"Anecdotal evidence suggests that trade unions succeed in ameliorating workplace health and safety, but no attempt has been made to link specific workplace injury rates with a respective union presence. Relying on WERS98, this paper establishes a cross-sectional link between trade unions and occupational injury rates, revealing that unions gravitate to accident-prone workplaces and react by reducing injury rates within these types of employment units. However, the ability for unions to reduce injury rates does not appear to increase monotonically as they progress along a workplace instrumentality continuum from recognition alone to a pre-entry closed shop."

Support: Digital