Is Earnings Nonresponse Ignorable?

24 Pages Posted: 23 Jan 2011

See all articles by Christopher R. Bollinger

Christopher R. Bollinger

University of Kentucky - Department of Economics

Barry T. Hirsch

Georgia State University; IZA Institute of Labor Economics

Multiple version iconThere are 3 versions of this paper

Date Written: December 19, 2010


Earnings nonresponse in the Current Population Survey is roughly 30% in the monthly surveys and 20% in the annual March survey. Even if nonresponse is random, severe bias attaches to wage equation coefficient estimates on attributes not matched in the earnings imputation hot deck. If nonresponse is ignorable, unbiased estimates can be achieved by omitting imputed earners, yet little is known about whether or not CPS nonresponse is ignorable. Using sample frame measures to identify selection, we find clear-cut evidence among men but limited evidence among women for negative selection into response. Wage equation slope coefficients are affected little by selection but because of intercept shifts, wages for men and to a lesser extent women are understated, as are gender wage gaps. Selection is less severe among household heads/co-heads than among other household members.

Keywords: response bias, imputation, earnings nonresponse, gender gap, CPS

JEL Classification: J31, C81

Suggested Citation

Bollinger, Christopher R. and Hirsch, Barry T., Is Earnings Nonresponse Ignorable? (December 19, 2010). RatSWD Working Paper No. 165, Available at SSRN: or

Christopher R. Bollinger (Contact Author)

University of Kentucky - Department of Economics ( email )

Lexington, KY 40506
United States

Barry T. Hirsch

Georgia State University ( email )

Department of Economics
Andrew Young School of Policy Studies
Atlanta, GA 30302-3992
United States
404-413-0880 (Phone)
404-413-0145 (Fax)


IZA Institute of Labor Economics

P.O. Box 7240
Bonn, D-53072

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Abstract Views
PlumX Metrics