The Impact of Bargaining Institutions on Employer-provided Training in Britain

Abstract: Using linked employer-employee data from the 1998 Workplace Employee Relations Survey, we investigate the impact of collective bargaining institutions on the incidence and intensity of formal training provided by employers to employees. Initially using only variables from the employee survey, we find that private sector men and full-time women covered by unions typically receive significantly more work-related training than their non-union counterparts, and that this impact is large. However part-time women in workplaces with union recognition are significantly less likely to be trained. When we add in various additional controls from the linked employer questionnaire, the private-sector union coverage effect for men in general remains significant but its magnitude is considerably reduced, while the effect for fulltime women is increased slightly. These findings suggest that the usual union coverage effect on training from individual-level surveys is biased, owing to omitted establishment-level characteristics. Finally, we estimate the impact on training incidence and intensity of three types of bargaining institution: the closed shop, the level at which pay bargaining takes place, and multiunionism. We find considerable heterogeneity across manual and non-manual, and male and female workers, in the impact of these institutions on training incidence and intensity. Unsurprisingly, the direct effect of union coverage on the male training probability is reduced when these additional union bargaining controls are included in the regressions.