Full bibliography

Using data from a large Ontario steel mill, it is argued that grievance outcomes provide a useful proxy to measure bargaining power. Three independent variables are constructed: 1. the match of a complaint with a specific grievable category in the collective agreement (the complaint/grievance match), 2. the general grievance type, and 3. the work site (works division). Various hierarchical log-linear/logit models are fitted to estimate grievance outcomes. The model that best fits the data is one that includes the interaction between the complaint/grievance category match and grievance type as well as the interaction between the complaint/grievance category match and works division. The results show that under most conditions, bargaining outcomes favor management. Under a specific constellation of conditions, however, bargaining outcomes favor labor. Thus it is only under this specific set of conditions that labor has greater bargaining power than management.