ABSTRACT
This paper investigates the wage assimilation of foreign immigrants and internal migrants in Italy, comparing them with stayers. Control for the selection in out-migration is performed using a new duration version of the Heckman correction and by taking into account both return migration and moves to other destinations. Internal migrants experience only minor wage differences when compared with stayers. By contrast, foreign immigrants earn about 8% less than stayers and internal migrants at the beginning of their careers, and the wage gap increases over time. Both language distance and job segmentation contribute to immigrants’ lack of wage assimilation.
ACKNOWLEDGMENTS
The authors thank David Card, Martin Ruhs, Herbert Brűcker, Dan-Olof Rooth, Tommaso Frattini, Letizia Mencarini, Emilio Reyneri, John Dagsvik, Tito Boeri, Daniele Vignoli and participants at the European Society for Population Economics (ESPE), European Association of Labour Economists (EALE), International Migration, Integration and Social Cohesion (IMISCOE) conferences, the European University Institute (EUI) Migration Working Group, the Centre d’Études Prospectives et d’Informations Internationales (CEPII) and Organisation for Economic Co-operation and Development (OECD) conference, and two anonymous referees for helpful comments.
DISCLOSURE STATEMENT
No potential conflict of interest was reported by the authors.
SUPPLEMENTAL DATA
Supplemental data for this article can be accessed http://dx.doi.org/10.1080/00343404.2017.1395003.
ORCID
Daniela Piazzalunga http://orcid.org/0000-0002-4765-4971
Alessandra Venturini http://orcid.org/0000-0001-8858-2743
Claudia Villosio http://orcid.org/0000-0002-1709-4395
Notes
1. For a survey, see Del Boca and Venturini (Citation2005).
2. Developed at LABORatorio Revelli (www.laboratoriorevelli.it/whip).
3. There is no attrition because it is compulsory for firms to provide information about their workers to the INPS.
4. For an analysis of the wage gap for female immigrants in Italy, see Piazzalunga (Citation2015).
5. This feature affects both immigrants and internal migrants.
6. Long-stay immigrants gained access to public employment only in 2013.
7. Often, this is a good proxy for the year of entrance in the country.
8. The summary statistics of the three groups are significantly different from each other.
9. Additional sources of data are described in Table A2 in Appendix A in the supplemental data online.
10. See Adserà and Pytliková (Citation2015) and Bakker et al. (Citation2009) for further details.
11. Following the literature, there is no correction on out-migration of native people. First, the gross emigration rate out of Italy was around 0.1% per annum during this period (Bonifazi, Heins, Strozza, & Vitiello, Citation2009). Second, the WHIP allows one to track and follow workers when they move across Italian regions (thus, internal migrants are always followed). Finally, the likelihood of workers definitively exiting Italian employment is 0.5 times higher for foreign immigrants than for stayers and internal migrants, even when individual, job and career characteristics are controlled for (see Table A4 in Appendix A in the supplemental data online).
12. The same results hold also for the north-east. The high degree of heterogeneity among Italian regions in terms of both levels and dynamics of foreign and internal migration discourages regional disaggregated analyses.
13. does not include periods spent outside employment, which have a negative effect only on stayers’ wages.
14. A test for common coefficient restrictions was run on a pooled regression of (1) immigrants and stayers, (1) immigrants and internal migrants, and (3) stayers and internal migrants. In (1) and (2) the null hypothesis that all the coefficients for immigrants are zero was rejected in both cases; in (3) the null was accepted for internal migrants.
15. In particular, Mocetti and Porello (Citation2010) show that foreign immigration is associated with inflows of highly educated internal migrants and the displacement of low-educated ones.
16. The authors estimate wage equations including cohort fixed effects (defined as the year of entrance into the labour market) instead of individual fixed effects, and the interactions between cohort and experience (see ).
17. NACE = Nomenclature statistique des activités économiques dans la Communauté européenne.
18. Using jobs with a high share of foreign immigrants (instead of LW jobs) yields similar results.