Using OLS and quantile regression methods and rich cross-section data sets for western and eastern Germany, this paper demonstrates that the impact of works council presence on labor productivity varies between manufacturing and services, between plants that are or are not covered by collective bargaining, and along the conditional distribution of labor productivity. No productivity effects of works councils are found for the service sector and in manufacturing plants not covered by collective bargaining. Besides demonstrating that it is important to look at evidence based on more than one data set, our empirical findings point to the efficacy of supplementing OLS with quantile regression estimates when investigating the behavior of heterogeneous plants.