La mobilità dei lavoratori da fonti amministrative e da surveys sulle famiglie: un'analisi comparata

Abstract

In Italy most of the research on worker and job mobility is carried out on social secu-rity (INPS) administrative databases. We investigate the possibility of using the Italian Quarterly Labour Force Survey (RTFL) as an additional source for currently measuring worker mobility. In principle, the motivation for using the RTFL is twofold: it covers (al-most) the whole employment, while INPS data are restricted to dependent employment in the private non-agricultural sector; it is timely, while INPS data are available with a lag of 4-5 years. The RTFL suffers, however, from a serious hindrance: the observation plan of the questionnaire provides fragmentary information on work histories. First, we identify and, to al large extent, control for the main reasons of discrepancy of worker mobility measures from the two sources, i.e., differences in the reference population and in the ob-servation plan. Second, we propose a method capable of largely reconciling the evidence from the two sources. The ‘three-waves linkage’ method – as we call it – exploits the ro-tating panel design of the survey and uses linked data over three waves of the RTFL within a calendar year. Its salient features are presented and some main results are discussed. The empirical analysis is carried out for the year 1995. The evidence suggests that the three-waves linkage method, and thus the corresponding longitudinal datasets from RTFL, cap-tures large part of the worker mobility documented by INPS databases: entirely as worker reallocation measures are concerned; with a significant, but moderate underestimation with respect to gross worker turnover.