Using high-frequency Italian administrative data, the author studies the heterogeneous effects of a reform raising the normal retirement age (NRA) from 60 years to 65 years for private-sector male employees. The analysis, based on a difference-in-differences (DD) method, shows that the NRA raise reduces pension benefit claims but does not lead to a one-to-one increase in the employment rate since workers also apply for more disability and unemployment benefits. Moreover, most of them simply retire without any benefit. The extent of the effects varies substantially across socio-economic groups, as individuals with poorer health, with lower occupational grades and lower pay levels are the most constrained by the reform, experiencing the highest delay in pension claims, increase in employment, and inactivity. All in all, this paper shows that raising the NRA could have unintended effects as it affects more negatively the most vulnerable in the labor market.