Recently the U.S. Department of Labor’s Office of Federal Contract Compliance Programs (OFCCP) has been issuing Notices of Violation to and subsequently settling cases with contractors based on the allegation that the contractors have been involved in “steering” their applicants or employees. In these cases the OFCCP alleges that the policy or practice of the contractor directed or steered one group of applicants (women) to take a certain targeted job and directed or steered another group of applicants (men) to take another targeted job. If the pay in one of these targeted jobs is higher than the other targeted job, then the OFCCP argues that the policy or practice has been discriminatory.
On November 4, 2013, the OFCCP announced that G&K Services Co. had agreed to settle a steering allegation. According to the OFCCP, women in this establishment were discriminated against when the company steered them to take lower-paying "light duty” laundry positions. At the same time, the company steered men to take higher-paid "heavy duty" jobs. G&K Services Co. agreed to pay $265,983 in back wages to 59 female workers who were allegedly steered into the lower paying jobs.
On September 18, 2014, the OFCCP announced that Hillshire Brands Co., formerly Sara Lee Food & Beverage, had agreed to settle another steering allegation. In this audit the OFCCP alleged that Hillshire steered men to take dumper/stacker jobs, while women were steered into taking biscuit assembler jobs. The OFCCP found that during the audit period, 98% of biscuit assembler positions went to women, and 99% of dumper/stacker positions went to men. The OFCCP argued that because there were fewer dumper/stacker positions available, Hillshire discriminated against men.
The existence of occupational gender segregation across some occupations is not a new phenomenon. Gender segregation in jobs is measured as the difference in the distributions of women and men across various jobs. If women and men are identically distributed across occupations, then gender segregation across occupations does not appear to exist. On the other hand, if women are concentrated in one set of occupations and men in another set of occupations, then gender segregation of occupations does appear to exist.
The Department of Labor statistics shows that, nationwide, there are substantial differences in the distribution of women and men in different occupations. For example, in the Production Sector, women were 76.1% of the sewing machine operator workforce; 72% of tailors, dressmakers, and sewers; and 60.1% of the laundry and dry-cleaning workforce. In contrast, women were only 4.6% of the welding, soldering, and brazing workforce; 5.1% of computer control programmers and operators; and 5.3% of machinists.1
Economists and sociologists have discussed extensively the causes of gender segregation in some occupations. Obviously, any employer policy or practice that contributes to gender segregation should be abandoned, and employers should make concerted efforts to provide equal opportunities across all jobs for all applicants regardless of their gender.
In the press release regarding the settlement for G&K Services Co., the OFCCP did not specify how the steering of women to “light duty” jobs actually happened. From the press release, however, it appears that both male and female applicants applied and were initially hired as general laborers. Among those hired, G&K subsequently assigned a proportionally larger number of women to "light duty" jobs and a proportionally larger number of men to "heavy duty" jobs. Did the female applicants want to take the "heavy duty" jobs, and the company denied them that right? One cannot tell from the press release.
Similarly, in the press release regarding the settlement for Hillshire Brands Co., the OFCCP did not specify how men were steered to take dumper/stacker jobs and how women were steered into taking biscuit assembler jobs. The press release also is silent on which job paid more. The press release only indicates that the biscuit assembler position had a higher number of hires, and that, therefore, by steering men to dumper/stacker jobs, Hillshire deprived men of more opportunities for employment. Again the question is whether the male applicants were really applying for biscuit assembler jobs and the company was ignoring those preferences, and instead steering the men to take jobs only in the dumper/stacker positions.
In an earlier article published here, I presented a hypothetical example in which a contractor in the Food Preparation and Serving industry hired both dishwashers and hosts. The contractor reviewed applications for 300 female applicants and 300 male applicants and hired 42 women and 98 men. On the surface, the crude statistical analysis showed that the contractor was hiring men at a statistically significant higher rate than women. The contractor hired men at a 43% rate (98/300) and hired women at only a 14% rate (42/300). Further analysis of the applications, however, showed that 240 women (majority) applied to host/hostess positions and only 60 women applied to the dishwasher position. In contrast, 240 men (majority) applied to dishwasher positions and only 60 men applied to host/hostess positions. Eventually, 17 women and 3 men were hired into the host/hostess position and 25 women and 95 men were hired into the dishwasher position. Even though the aggregated data showed statistically significant disparity between hiring rates of men and women, when the fact that men and women had different preferences for the two jobs was taken into account, statistically the hiring rate for women appeared to be slightly better than the hiring rate for men in both positions. The moral of that story is that even when employers are hiring applicants into entry-level general laborer type positions, they need to maintain hiring records at a less aggregated level, probably at each position level. The best practice is to tie the applications to each requisition number for each job vacancy so that the pools for each vacancy can be identified when the OFCCP conducts its audits, usually a few years after the original selection decisions were made.
Let’s see how the new OFCCP focus on steering issues impacts the above hypothetical example. In the above example, the OFCCP finds that on aggregate the hiring rate for men is higher than the hiring rate for women. The contractor, however, uses the information regarding which job each applicant was specifically applying for to show that the hiring rates, properly calculated, are not statistically significant different for men and women. With the steering allegation, the OFCCP has more opportunities to find problems with a contractor’s hiring practice. The fact that a higher proportion of women were hired into host/hostess positions and a higher proportion of men were hired into dishwasher positions opens the door for OFCCP to argue that the company was steering women into host/hostess positions and men into dishwasher positions.
Nationally, hosts/hostesses have higher earnings than dishwashers. The median weekly earning for dishwashers is $356 and the median weekly earning for hosts/hostesses is $393.2 Therefore, the OFCCP can argue that the steering was discriminatory toward men since it prevented them from getting a higher paying job.
In the example above, the OFCCP can also argue that women were discriminated against. Since 120 of the hires went to dishwashers and only 20 went to hosts/hostesses, the OFCCP can argue that by steering women to host/hostesses positions the contractor reduced women’s chance of finding employment.
As the above example illustrates, the new creative steering theory opens many new opportunities for the OFCCP to argue that a contractor’s hiring practice is not appropriate. It is up to the contractor to present the evidence to show that its policies and practices were neutral and that they did not steer the applicants toward any particular position.
How can the contractor show that their practice is neutral with respect to steering? The contractor needs to track and preserve the expression of interest of all applicants for specific positions. For example, in the above example, the contractor needs to announce the vacancies in dishwasher and host/hostess positions separately by issuing a separate requisition number for each position. These types of vacancies should not be announced under the general entry-level laborer job category. By separating these types of positions and by issuing separate requisition numbers for each vacancy, the contractor can collect and track the information on each vacancy announcement and also the information on the life cycle of each application which relates to each vacancy. This practice will enable the contractor to simultaneously track and record the employer’s decisions regarding each applicant as well as each applicant’s voluntary decisions regarding each position. If the contractor’s practices are neutral with respect to steering, they can effectively illustrate that neutrality after the voluntary decisions by the applicants are taken into account.
1. See: Employed persons by detailed occupation, sex, race, and Hispanic or Latino ethnicity – 2013, Current Population Survey, (http://www.bls.gov/cps/cpsaat11.htm). ↩
2. See: Median weekly earnings of full-time wage and salary workers by detailed occupation and sex -2013, Current Population Survey, (http://www.bls.gov/cps/cpsaat39.htm). ↩