LIBRI DI GIOVANNI BARBIERI

The Luxembourg process placed strong emphasis on the reorganisation of the Pes as essential contribution for a successful reform of labour market. Being in Italy the reform process at an early stage, this paper explores the decision to enrol to the old Pes and its effects in terms of job finding opportunities. Following a non-experimental approach, the evaluation of the impact of the enrolment, obtained by comparing the treatment group with a control group, is likely to be biased because of self-selection problems. In order to overcome this problem, two different techniques are applied: a matching approach based on propensity score, that control for differences between the two groups in terms of observed variables, and an instrumental variables methodology, that takes into account also unobserved heterogeneity. Both experiments show negligible effects of the Pes status upon job chances. The approach we present, applied at Lfs longitudinal data, may also help in the forthcoming future to evaluate the effectiveness of the Pes reform.