How Much Wage Variation Can We Explain?

  • Earnings functions are an important tool in labor economics as they allow to test a variety of labor market theories. Most empirical earnings functions research focuses on testing hypotheses about sign and magnitude for the variables of interest. In contrast, there is little attention for the explanation power of the econometric models employed. Measures for explanation power are of interest, however, for assessing how successful econometric models are in explaining the real world. Are researchers able to draw a complete picture of the determination of earnings or is there room for further theories leading to alternate econometric models? This article seeks to answer the question with a large microeconometric data set from Germany. Using linear regression estimated by OLS and R2 as well as adjusted R2 as measures for explanation power, the results show that up to 60 percent of wage variation can be explained using only observable variables.

Download full text files

Export metadata

Additional Services

Share in Twitter Search Google Scholar
Metadaten
Author:Andreas Tiltag
URN:urn:nbn:de:hbz:385-10919
Document Type:Working Paper
Language:German
Date of completion:2017/09/20
Release Date:2017/09/20
Tag:Labor economics; R2; earnings functions; theories of wage determination; wage variation
GND Keyword:Arbeitsmarkttheorie; Lohnunterschied; ├ľkonometrisches Modell
Comment:
DOI: https://doi.org/10.25353/UBTR-2161-5058-91XX
Institutes:Fachbereich 4 / Wirtschaftswissenschaften
Dewey Decimal Classification:3 Sozialwissenschaften / 33 Wirtschaft / 330 Wirtschaft
JEL-Classification:J Labor and Demographic Economics / J3 Wages, Compensation, and Labor Costs / J30 General
J Labor and Demographic Economics / J3 Wages, Compensation, and Labor Costs / J31 Wage Level and Structure; Wage Differentials
J Labor and Demographic Economics / J3 Wages, Compensation, and Labor Costs / J39 Other

$Rev: 13581 $