" We study whether there is increased reliance on interest arbitration, that is, a narcotic or addictive effect or, alternatively, positive state dependence, in public sector contract settlements. We use contract data from three sectors (police, firefighters and hospitals) in the Canadian province of Ontario, which covers 1981 to 2012. The length of our study period yields much longer bargaining histories than previously used, which should provide more compelling evidence on whether there is increased reliance on interest arbitration to settle bargaining impasses over time. We obtain our estimates using a dynamic probit model with random effects that models the initial conditions. Our estimates indicate — across all the sectors we consider and some robustness checks — that there is a narcotic effect in interest arbitration usage despite very different average propensities to use arbitration across sectors."

Support: Digital