SUPREME COURT ADDRESSES USE OF RACE IN ADMISSIONS
Much has been written about the potential impact the U.S. Supreme Court's split decisions in the two University of Michigan affirmative action cases will have on employment law issues, particularly on those involving private sector affirmative action programs. (Grutter v. Bollinger and Gratz v. Bollinger, decided June 23, 2003). The analysis used by the justices, however, does not apply to private sector affirmative action programs aimed at creating diversity in employment.
In Grutter, an applicant was denied admission to the University of Michigan Law School, which based its admissions decisions on undergraduate achievement, LSAT scores, work experience, leadership and services, and letters of recommendation. The law school considered whether the applicant was an underrepresented minority. In the companion case, Gratz, the University of Michigan's undergraduate College of Literature, Science and Arts rejected a candidate. The college assessed applicants based on a 150 point system in which 20 points could be awarded to applicants who were members of underrepresented minority groups (African-American, Hispanic, or Native American). The Supreme Court struck down the points-based undergraduate system because its use of race was not narrowly tailored to achieve the college's interest in diversity and, therefore, was unconstitutional. It upheld the law school's process, however, because it was narrowly tailored. The Supreme Court justices relied on cases defining the scope of permissible affirmative action in the context of public education and concluded that race (and gender) can under certain narrow circumstances be viewed as a "plus" factor in making individualized decisions as to whom should be admitted to a public university in order to promote educational diversity.
For affirmative action in the employment context, the most recent United States Supreme Court pronunciation on this issue remains a 1987 decision in Johnson v. Transportation Agency, Santa Clara County, CA, 480 U.S. 616 (1987). In Johnson, the Supreme Court held in a split decision that the county agency did not violate Title VII by taking a female employee's sex into account and promoting her over a male employee with a higher test score. The Court reasoned the action was permissible because the decision was made pursuant to an affirmative action plan directing that sex and race be considered to remedy under-representation of women and minorities in traditionally segregated job categories. The Court also found that the plan did not unnecessarily trammel rights of male employees or create an absolute bar to their advancement.
Many of the justices who upheld the county agency's affirmative action plan in the Johnson decision are no longer on the Court. Like reading tea leaves, predicting how the Supreme Court will rule on a future case involving an employment preference for a minority or woman based on an affirmative action plan, is a risky proposition, particularly given the possible changes in the composition of the Court in the near future. We believe it would be a very risky proposition for an employer to give a preference to a minority or woman based solely on the existence of determined "underutilization" in a particular "job group" for several reasons, particularly if the method of determining underutilization is the "any difference" or "whole person" test, as those terms are used in connection with affirmative action programs mandated by Executive Order 11246 and the implementing regulations enforced by the Department of Labor's Office of Federal Contract Compliance Programs ("OFCCP").
First, the "swing" vote in the 2003 University of Michigan Law School case was Justice O'Connor. The four justices who voted against the University of Michigan Law School almost certainly will vote against any such affirmative action in the employment context if faced, for example, with an unsuccessful white or male applicant or employee who brought a "reverse discrimination" claim where the employer acknowledged that the plaintiff male candidate or plaintiff white candidate would have been selected, if the successful candidate was not female or a minority.
Second, in the 1987 Johnson decision, Justice O'Connor concurred in upholding the employer's action based on its affirmative action plan, but made clear that the plan could only be justified because of a statistical disparity that was so great as to make out a prima facie case of discrimination against the employer. In other words, a determination by an employer that underutilization exists and that a goal is necessary based on a minor statistical difference or based on the "whole person" test, would be insufficient. Rather, the employer would need to show that the difference between the expected and actual number of minorities or women in the employer's workforce (given the percentage of minorities and women in the relevant labor market who possess the requisite skills to fill the position at issue) was sufficiently large to be statistically significant at the two standard deviation level or above.
Third, the 1987 Johnson decision involved a preference given to an employee seeking a promotion. The impact on the unsuccessful male candidate obviously was less than if the decision had involved hiring, in which case the unsuccessful candidate would have been without a job entirely.
Fourth, the governmental entity involved in the Michigan cases (the University itself) had determined that providing a preference for the sake of diversity was necessary to its educational mission. In the private sector affirmative action employment context, the governmental entity charged with enforcement of the Executive Order, the OFCCP, has made no such finding and in fact has stated in its regulation that "Placement goals do not provide the contractor with a justification to extend a preference to any individual, select an individual, or adversely affect an individual's employment status, on the basis of that person's race, color, religion, sex, or national origin. . . . Placement goals may not be used to supersede merit selection principles. Affirmative action programs prescribed by the regulations in this part do not require a contractor to . . . hire a less qualified person in preference to a more qualified one." See 41 C.F.R. § 60-2.16(e).
Lesson for Employers
It therefore continues to be the safest course to limit affirmative action efforts to: (1) outreach in order to ensure that women and minorities are drawn into the applicant pool for positions for which they are under-represented, (2) monitoring to ensure non-discriminatory treatment of female and minority candidates during the hiring process, (3) oversight to ensure that women and minorities, once hired, are appropriately and non-discriminatorily considered for pay increases and for special projects and growth opportunities so that they will have a fair chance for promotion and (4) review of decisions concerning the termination of women and minorities to ensure those decisions are made in a non-discriminatory fashion.
