Written Testimony of Fatima Goss Graves Vice President for Education and Employment National Women's Law Center

Meeting of February 16, 2011 - EEOC to Examine Treatment of Unemployed Job Seekers

My name is Fatima Goss Graves, and I am the Vice President for Education and Employment at the National Women's Law Center. Since 1972, the Center has been involved in virtually every major effort to secure and defend women's legal rights. I appreciate your invitation to testify before the Commission on such a profoundly important issue, employer restrictions on applications from unemployed job seekers and their impact on women in the workforce.

Businesses increasingly exclude unemployed job seekers from their applicant pools, either limiting applicants to those who are currently employed or those who have been employed within a recent period of time. Employers' use of such a requirement threatens to exacerbate long-term unemployment and undermines broader economic recovery. Moreover, creating barriers to employment for those who are out of the workforce could have a serious negative impact on women.

My remarks will begin by focusing on the effect of the recession and economic recovery on women workers, especially in nontraditional fields where women were already underrepresented. I will then discuss how restricting job applications from unemployed job seekers may have a disparate impact on women, particularly women in nontraditional fields, certain categories of older women, women of color, and women returning to the workforce after caregiving. I also will discuss why the practice of excluding the unemployed from applicant pools is not job-related and cannot be justified under Title VII by a "business necessity" rationale.

I. THE RECESSION HURT WOMEN WORKERS AND THE ECONOMIC RECOVERY HAS LEFT WOMEN BEHIND.

An understanding of the role of women in the workforce before the recession and how women have fared during the most recent downturn and economic recovery is essential background for understanding the way in which excluding the unemployed will impact women. In 2007, just before the recession began, women were 46.2 percent of all workers.¹ But that relative parity between men and women workers in the aggregate masked significant disparities in women's employment in certain occupations. Some of those occupations offered relatively high pay for workers with relatively low educational levels, and so provided particularly desirable jobs. To give you some specific examples, in construction and extraction positions— which include positions like construction laborers and carpenters—women made up just 2.2 percent of workers in 2007, an exceedingly small share.² Although these women workers were paid less on average than men in the same occupations in 2007, making 88.4 percent of men's wages, they still enjoyed a median weekly wage of $573 ³ in occupations where nearly 70 percent of all workers had a high school degree or less.⁴ Likewise, among protective service occupations—a category into which security guards fall—women composed 20.5 percent of workers and made only 78.0 percent of men's wages.⁵ However, they still earned $588 in median weekly wages.⁶ The occupations I've just described have two things in common: They offered women an opportunity to make decent wages with relatively modest levels of education and positions in these occupations disproportionately went to male workers.

Thus, as women struggled to gain a foothold in certain occupations where they were historically underrepresented, we fast forward to the recession: The U.S. economy shed nearly 7.5 million jobs, entire industries disappeared, and failing businesses made painful layoffs.⁷ Men lost seven of every ten jobs between December 2007 and June 2009, while women lost only three during the same period,⁸ leading some to call the most recent recession a "mancession."⁹

There is no doubt that the recession had a dramatic impact on male workers, and men of color in particular. Although the economic recovery that followed has benefited men overall, it has been noted pretty widely that black and Hispanic men have seen their unemployment rates rise during the recovery.¹⁰ What has received less attention, however, is that the recovery is also leaving women behind. Since the beginning of the recovery in July 2009 through January 2011, men gained 438,000 jobs but women actually lost 366,000 jobs.¹¹ And from January 2010 to January 2011, a period during which the recovery accelerated, women filled fewer than one in every twenty of the 984,000 jobs added, even though they lost roughly six in twenty during the recession.¹² With this lopsided recovery, it is thus not surprising that women's overall unemployment rate has actually risen during the economic recovery, while men's has declined.¹³

