Should Workers Care about Firm Size?

The authors analyze how firms of different sizes reward measured skills and unmeasured ability. The empirical methodology, based on nonlinear instrumental variable estimation, permits direct estimation of the returns to unmeasured ability by firm size. An analysis of panel data from the Canadian Survey of Labour and Income Dynamics for two periods, 1993–1998 and 1996–2001, reveals statistically significant differences between firms of different sizes. In particular, returns to unmeasured ability are higher in medium-sized firms than in either small firms or large firms. The authors find that the firm-size wage gap and the differential in returns to unmeasured ability between small and medium-sized firms is mainly explained by ability sorting. The fact that larger firms reward ability less than medium-sized firms is consistent with an explanation based on monitoring costs. When firms become “too large,†monitoring costs may prevent them from rewarding ability directly through wages.