Dynamics and inertia in the Italian labour market and policy evaluation (databases, measurement issues, substantive analyses)


Research Project supported by the Italian Ministry of Education and Scientific Research, 2001

[Home] [Versione italiana] [Printer-friendly version]

Projects | 1997 | 1999 | 2001 | 2003 | 2005 



The goals of the programme are described with reference to each of the four broad classes of issues we plan to pay attention to. A partly intermediate goal, but with an autonomous relevance too, has to do with the development of suitable micro databases, in some case by record linkage and integration of different data sources.On this latter topic, see the projects of the various Research Units (RU).


(A1) Our first task will be to improve on the formulation of a model of labour market attachment in the presence of fallible indicators of the labour force state, proposed by Rettore and Trivellato (1998). We will extend the analysis of the relevance of the classification of not-employed persons in a dynamic setting, along the lines suggested by Jones and Riddell (1999) and within a dynamic latent class analysis set-up. By building a panel database from the ILFS, we will also proceed to improve the modelling of unemployment duration in the presence of measurement errors.

(A2) Secondly, we will focus on worker mobility, as documented in the SS longitudinal database. We will extend the analysis to all the working forms, atypical contracts and self-employment included, and we will pay attention to geographical differentials and cyclical features of associations and separations. We will analyse also wage mobility.

(A3) Thirdly, we will compare mobility estimates resulting the ILFS and from the SS longitudinal database, respectively, with several purposes: to control for various heterogeneity factors; to ascertain to what extent we can reconcile mobility indicators from the two sources; to have more "robust" indicators of mobility.

Research units: Padova for (A1); Torino for (A2); Padova and Torino for (A3).


(B1) A first set of objectives deals with the study of workers' careers, on the basis of a SS database of the entire working population of two provinces of the Veneto region, 1974-1997. We plan to study: (i) the different mobility paths of prime age workers and, by contrast, of some weak segments of the labour force; (ii) the different transitions patterns to self-employment of two samples of people at present self-employed: one of people fired, another of people that left voluntary their work.

(B2) Based on the Netlabor database for the Veneto region, we will analyse the paths of some specific segments of workers. Specifically, we will pay specific attention to (i) the work provided through temping agencies - a type of work recently introduced in the Italian legislation - and to the dynamics of entry and the work histories of immigrants from outside the European Union.

(B3) Finally, we will carry out a pilot-study about the linking of individual records from two sources, Netlabor and the ILFS, still for the Veneto region. The main goal of the exercise will be to check differences in coverage and to derive elements for assessing the quality of the register.

Resarch units: Venezia for (B1); Padova for (B2) e (B3).


(C1) Large part of this research line will be carried out using the ECHP four-wave panel. The main objectives are: (i) methodological refinements in various directions: appropriate definition of the longitudinal unit of analysis (the "longitudinal household"), variable measurement (especially for variables defining the degree of participation at work and the degree of poverty), specification of suitable causal models for panel data; (ii) using the above instruments for empirical analyses, with Italian and other European countries data.

(C2) Secondly, we plan to extend a model of poverty persistence recently proposed by Giraldo, Rettore e Trivellato (2001). We will introduce explanatory variables into the model and using Chamberline-type estimators, in order to identify the relevant events - mainly associated with labour force participation and the dynamics of household composition - which affect transitions to/out of the poverty and their permanent or transitory impact.

(C3) In order to study the inter-relationships between labour and living conditions at the individual and household level, we will complete the construction of an integrated micro database, resulting form the linkage of individual records from two sources: ILFS and SS files. Potentially, the integrated database is quite interesting, because the two sources provide largely complementary information. We plan to start working on this integrated database with the target of an in-depth quality assessment. Then, we will use it for specifying and estimating dynamic model accounting for the role of the household context and models of job search using information on labour income.

Research units: Siena for (C1); Padova for (C2) and (C3) (possibly with interactions).


(D1) A first policy investigated by the research project is a number of programs that provide incentives to businesses that operate in areas with declining industrial production (EU "Objective 2" areas). We aim at evaluating the employment impact of those programs. Within this framework, the impact analysis should be focused on the comparison between employment patterns of "Objective 2" areas and those of areas excluded from the program. At this level of aggregation (not at the firm level) it is possible to detect the impact of the availability of a more generous package of incentives. To that purpose we will use a methodology based on a non experimental comparison group design, in which the unit of observation is not the individual firm but the geographic areas targeted by the program (see, e.g., Bondonio e Engberg, 1999).

(D2) A second programme we will evaluate is known as "Mobility Lists". It combines income support to eligible dismissed employees with substantial benefits to the employers who hire them (for a recent paper, see Paggiaro and Trivellato 2000). We plan to link the information from the administrative database used for running the programme with information from the Netlabor archives, at the individual level. This will allow us to have a richer description of the segments of work histories of the workers registered in the List, temporary contracts included; to analyse their stay in/exit from the List in a more flexible way by means of multi-state models; to assess in a better way the operating of the programme and the impact of its various ingredients

(D3) The evaluation of training-job contracts is our third case study. We will exploit the variability of the regulations on this and other contracts over time, and of the associated benefits for the firms using them, as a sort of natural experiment in order to assess the differential effects of the programme on youth employment.

(D4) The last type of policy we will consider for our evaluation exercises will be vocational training programmes. We will analyse a set of vocational training programmes for the province of Bozen (with data collected through follow-up surveys on trainees), and we will possibly extend the analysis to vocational training programmes in the Veneto region (if - as it seems likely - we will be able to get the list of trainees and to match information on programme participation with their work history resulting from the Netlabor archives).

Research units: Piemonte Or. for (D1); Padova for (D2) and (D4); Padova and Torino for (D3).


Info: anna.giraldo@unipd.it