In this paper, we use linked employer-employee administrative tax data from Canada to estimate the impact of payroll taxes on a variety of firms and workers outcomes. At the firm level, we use geographic and time variations in tax rates to identify the effect of payroll taxes on wage growth at the worker level. For one province, we exploit a clean overtime change in the payroll tax rate to estimate its impact on the firm's level of employment, average wage and productivity, with difference-in-differences models, taking into account firm-level unobserved heterogeneity. Additionally, taking advantage of the nature of linked data, we estimate wage equations with both fixed worker and firm fixed effects. We find no impact on employment, productivity and profits, but significant impacts on wages, implying that payroll taxes are passed almost entirely to workers in the form of lower wages.