This paper demonstrates empirically that the limited dependent variable technique for sample selectivity bias correction provides two very different estimates of the gap between union and non-union wages. Our findings, based on the Canadian Survey of Work History, suggest that selectivity...

This paper specifies an alternative model of union growth which builds on the strength of the popular A/P and b/E models and incorporates several hypotheses relating aggregate union membership growth to changes in key business cycle variables, the legislative framework, and the inter-industry...