Chart Detail

Dimensions of wage inequality, by gender, 1973–2011

Wage gap* Change
1973 1979 1989 1995 2000 2007 2011 1973–1979 1979–1989 1989–2000 2000–2007 2007–2011
A. Total wage inequality**
90/10 (x/y) Men 128.0% 130.0% 144.3% 151.1% 150.3% 155.3% 160.9% 2.0 14.3 6.0 5.0 5.7
Women 115.9 103.2 134.9 137.6 137.7 143.7 145.9 -12.7 31.8 2.8 6.0 2.1
90/50 Men 60.3 58.8 69.2 76.1 79.5 83.0 87.7 -1.5 10.4 10.2 3.6 4.7
Women 59.2 60.6 70.5 76.5 78.2 81.3 83.2 1.4 9.9 7.7 3.1 1.9
50/10 Men 67.6 71.1 75.1 75.0 70.8 72.3 73.3 3.5 3.9 -4.2 1.4 1.0
Women 56.7 42.5 64.4 61.1 59.5 62.4 62.7 -14.2 21.9 -4.9 2.9 0.2
B. Between-group inequality***
Education
College/high school Men 25.1% 20.2% 34.0% 37.1% 42.0% 44.1% 44.8% -4.9 13.8 8.0 2.1 0.7
Women 36.5 25.0 40.0 46.7 47.9 48.5 48.7 -11.5 15.0 7.9 0.6 0.2
High school/less than high school Men 22.3 22.0 22.1 26.5 26.0 25.2 28.7 -0.3 0.1 3.9 -0.7 3.4
Women 26.2 21.3 26.4 29.8 29.5 27.7 26.2 0.0 0.1 0.0 0.0 0.0
Experience****
Middle/young
(35 yrs/25)
Men 22.0% 21.5% 25.7% 26.9% 22.9% 24.3% 27.5% -0.5 4.1 -2.8 1.4 3.2
Women 8.0 9.5 17.8 21.7 18.4 20.9 22.4 1.5 8.3 0.6 2.5 1.5
Old/middle
(50 yrs/35)
Men 3.4 8.2 12.4 12.7 8.8 9.4 11.8 4.7 4.3 -3.7 0.7 2.4
Women -2.0 0.4 2.1 5.3 4.7 8.3 8.6 2.4 1.7 2.5 3.7 0.3
C. Within-group inequality*****
Men 42.3% 42.8% 46.7% 47.8% 48.1% 50.1% 50.7% 1.4% 9.0% 3.0% 4.2% 1.1%
Women 41.8 40.2 44.7 46.7 45.8 48.4 48.5 -3.8 11.4 2.4 5.7 0.2

* Log wage differential

** Log wage ratio of x/y

*** Simple human capital regression of log wages; see table notes

**** Ratio x/y

***** Mean square error from same regressions as education and experience

Source: Authors' analysis of Current Population Survey Outgoing Rotation Group microdata

Updated May 14, 2012

Documentation and methodology

All of the data are based on analyses of the CPS-ORG data described in Appendix B and used in various tables. The measures of “total wage inequality” are natural logs of wage ratios (multiplied by 100) computed from Tables 4.5 and 4.6. The exception is 1979 data for women, which are 1978–1980 averages; we use these to smooth the volatility of the series, especially at the 10th percentile. The “between-group inequalities” are computed from regressions of the log of hourly wages on education categorical variables (advanced, college only, some college, less than high school with high school omitted), experience as a quartic, marital status, race, and region (4). The college/high school and high school/less-than-high-school premiums are simply the coefficient on “college” and “less than high school” (expressed as the advantage of “high school” over “less than high school” wages). The experience differentials are the differences in the value of age (calculated from the coefficients of the quartic specification) evaluated at 25, 35, and 50 years old. “Within-group wage inequality” is measured as the root mean square error from the same log wage regressions used to compute age and education differentials.

Demographic

Get the data

Get the data behind the charts.

Explore Our Charts

Subject

Demographic

Select a subject or demographic

Interactive feature

Interactive feature promo image When income grows, who gains?

For the media

Economic Policy Institute Media Relations Department (202) 775–8810 |