Work Experience as a Source of Specification Error in Earnings Models: Implications for Gender Wage Decompositions

This paper models the bias from using potential vs actual experience in log wage models. The nature of the problem is best viewed as specification error as opposed to classical errors-in-variables. We correct for the discrepancy between potential and actual work experience and create a predicted mea...

Full description

Saved in:
Bibliographic Details
Published in:Journal of population economics 2009-04, Vol.22 (2), p.463-499
Main Authors: Regan, Tracy L., Oaxaca, Ronald L.
Format: Article
Language:English
Subjects:
Bias
Datasets
Demography
Earnings
Economic models
Economic theory
Economics
Economics and Finance
Employment
Endowment effect
Error
Estimation bias
Gender
Gender bias
Gender discrimination
Gender equity
Gender wage gap
Head of household
Human capital
Labor Economics
Labor force
Labor market
Labor markets
Literature reviews
Longitudinal studies
Males
Modeling
Original Paper
Population Economics
Rates of return
Social Policy
Studies
U.S.A
Variables
Wage differential
Wages
Women
Work experience
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper models the bias from using potential vs actual experience in log wage models. The nature of the problem is best viewed as specification error as opposed to classical errors-in-variables. We correct for the discrepancy between potential and actual work experience and create a predicted measure of work experience. We use the 1979 National Longitudinal Survey of Youth and the Panel Study of Income Dynamics and extend our findings to the Integrated Public Use Microdata Sample. Our results suggest that potential experience biases the effects of schooling and the rates of return to labor market experience. Using such a measure in earnings models underestimates the explained portion of the male-female wage gap. We are able to separately identify the decomposition biases associated with incorrect experience measures and biased parameter estimates.
ISSN:0933-1433
1432-1475
DOI:10.1007/s00148-007-0180-5