II. CURRENT OR RECENT EMPLOYMENT AS A JOB QUALIFICATION WILL HAVE AN ADVERSE IMPACT ON WOMEN.

The exclusion of job applicants who are not currently or were not recently employed may violate federal employment laws.¹⁴ Title VII does not identify the unemployed as a protected category of workers, but it does prohibit employers from using a practice that disproportionately excludes people based on their sex, race, national origin, or religion unless the employer has a business need for the practice – a standard that would be difficult for an employer to meet without evidence supporting the validity of a practice in predicting employee performance. And even if the employer can identify a business need, it must explore a less discriminatory alternative to the practice.¹⁵ Likewise, the Age Discrimination in Employment Act (ADEA) prohibits the use of a selection device in hiring that has a disparate impact on workers aged 40 and over unless the employer adopts the device based on a "reasonable factor[] other than age."¹⁶

There are at least three reasons that employers' exclusion of the unemployed from applicant pools may have an adverse impact on women. First, when an employer restricts applicants to those who have been employed at some point within a recent period of time, such as the last three or six months, the restriction is likely to have a disparate impact on certain categories of older women and women of color. Unemployed women were more likely than men to be long-term unemployed for 27 weeks or more in January 2011.¹⁷ And certain categories of older women are especially disadvantaged. Indeed, the age group with the largest gender disparity is workers aged 45 to 54. Among unemployed workers in this age range, women were unemployed more than two months longer than their male counterparts.¹⁸ Additionally, women aged 55 to 64 were unemployed three weeks longer than men of the same age during this time period.¹⁹ The difference in length of unemployment is also particularly pronounced for women of color. For example, black women were unemployed more than two months, 8.5 weeks, longer than white males in January 2011.²⁰

Second, excluding the unemployed from applicant pools may have a disparate impact on women in occupations for which women experience greater unemployment than men, many of which are nontraditional fields for women. Although the overall unemployment rate is lower for women than men,²¹ women's unemployment rate exceeds men's in several fields that are nontraditional for women, including construction or extraction; protective service occupations; production occupations; and architecture and engineering.²² As noted above, the share of women workers in some of these and other nontraditional fields for women was already very small before the recession. Thus, even in occupations that may have contracted during the recession, such as construction, it is critical that as those sectors recover, women do not encounter yet another obstacle to employment in fields that can offer high wages.

Third, employers that restrict applicant pools to those who are already employed risk imposing a disparate impact on caregivers, some of whom leave the workforce temporarily to care for young children, ailing parents, or disabled family members. Although women now make up nearly half of the workforce, mothers in households with infants, for example, are far less likely than coupled fathers with infants in the household to be employed.²³ So, based on pre-recession data about middle socioeconomic status individuals, 98.1 percent of coupled fathers of infants were employed, whereas 46.0 percent of coupled mothers and 66.4 percent of single mothers were.²⁴ These women spent substantially more time than coupled men providing childcare, and they "significantly scaled back working time" in order to "finance[] the additional time they need[ed] for childcare."²⁵ Moreover, women account for 66 percent of caregivers for elderly or disabled family members.²⁶ A greater share of these female caregivers than male caregivers report that they have given up working entirely as a result of their caregiving responsibilities, 12 percent versus 3 percent respectively.²⁷

Caregivers who have not actively sought employment in the previous month are not reflected in unemployment statistics.²⁸ Nevertheless, like their formally "unemployed" counterparts, when female caregivers do decide to return to the workforce, they will be shut out of positions open only to individuals with current employment.

III. HIRING RESTRICTIONS EXCLUDING THE UNEMPLOYED ARE NEITHER JOB-RELATED NOR REQUIRED BY BUSINESS NECESSITY.

Under Title VII, an employer can defend against a disparate impact claim by demonstrating that a "challenged [employment] practice," in this case restricting applicants to individuals who are currently or recently employed, "is job related for the position in question and consistent with business necessity."²⁹ The statutory language makes clear that the "business necessity" defense involves a fact-intensive inquiry tethered to a specific job.³⁰

