I use the Polish anonymized tax data for 24 million individuals observed in the period 2004-2016 to analyse the employment effects of minimum wage. In contrast to most studies, the longitudinal dimension of the dataset allows me to control for unobserved characteristics of employees and to assess the long-term effects of minimum wage hikes. I find that minimum wage has a moderate impact on job separations in the long-run, while the short-term effects are negligible. Another important result is that young workers earning around the level of minimum wage have a significantly lower probability of returning to employment after a job loss than their peers from higher part of the income distribution. This effect has been in place since 2008, when there was a substantial increase in the minimum wage.