Affirmative Action & Gender Equity: New Rules Under Title IX?

Civil Rights Practice Group Newsletter - Volume 3, Issue 1, Spring 1999

May 1, 1999

Jennifer C. Braceras

Although the plain language of Title IX states unequivocally that schools are not required to grant "preferential or disparate treatment" to members of one sex in order to rectify a gender imbalance in an educational program or activity, the Department of Education regulations implementing Title IX have indeed created a vast and rigid quota system. A proposed common rule recently floated by the Clinton administration threatens to further expand the grip of quotas on our nation's educational institutions and on any and all institutions which receive federal funds and conduct any educational programs or activities.

There are two ways that the proposed common rule encourages, and, indeed, even mandates, gender-based quotas. First, the regulations contemplate the imposition of mandatory affirmative action programs by the enforcing agency as a remedy for discrimination and encourage recipients of federal funds to adopt voluntary affirmative action programs even where there has been no prior discrimination. Second, under Title IX, the unequal representation of women (or men) in a particular program or activity may be "consider[ed] in any hearing or proceeding under this Chapter." Significantly, however, the regulations implementing Title IX eviscerate well-accepted legal standards governing the burdens of proof in "disparate impact" cases. Because the regulations make it almost impossible for a recipient to justify the under-representation of one sex in a particular program or activity, recipients will naturally adopt quotas in order to head off such complaints in the first instance.


A. Affirmative Action

Subpart A of the proposed common rule deals directly with remedial and affirmative action. It provides that, where an agency makes a finding that a recipient has discriminated on the basis of sex, "such recipient shall take such remedial action as the designated agency official deems necessary to overcome the effects of such discrimination." (Section ___.110(a)) (emphasis supplied). Significantly, this provision not only grants the federal government the power to order a recipient to terminate the discriminatory practice in question (in order to level the playing field), but it also provides a license for federal agencies to mandate quotas in order to eradicate any statistical disparity between the sexes which the agency deems "attributable" to the previous unlawful policy.

In practice, this means, for example, that if an administering agency finds that a college has been unlawfully limiting enrollment in certain courses (i.e., feminist studies courses) to members of one sex (i.e., women), the agency can not only order the college to open the courses to all students, but can require that the school adopt an affirmative action plan to ensure the proportional representation of the excluded sex in the courses which were previously (and unlawfully) off limits to that sex. This provision, thus, vests virtually unlimited power in the administrative agencies implementing Title IX to require quotas as a remedy for under-representation of one sex or another. The imposition of quotas by a government entity (such as a state college) would, of course, run afoul of the Equal Protection Clause of the United States Constitution. Cf. City of Richmond v. J. A. Croson, Co., 488 U.S. 469 (1989); Wessman v. Gittens, 160 F.3d 790, 796 (1st. Cir. 1998) (racial balancing and the desire to correct under-representation of a protected class do not justify race-conscious policies by state actors).

Likewise, Subpart A encourages recipients of federal funds voluntarily to "take affirmative action consistent with law to overcome the effects of conditions that resulted in limited participation therein by persons of a particular sex." (Section ____.110(b)) (emphasis supplied). This section is modified slightly from the current regulation by adding the phrase, "consistent with law." The supplementary information accompanying the regulations states that "given recent, numerous decisions by the Supreme Court and lower courts concerning affirmative action, agencies should consult with the Department of Justice regarding interpretation of this section." The phrase, "consistent with law" is, of course, purposely vague. Rather than codifying governing judicial decisions regarding affirmative action, the proposed common rule advises the enforcing agencies to defer to interpretations of law advanced by political bureaucrats.

Although Subpart C of the proposed common rule explicitly prohibits gender preferences and/or the imposition of numerical limits on the number or proportion of persons of either sex who may be admitted, the regulations do not prohibit (and, indeed, encourage) gender preferences and numerical limits with respect to particular programs and activities.


B. Disparate Impact

1. Tests and Other Admission or Hiring Criteria:

Two of the most troubling aspects of the proposed common rule are found in Subpart C, Section ____.300(b)(2) and Subpart E, Section _____.505, which prohibit the use of tests or other criteria with a disparate impact on a particular sex. Specifically, Subpart C, Section _____.300(b)(2) states:

A recipient shall not administer or operate any test or other criterion for admission that has a disproportionately adverse effect on persons on the basis of sex unless the use of such test or criterion is shown to predict validly success in the education program or activity in question and alternative tests or criteria that do not have such a disproportionately adverse effect are shown to be unavailable. (emphasis supplied).

