By Giulia Bovini (London School of Economics & Political Science (LSE) – Department of Economics)
This paper studies life-cycle labour supply responses to lower social security wealth. An over-arching pension reform implemented in Italy in 1995 (the Dini pension reform) provides the setting for the analysis. While ushering the transition from a defined-benefit (DB) to a notional defined-contribution (NDC) scheme, it introduced discontinuities in the social security replacement rate based uniquely on years of qualifying retirement contributions accrued by the end of 1995. Within a dynamic difference-in-difference framework and drawing on the combination of workers’ contribution histories with matched employer-employee records, the paper compares labour supply trajectories of prime-age workers (aged 35 to 50 at the time of the reform) who were barely assigned to the less generous transitional mixed regime (treated) and who barely remained under the more generous fully DB regime (control), over an extended period of time (1990-2015). Preliminary findings suggest that treated workers adjust labour supply: in the post-reform period, their labour earnings are higher. This is explained by an increase in the probability of working as well as in the number of full-time days worked, while modest to null differences emerge with respect to days worked part-time and daily earnings. Days of inactivity decline as well and and the magnitude of all coefficients is increasing over time. A 1000 euros drop in expected yearly pension benefits is associated to up to 4 more days worked per year. A cohort-by-cohort analysis that compares treated and control workers born in the same year confirms that responses in most cases become larger as workers grow older. Women are found to be more responsive than men.

Source: SSRN