I am not aware of any evidence demonstrating that the exclusion of unemployed applicants is job-related for particular occupations, nor can I even imagine a rationale that would be consistent with Title VII. Instead, the exclusion of the unemployed from applicant pools is comparable to job selection devices that courts routinely conclude violate Title VII. For example, in the past, security and related occupations often imposed height or weight requirements on job applicants, which had the effect of locking most women, who are on average smaller than men, out of these positions. Court after court has held such requirements unrelated to the job in question and not warranted by business necessity.³¹

Like height, weight, and other unrelated selection devices,³² the exclusion of unemployed applicants seems unrelated to bona fide job qualifications. Media reports suggest that employers assume the practice will winnow the applicant pool to a manageable size, leaving those who are most qualified or desirable as workers.³³ There are a number of problems with this rationale. First, it is widely recognized that employers now face one of the most qualified applicant pools in recent memory because of high unemployment rates and sluggish hiring.³⁴ And even assuming that employers who underwent only limited layoffs retained their most valuable employees, many individuals who are not employed lost their jobs as part of mass layoffs or business closures, or left the workforce temporarily to become caregivers, all entirely unrelated to the employee's work performance. Second, these unsupported generalizations about the unemployed are reminiscent of—and are likely to exacerbate—stereotypes about caregivers, i.e., the unfounded assumptions that caregivers, especially mothers, are less competent or committed to their employers.

Finally, to the extent that limiting job applicants to those who are currently or recently employed helps employers winnow the applicant pool, employers could adopt alternative practices that are standard for many businesses—such as screening by a human resources specialist or calling references—that would have the same effect without having a disparate impact on women.

IV. EEOC GUIDANCE AND ENFORCEMENT ARE NEEDED.

The use of an individual's current or recent unemployment status as a hiring selection device is a troubling development in the labor market. As noted earlier, I have serious policy concerns about the use of this practice, especially during a time of such high, sustained unemployment. The deleterious effect of this practice on unemployed workers' economic security may well act as a negative counterweight to the positive work that government at all levels is doing to get people back to work. To that extent, I think this practice warrants further scrutiny from legislators and other policymakers.

However, even setting aside these policy concerns, the fact remains that restricting applications from unemployed job seekers already runs afoul of Title VII where it has a disparate impact on women and other protected groups. As discussed earlier, this practice likely does have such an impact on women in nontraditional fields where women's unemployment rates are higher than men's, certain categories of older women, women of color, and female caregivers. I urge the EEOC to provide guidance to the corporate community, making clear that discrimination against the unemployed may violate Title VII under a disparate impact theory. I also urge the EEOC to enforce Title VII vigorously when employers restrict the hiring of unemployed workers in a way that disparately impacts women and other protected groups.

Again, thank you very much for the opportunity to testify this afternoon, and I look forward to your questions.

Footnotes

¹ NWLC calculations from U.S. Dep't of Labor, Bureau of Labor Statistics, Labor Force Statistics from the Current Population Survey, Table A-1: Employment status of the civilian population by sex and age, seasonally adjusted, http://www.bls.gov/news.release/empsit.nr0.htm (last visited Feb. 11, 2011).

² U.S. DEP'T OF LABOR, WOMEN IN THE LABOR FORCE: A DATABOOK 55-67 (2008), Table 18, available at http://www.bls.gov/cps/wlf-databook2008.htm.

³Id.

⁴ U.S. Dep't of Labor, Bureau of Labor Statistics, Table 3: Percent employed in major occupational groups, by educational attainment and education or training category, 2008, https://www.bls.gov/cps/tables.htm (last visited Feb. 8, 2011).

⁵ WOMEN IN THE LABOR FORCE 55-67, supra n.2.

⁶Id.

⁷ NWLC calculations from U.S. Dep't of Labor, Bureau of Labor Statistics, Labor Force Statistics from the Current Employment Statistics Survey, Table B-5: Employment of women on nonfarm payrolls by industry sector, seasonally adjusted, http://bls.gov/ces/cesbtabs.htm (last visited Feb. 8, 2011).

⁸Id.