Likewise, Subpart E, Section ____.505, which applies to employment standards of recipients of federal funds, prohibits the use of employment tests or criteria that adversely impact one sex unless the test or criteria can be proven to "predict validly" successful performance in the particular job at issue and the recipient can demonstrate that alternate tests, without a disparate impact, are "unavailable."

These provisions are blatantly at odds with federal law regarding "disparate impact" analysis under federal employment discrimination statutes. Under Title VII, a plaintiff attempting to demonstrate that an employer's policies have an unlawful disparate impact on a protected class must first demonstrate that a particular practice has caused a disparity. If the employer can demonstrate that the disparity was caused by a factor other than the challenged employment practice (i.e., by free choice), the claim of discrimination will fail, and the claim will not proceed to the next stage. 42 U.S.C. § 2000e 2(k)(i)(B)(ii). If, however, the plaintiff is able to raise an inference that the particular practice is the cause of the disparity, the employer is then required to demonstrate that the challenged practice is "job related" and "consistent with business necessity." 42 U.S.C. § 2000e 2(k)(i)(A). A plaintiff may still prevail if he can demonstrate that other tests which do not have a disparate impact would serve the employer's business interest just as well as the challenged practice, yet the employer refuses to adopt them. This standard was crafted to prevent employers from having to choose between abandoning legitimate selection criteria and adopting quotas in order to avoid liability. By contrast, under the regulations implementing Title IX, a recipient of federal funds is excused from implementing quotas only if it can prove that the challenged selection practice accurately predicts an applicant's successful performance and there are no other tests available.

To illustrate the difference in the two standards for imposing liability, consider the following example: An elite law school is hiring tenure track faculty members to teach first year students. The law school will only consider those candidates who, while in law school, were members of the law review and who, after graduating from law school, clerked on the Supreme Court of the United States. Although these criteria are certainly "job related" (both criteria indicate that an applicant has an academic and scholarly interest in the law), they may not "predict validly" an applicant's success at teaching first year students. Indeed, a more accurate predictor of teaching skills might be candidates' level of experience in public speaking — yet certainly no law school would base its hiring decisions on such a variable.

Likewise, suppose a law school, in selecting students for its incoming first year class, chooses to use as one of its admission criteria the applicant's commitment to public service. Although this criteria certainly will not "predict validly" whether applicants will succeed academically at the law school, it is arguably "related" to the mission of the school, which is to train attorneys who will uphold the law and advance justice.

Moreover, while a plaintiff in an employment discrimination case brought pursuant to Title VII has the burden of proving that other tests or criteria, without a similarly undesirable disparate impact, would equally serve the employer's business interest, Title IX requires recipients to prove that no other test or criterion is available. It is, of course, axiomatic that it is often impossible to "prove a negative" such as this (no other available criteria). There will almost always be some other test or criteria out there, even if it is not as accurate or more expensive to administer than the challenged practice and even if it fails to screen for the attribute deemed most important by the recipient. The regulations make no allowances for such concerns.

Take, for example, the 1996 case in which the ACLU and FairTest, an anti-standardized testing group, filed a complaint with the Department of Education's Office of Civil Rights alleging gender bias in the Preliminary Scholastic Aptitude Test (the PSAT). The complaint charged that the PSAT discriminated against girls in violation of Title IX because the 15th percentile of test-takers is generally made up of approximately 60% boys and 40% girls. Because ranking in the 15th percentile is one of the requirements for National Merit Scholarships, plaintiffs argued that the test discriminatorily denied girls millions of dollars in scholarships. Although the questions on the PSAT are certainly "related" to the test-takers' educational potential and ability, it is unlikely that the PSAT could have proven that the test "predicts validly" student success at college (particularly when "success" in college depends not only on which college a student chooses to attend, but also on which courses the student takes, and how much time the student spends studying). Moreover, it would certainly have been impossible for the PSAT to prove that other test questions, that do not have a disparate impact on girls, were "unavailable." Thus, under the threat of losing federal financial assistance, the PSAT settled the case by agreeing to alter the exam by adding thirty-nine questions on which girls generally score higher than boys. The remedy for a test that was allegedly biased against girls, then, was to add questions that are biased against boys.

