Working Paper: NBER ID: w1624
Authors: George J. Borjas
Abstract: This paper investigates whether the parameters of labor demand functions are sensitive to alternative methods of estimation. The assumption that the production technology is of the Generalized Leontief type implies that the demand system can be estimated by analyzing cross-section differences in earnings across labor markets, by studying longitudinal changes in earnings within a labor market, or by investigating cross-section differences in labor force participation rates across labor markets. The estimation of these models on the 1970 and 1980 Public Use Samples from the U.S. Census reveals that the estimates of labor demand functions are indeed quite robust to major specification changes.
Keywords: No keywords provided
JEL Codes: No JEL codes provided
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| increase in female labor participation (J21) | decrease in earnings of white men (J79) |
| male immigrants (J11) | small negative effect on earnings of white native men (J79) |
| male immigrants and white males are substitutes in production (J79) | small negative effect on earnings of white native men (J79) |
| male immigrants and black males are complements (J79) | no negative impact on earnings of black men (J79) |
| increase in capital stock (E22) | increase in participation rates of labor inputs (J49) |