Working Paper: CEPR ID: DP1360
Authors: Alison L. Booth; Gylfi Zoega
Abstract: This paper looks at the effect of quitting on the number of workers trained under conditions of uncertainty about future productivity when workers have both firm-specific and industry-specific skills. A new effect is found which works in the opposite direction to the undertraining result of Stevens (1994, 1995): A high quit rate makes investment in training less irreversible in the presence of firing costs and hence also less risky. This effect makes firms start hiring new workers at a lower level of productivity and hire more workers for a given increase in productivity. A rise in the quit rate can now either decrease or increase the number of trained workers.
Keywords: quitting externalities; uncertainty; underinvestment
JEL Codes: E32; J23; J24; J41
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| higher quit rate (J26) | increase in number of workers trained (J24) |
| higher quit rate (J26) | reduced irreversibility of training investments (D25) |
| reduced irreversibility of training investments (D25) | increase in number of workers trained (J24) |
| higher quit rate (J26) | hiring new workers at lower productivity levels (J29) |
| hiring new workers at lower productivity levels (J29) | hiring more workers for a given increase in productivity (J24) |