The disparate impact standard articulated by the proposed common rule, therefore, unacceptably infringes on the educational and business prerogatives of recipients in an attempt to alter current federal anti-discrimination law.


2. Recruiting:

Also of concern is Subpart C, Section _____.310 which prohibits recipients from "recruit[ing] primarily or exclusively at educational institutions, schools, or entities that admit as students only or predominantly members of one sex, if such actions have the effect of discriminating on the basis of sex. . . ." Thus, a public service agency that conducts basic health classes for the poor could not recruit primarily at nursing schools or schools of social work if the majority of students at such schools are women. Likewise, under this section, a federally funded day care center could not recruit teachers primarily from all female colleges that specialize in early childhood education degrees.


3. Course Offerings:

Subpart D addresses nondiscrimination in educational program or activities, but — unlike Subpart C (which deals with admissions), does not contain any language prohibiting quotas. Specific areas covered in this subpart include "access" to course offerings, athletics, and other educational programs. Significantly, the term "access" remains undefined. The proposed regulations fail to state clearly that a statistical disparity in the numbers of men and women in different courses, majors, or athletics, would not be enough to demonstrate that "access" to such programs and activities has been limited on the basis of sex.

Where a recipient finds that a particular class contains a substantially disproportionate number of individuals of one sex, Subpart D, Section ____.425(c) requires that the recipient "take such action as is necessary to assure itself that such disproportion is not the result of discrimination on the basis of sex in counseling or appraisal materials or by counselors." It is, not surprisingly, unclear how a recipient should go about "assuring itself" that discrimination was not the cause of the disparity. For this reason, Subpart D, Section ____.425(c) should be eliminated.


The regulations also state that: "[w]here use of a single standard of measuring skill or progress in a physical education class has an adverse effect on members of one sex, the recipient shall use appropriate standards that do not have such effect." Subpart D, Section _____.415


4. Athletics:

With respect to participation in athletics, Subpart D, Section ___.450 requires that the opportunities to participate at each level of competition be made available on a nondiscriminatory basis, and that there be no discrimination on the basis of sex in the treatment of athletes competing within each level. This subsection sets forth a (non-exclusive) list of factors to consider in determining whether discrimination exists in an athletic program or activity, including, "whether the selection of sports and levels of competition effectively accommodate the interests and abilities of members of both sexes.

Although the regulation implies that the relevant inquiry is whether men and women who want to participate in competitive sports have the opportunity to do so, cases interpreting the regulations (and the Department of Education's non-binding "Policy Guidance" on the subject) have held that the participation ratio in athletic programs is to be based upon the number of males and females enrolled at an education institution at large. The case law has made no exception for differences in interest, ability, and choice. See Cohen v. Brown Univ., 991 F.2d. 888 (1st Cir. 1993), cert. denied, 520 U.S. 1186 (1997). Thus, under current case law, if a college's student body is 54% female, its athletes must also be 54% female, even if the pool of students who want to participate in competitive sports is only 30% female. As a result, colleges across the country have been eliminating systematically male sports teams in order to achieve gender parity on the playing fields.

This proportional to enrollment test is, of course, difficult to reconcile not only with the text of Title IX and the Athletic regulation itself, but also with prevailing non-discrimination standards. Indeed, in the context of employment discrimination claims under Title VII, the courts have recognized that a difference between the proportion of male and female workers in a particular position and their position in the general population is not proof of discrimination. See International Brotherhood of Teamsters v. United States, 431 U.S. 324, 340 n.20 (1977) (Title VII "imposes no requirement that a work force mirror the general population"); Wards Cove Packing Co. v. Atonio, 490 U.S. 642 (1989) (the relevant statistical comparison in the Title VII context is between the composition of qualified persons in the labor marker and the persons holding the jobs at issue); see also Piva v. Xerox Corp., 654 F.2d 591, 596-97 (9th Cir. 1981) (disparity between percentage of employer's regional sales representatives who were women and percentage of women in the general workforce not highly probative); EEOC v. United Virginia Bank/Seaboard National, 615 F.2d 147 (4th Cir. 1980).

Jennfier Cabranes Braceras is an attorney at Ropes & Gray.