Abstract
Estimates of union wage and productivity effects are derived using primary micro-level panel data for a sample of firms in the UK engineering industry. Union wage differentials of the order of 10% are suggested from the results, whereas union productivity impacts appear to be non-linear with respect to union density.