Working Paper: NBER ID: w2350
Authors: William T. Dickens; Kevin Lang
Abstract: We subject our dual labor market model to a goodness of test fit and compare the results with those obtained using a single equation model with a complex error structure. The dual labor market does an excellent job of predicting the wage distribution except for failing to explain bunching at $7.50 and $10.00 per hour. The null hypothesis that the model is correct cannot be rejected at the .05 level. In contrast, the wage distribution predicted by the single labor market model differs significantly from the observed distribution.
Keywords: dual labor market; wage distribution; goodness of fit; labor market segmentation
JEL Codes: J31; J41
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| dual labor market model (J29) | wage distributions (J31) |
| dual labor market model (J29) | predictive accuracy (C52) |
| single equation model (C20) | empirical wage distribution (J31) |
| dual labor market model (J29) | single equation model (C20) |