Sergio Destefanis (University of Salerno)
Fernanda Mazzotta and Lavinia Parisi
This study estimates a Beveridge curve for Italy at the provincial (NUTS3) level over the 2017–2024 period, distinguishing between female and male unemployment. The main contribution is the adoption of a transformation function approach that allows for heterogeneous gender-specific slopes and incorporates demographic, institutional, and structural factors, such as childcare provision and female entrepreneurship, as shift determinants of the curve. Building on labour-market matching theory, the model accounts for the joint determination of male and female unemployment in relation to vacancies, thereby capturing potential substitution and complementarity effects across genders. The model is estimated using a System GMM approach to address endogeneity concerns and to account for the dynamic nature of the relationship.Preliminary results indicate a stable downward-sloping Beveridge curve across all specifications, with no robust evidence of systematic gender differences in its slope once full interactions are included. Labour market matching is significantly influenced by structural and institutional factors, particularly childcare provision. Both private and public childcare are associated with improved matching efficiency, although private childcare displays stronger and more persistent gender-specific effects in favour of women. Results for total childcare remain positive but with weaker gender differentials. By contrast, female entrepreneurship does not appear to exert a statistically significant effect.