Working Paper: CEPR ID: DP2200
Authors: Alison L. Booth; Gylfi Zoega; Marco Francesconi
Abstract: We investigate two dimensions of investment in general human capital on-the-job: the number of workers trained and the intensity of training for each worker. In the benchmark case, we consider wage and training decisions made by firms in an imperfectly competitive labour market. The benchmark case generates two types of market failure: too few workers are trained, and the workers who are hired receive too little training. This is caused by the firms' discount rate exceeding the social discount rate, due to a 'quitting externality'. We show that the presence of labour unions can increase social welfare by increasing training intensity, while reducing welfare by lowering the number of workers trained. Using the British Household Panel Survey, we confirm the predictions of the model.
Keywords: general training; human capital; labour unions; quitting externalities
JEL Codes: J24; J31; J41
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Union presence (J45) | Training intensity (M53) |
| Union presence (J45) | Number of workers trained (J24) |
| Union presence (J45) | Worker outcomes (J28) |
| Union coverage (J51) | Training probability (C45) |
| Union coverage (J51) | Days of training (M53) |
| Union coverage (J51) | Wages (J31) |
| Training incidence (C59) | Wages (J31) |
| Union coverage (J51) | Wage growth (J31) |
| Union coverage (J51) | Quitting probability (J26) |