⁹See, e.g., Michael Gerson, Recession Takes a Toll on Men's Role as Providers, WASH. POST, Jan. 20, 2010, available at http://www.washingtonpost.com/wpdyn/content/article/2010/01/19/ AR2010011903468.html; Catherine Rampell, The Mancession, ECONOMIX, Aug. 10, 2009, http://economix.blogs.nytimes.com/2009/08/10/the-mancession; see also What Mancession?, NEWSWEEK, July 15, 2009, available at http://www.newsweek.com/2009/07/15/what-mancession.html (noting that the term "mancession" had 13,500 hits on Google in 2009).

¹⁰ NWLC calculations from U.S. Dep't of Labor, Bureau of Labor Statistics, Labor Force Statistics from the Current Population Survey, Table A-2, Employment status of the civilian population by race, sex, and age, and Table A-3, Employment status of the Hispanic or Latino population by sex and age, http://www.bls.gov/news.release/empsit.nr0.htm (last visited Feb. 11, 2011).

¹¹ NWLC calculations from U.S. Dep't of Labor, Bureau of Labor Statistics, Labor Force Statistics from the Current Employment Statistics Survey, Table B-5, supra n.7.

¹²Id.

¹³ NWLC calculations from U.S. Dep't of Labor, Bureau of Labor Statistics, Labor Force Statistics from the Current Population Survey, Table A-1, supra n.1.

¹⁴ There has been some legislative movement at the state level to prohibit such discrimination. Last year the New Jersey Assembly and Senate passed a bill that would fine employers who post job advertisements excluding unemployed individuals from the applicant pool. See Bills 2010-2011, A3359, http://www.njleg.state.nj.us/bills/BillView.asp?BillNumber=A3359 (last visited Feb. 10, 2011); see also Laura Bassett, Proposed New Jersey Law Would Penalize Employers for Discriminating Against Unemployed, HUFFINGTON POST, Oct. 13, 2010, http://www.huffingtonpost.com/2010/10/13/new-law­proposed-in-new-j_n_761570.html (last visited Feb. 8, 2011) (describing the background of the legislation). In January 2011, the New Jersey Governor conditionally vetoed the bill based in part on the perceived impact it would have on businesses and sent it back to the Assembly with recommended changes for reconsideration. See Bills 2010-2011, supra.

¹⁵See 42 U.S.C. § 2000e-2(k); see also Griggs v. Duke Power Co., 401 U.S. 424, 431-32 (1971).

¹⁶ 29 U.S.C. § 623 (emphasis added); see Smith v. City of Jackson, 544 U.S. 228 (2005).

¹⁷ NWLC calculations from U.S. Dep't of Labor, Bureau of Labor Statistics, Labor Force Statistics from the Current Population Survey, Table A-36: Unemployed persons by age, sex, race, Hispanic or Latino ethnicity, marital status, and duration of unemployment, http://www.bls.gov/web/empsit/cpseea36.htm (last visited Feb. 8, 2011). Comparisons are for median duration of unemployment.

¹⁸Id.

¹⁹Id. Women aged 65 and older, however, were unemployed for much shorter median periods of time than men in the same age range. Id.

²⁰Id. However, black women's median period of unemployment was one week shorter than black men's, 26.2 weeks versus 27.2 weeks respectively. Id.

²¹ NWLC calculations from U.S. Dep't of Labor, Bureau of Labor Statistics, Labor Force Statistics from the Current Population Survey, Table A-1, supra n.1.

²² NWLC calculations from U.S. Dep't of Labor, Bureau of Labor Statistics, Labor Force Statistics from the Current Population Survey, Table A-30: Unemployed persons by occupation and sex, http://www.bls.gov/web/empsit/cpseea30.htm (last visited Feb. 8, 2011).

²³See Robert Drago, The Parenting of Infants: A Time-Use Study, MONTHLY LABOR REV. 33, 39 (Oct. 2009), available at http://www.bls.gov/opub/mlr/2009/10/art3full.pdf. The study did not report data for single fathers with infants in the household because of the small sample size of that group. Id. at 34.

