Union wage effects: Does membership matter?

Using a matched employer-employee data set for Norway, we exploit rare information on the union status of both individual employees and their workplaces. We establish two key results. First, we find a positive effect of workplace trade union density on the level of the individual's pay in establishments covered by collective agreements. Second, we find that, conditioning on coverage, the individual union membership differential disappears after controlling for establishment-level union density. The union wage effect is therefore a pure public good, with individual membership conveying a positive wage externality.