of entry, matched employer-employee data, dynamic multinomial logit models, state dependence, fixed effects