²⁴Id. at 39.

²⁵Id. at 41.

²⁶ NATIONAL ALLIANCE FOR CAREGIVING & AARP, CAREGIVING IN THE U.S. 2009, at 4, available at http://www.caregiving.org/data/Caregiving_in_the_US_2009_full_report.pdf. The share of caregivers who are women reaches 96 percent for recipients of care who are veterans, and more than four in ten caregivers for veterans report that they stop working due to their caregiving responsibilities. NATIONAL ALLIANCE FOR CAREGIVING, CAREGIVERS OF VETERANS 2010, at 25, 39, available at https://www.caregiver.org/caregiver-statistics-demographics.

²⁷ CAREGIVING IN THE U.S. 2009, at 55, supra n.26.

²⁸See U.S. Dep't of Labor, Bureau of Labor Statistics, Economic News Release, Employment Situation Technical Note, http://www.bls.gov/news.release/empsit.tn.htm (last visited Feb. 8, 2011).

²⁹ 42 U.S.C. § 2000e-2(k).

³⁰See id.; see also, e.g., EEOC v. Dial Corp., 469 F.3d 735, 742 (8th Cir. 2006) (noting that an "employer using the business necessity defense must prove that the practice was related to the specific job and the required skills and physical requirements of the position"); Anderson v. Zubieta, 180 F.3d 329, 345-47 (D.C. Cir. 1999) (identifying the business necessity inquiry as a question of fact and analyzing the relevant employee positions and the employer's workforce as part of such an inquiry).

³¹See, e.g., Dothard v. Rawlinson, 433 U.S. 321, 331-32 (1977) (height and weight requirement for correctional counselors); see also Blake v. City of L.A., 595 F.2d 1367, 1377-81 (9th Cir. 1979) (holding that defendant had not shown for purpose of summary judgment that height requirement for jobs with Los Angeles Police Department was warranted by business necessity); cf. United States v. Virginia, 620 F.2d 1018, 1024 (4th Cir. 1980) (concluding that a height and weight requirement for state troopers had an adverse impact on women that was not justified by business necessity and so violated the Crime Control Act, to which Title VII principles apply).

³²See, e.g., Dial, 469 F.3d at 742-43 (upholding a district court's conclusion that a strength test for sausage factory worker applicants had a disparate impact on women and was not justified by business necessity); Harless v. Duck, 619 F.2d 611, 614, 616-17 (6th Cir. 1991) (holding that a police department's use of structured oral interviews had a disparate impact on women and was not job-related); Bouman v. Block, 940 F.2d 1211, 1227-28 (9th Cir. 1991) (holding that a sheriff department's written promotion examination violated Title VII because it had a disparate impact on women candidates and had no "legitimate business justification").

³³See, e.g., Dan Chapman, Outlook Poor for Long-Term Unemployed, ATLANTA J. CONST., Oct. 4, 2010, available at http://www.ajc.com/business/outlook-poor-for-long-657702.html; Laura Bassett, Disturbing Job Ads: 'The Unemployed Will Not Be Considered', HUFFINGTON POST, June 8, 2010, http://www.huffingtonpost.com/2010/06/04/disturbing-job-ads-the-un_n_600665.html.

³⁴See, e.g., Steve Matthews, Jobless Producing U.S. Profit on Higher Productivity, BUSINESS WEEK, June 28, 2010, available at https://www.bloomberg.com/news/articles/2010-06-27/jobless-produce-u-s-investor-profits-on-productivity-with-less-inflation (noting that "[t]he 6.8 million Americans out of work for 27 weeks or longer . . . are providing U.S. companies with an eager, skilled and cheap labor pool"); Alison Green, How the Recession Has Changed Hiring, USNEWS.COM, June 15, 2009, available at http://money.usnews.com/ money/blogs/outside-voices-careers/2009/6/15/how-the-recession-has-changed-hiring (noting that the recession has led to more "high-quality applicants").