Gender pay inequality in the federal government

The new OPM report says:

there has been a large reduction in Clerical employees and a large increase in Administrative employees. The percentage of females in Professional jobs increased significantly. In terms of education level, the percentage of employees and the percentage of females with a Bachelor Degree or higher increased significantly.

and

  • Over the study years (1992-2002-2012), the gender pay gap has dramatically shrunk from about 30 percent to 13 percent (for All White Collar) and to 11 percent (for GS only).
  • The differences in the distribution of males and females across occupational categories appear to explain much of the pay gap. This finding was reinforced by multivariate regression and decomposition analysis, which showed that 70 percent of the White Collar pay gap was explained by the factors used in our analysis and that the occupational factor accounted for 76 percent of the explained portion of the gap in 2012. A separate analysis of the GS population produced similar results—67 percent of the gap was explained in 2012, with 93 percent of that explained portion attributable to the occupational factor. While occupational distribution explains much of the pay gap, we are not ruling out the possibility that discriminatory influences played a role in occupational distribution.
  • The regression and decomposition analysis shows that the unexplained portion of the pay gap in 2012 was 30 percent of the total pay gap for All White Collar and 33 percent of the total pay gap for the GS (less than 4 percentage points). (This unexplained portion could be attributable to factors that may or may not be discriminatory that were not accounted for in our analysis (e.g., non-Federal work experience, personal obligations).)
  • The pay gap was smaller in younger age groups. Pay gaps at different ages may reflect the differences in occupational distribution at those ages.
  • In 2012, pay gaps were found at all education levels, almost all in the 8-10 percent range.
  • In 2012, for supervisors and managers, the average female salary was 95.6 percent of the average male salary; however, females made up only 36 percent of supervisors and managers. Among members of the Senior Executive Service (SES), the female salary percentage was 99.2 percent; however, females made up only 33 percent of SES members.
  • When we examined pay gaps by grade level for the GS population, we found that there was no significant gap between female and male salaries. However, more females were found in lower grades, which may be a reflection of differences in occupational distribution.
  • For GS employees, a discretionary authority to set pay for new hires above the step 1 minimum rate was used more frequently (on a percentage basis) for males than females in all 3 study years. Closer analysis revealed that these actions are most heavily used in three occupational categories that are male-dominated, which affected the overall usage rates.
  • For GS employees, females received out-of-cycle “quality” step increases for outstanding performance more frequently (on a percentage basis) than males in all 3 study years.
  • Starting salaries were lower for females than males, on average, in all 3 years—roughly 10 percent lower in 2012. When we analyzed White Collar starting salaries for the 37 more-specific occupational categories in 2012, we found that female starting salaries exceeded male starting salaries for 14 categories and were within 5 percent of male starting salaries for another 12 categories. Only 4 categories had pay gaps of more than 10 percent (no more than 12 percent). Differences in occupational distribution between males and females appear to explain much of the overall starting salary pay gap. When we examined GS starting salaries by GS grade level, we found that male and female average starting salaries were quite close in all three years.
  • Promotions were received more frequently (on a percentage basis) by females than males in all 3 study years. When we examined White Collar promotion rates for the 37 more-specific occupational categories in 2012, we found that the female promotion rate exceeded the male rate for 27 of 37 categories.