Using a large-scale linked-employer-employee data set from western Germany, this paper presents new evidence on the wage premium of collective bargaining contracts. In contrast to previous studies, we seek to assess the extent to which differences in wages between workers in covered and uncovered firms arise from the non-random selection of workers and firms into collective bargaining coverage. By measuring the relative wage changes of workers employed in firms that change contract status, we obtain estimates that depart considerably from previous results relying on cross-sectional data. Results from analysing separate transitions show that leaving industry-level contracts is associated with subsequent wage losses. However, the results from a trendadjusted difference-in-difference approach indicate that the particularly the transitions to no-coverage appear to be associated with negative shocks. Overall, our findings provide no evidence of a 'true' wage effect of leaving wage bargaining, once differences in pre-transition wage growth are accounted for.