We study the impact of a large payroll tax cut for older workers on employment and wages in Hungary. By exploiting administrative data and applying a difference-in-differences empirical strategy, we document a modest employment increase equivalent to a labor demand elasticity of -0.3 and a pass-through rate of 22%. These average effects mask large heterogeneity across different firms. Employment mainly increases at low-productivity, low-paying firms, while no jobs are created at high-productivity, high-paying firms. At the same time, the tax cut is passed through to wages only at high-productivity, high-paying firms, while low-productivity, low-paying firms do not share the benefits of the tax cut with their workers. These results point to important heterogeneity in the incidence of payroll tax cuts across firms, highlighting that workers at different firms benefit differently from payroll taxes. They also demonstrate that payroll taxes can have a significant impact on the composition of jobs in the labor market.