We study the effects of salary caps for top public managers in Italy, focusing on the 2011 and 2014 reforms implemented during the post-crisis fiscal consolidation period. The first cap, set at e290,000 and later reduced to e240,000, imposed a 100% marginal tax rate above the threshold. Using linked employer–employee administrative data and an event-study difference-in-differences design, we estimate the causal impact on earnings and participation. The 2011 reform significantly compressed salaries at the top of the public-sector distribution and generated sizable fiscal savings. Intensive-margin responses are modest: the top-bracket elasticity of taxable net earnings is about 0.214, and the resulting efficiency loss for stayers is small, reflecting second-order Harberger-triangle distortions. In contrast, extensive responses are large: the reform significantly increased early retirement and switching to the private sector, with participation elasticities far exceeding the intensive elasticity. Welfare analysis shows that overall efficiency costs are driven primarily by these extensive margins.