Quantile regression and structural change in the Italian wage equation

A test for structural break based on quantile regressions (QR) reveals the impact of a break in the tails of the conditional distribution, unveiling an opposite behavior in the tails that balances at the mean and that cannot be found using OLS. By repeatedly computing the QR test it is possible to check for stability of the coefficients both over time and with respect to dichotomies in the explanatory variables, such as generations or regions. This provides evidence of a break in earnings i) over time, with wages decreasing in the latest period; ii) across generations, where young workers experience a decrease in returns to education; and iii) across regions, with an opposite pattern of earnings, unstable at the lower quantiles in the southern regions and unstable at the higher wages in the north-center area.

► Quantile regressions analyze different points of the conditional distribution. ► A test analyzes parameter instability over time, across regions, and across generations. ► The instability is due to institutional changes introducing labor flexibility. ► The regional divide is a large source of discrepancy in the coefficient estimates. ► Pooling the two regions masks different behavior of the regional earning profiles.