| Abstract: | Estimates of the impact of union membership on job satisfaction suffer from nonrandom self‐selection of employees into unions. In this paper, we circumvent this problem by examining the impact on satisfaction of collective bargaining representation, rather than of union membership. We use a two‐stage technique that controls for nonrandom selection of faculty into institutions, and apply that to a panel of faculty at repeatedly observed four‐year universities. We find that bargaining agreements increase satisfaction with compensation but reduce satisfaction with faculty workload. Bargaining has no statistically measurable impact on overall job satisfaction or on faculty's satisfaction with their authority to make decisions regarding their instructional duties. |