NEW PROPOSED GUIDELINES ON WHO IS AN "APPLICANT" ARE, ONCE AGAIN, DELAYED
Proposed guidelines from the four agencies, including the OFCCP and the EEOC ,charged with the task of proposing new guidance on the definition of "applicant" as that term is used in the Uniform Guidelines on Employee Selection Procedures have, once again, been delayed. The proposed guidelines on the definition of an applicant will not be expected until fall of 2003, at the earliest. The definition is critical to federal government contractors and subcontractors for purposes of the tracking and analysis they must do as part of their affirmative action programs (AAPs). Because of the advent of new technology and use of the Internet, the OFCCP's old approach to defining who is an applicant has become unworkable.
At a recent affirmative action seminar, Harold Bush, the Director of OFCCP's Division of Program Operations, suggested that, as a practical matter, until new guidelines on the definition of applicant are adopted, OFCCP compliance officers have been accepting, a "provisional" definition of applicant if used consistently by an employer. The provisional definition is as follows: an applicant is an individual (1) who is minimally qualified, (2) has been considered as part of a pool of candidates who have survived some initial screening prior to the formal interview process, and (3) is in the final pool from which the employer performs its impact ratio analysis comparing hires to applicants.
Lesson for Employers
The OFCCP appears to be suggesting that, at least temporarily, contractors and subcontractors will be permitted to exclude from the definition of applicant those candidates who they determine fail to meet minimal qualifications and, perhaps, those candidates who are excluded after telephone screens; provided, it appears, that the remaining pool of candidates, each of whom will be defined to be an "applicant," is larger than the group who will be interviewed.
Despite this last proviso from the OFCCP that the applicant group should be larger than the interviewee group, we have assisted clients recently in a number of audits where the employer's definition of applicant was limited to the group of candidates who were formally interviewed by the employer and the OFCCP did not find a violation. Taking such a narrow approach to the definition of applicant is, however, a risky proposition and consideration certainly should be given to using a broader approach, such as that suggested by Mr. Bush of the OFCCP. While the OFCCP may chose not to pursue the applicant definition issue at this point in time in a situation where the only violation found would concern the employer's definition of applicant, if the OFCCP intends to cite an employer for other violations, it may well take a more aggressive approach on the applicant issue if the employer has limited the definition of applicant to those who have been interviewed.
EEOC PROPOSES CHANGES TO INDIVIDUAL AND JOB CATEGORIES USED IN THE EMPLOYER INFORMATION REPORTS (EEO-1)
In June of 2003, the EEOC proposed changes to the EEO-1 form and is seeking comment on its proposal through August 11, 2003. The primary change in individual categories is that employees will be able to identify themselves both by their ethnicity and by their race or races. Employees will be asked to identify whether they are or are not Hispanic or Latino and, regardless, whether they are White, Black or African-American, Native Hawaiian or Other Pacific Islander, Asian, American Indian or Alaska Native, or two or more races (anyone who identifies him or herself as being Hispanic or Latino will automatically be classified in the "two or more races" category).
In addition, the EEOC has proposed changes to the EEO-1 job categories. The EEOC has proposed dividing the "officials and managers" job category into three subcategories: (1) executive/senior level officials and managers, (2) mid-level officials and managers and (3) lower-level officials and managers. The EEOC also has proposed changing the order of a number of the other categories and making minor adjustments to the titles of some of the other categories.
Q&B Comment
If these changes pass, it could become a major headache for employers, although the proposed changes are less burdensome than a possible change that has been floated earlier, which would have allowed employees to check off as many different race and ethnic categories as they wanted and thus would have created hundreds of different combinations of categories into which an employee could fall. However, to help ease the transition to a new EEO-1 reporting system, numerous software vendors, without doubt, will be rolling out new HRIS systems and patches to existing systems to address whatever changes ultimately are adopted by the EEOC. In addition, almost certainly, the EEOC will provide employers with an adequate grace period within which to revise their HRIS systems after the changes are finalized. In other words, it's too early to start panicking. Instead, employers should either comment on the proposals directly or ensure that their trade associations are filing appropriate comments.
OFCCP SCHEDULES 2,000 NEW COMPLIANCE REVIEWS BASED ON THE EO SURVEYS COMPLETED BY EMPLOYERS EARLIER THIS YEAR AND WILL FOCUS THOSE REVIEWS ON COMPENSATION
In January of 2000, 49,000 federal government contractors and subcontractors were sent EO Surveys to obtain information about their applicants and workforce that went far beyond the workforce profile snapshot contained in the EEO-1 form. In addition to the snapshot, the EO Survey required contractors and subcontractors to provide information about the prior year's applicant flow, hires, promotions, terminations and annual monetary compensation by gender and minority/non-minority status by EEO-1 category. Hundreds of thousands of hours were spent by employers preparing this information and providing it to the government. It now appears that none of the information will be used. Instead, the government will be relying on the 10,000 EO Surveys sent to selected federal government contractors and subcontractors in January of 2003.
Two thousand of those 10,000 contractors who responded to the EO Survey earlier this year will be selected randomly for audit by the OFCCP during the balance of the year. Many have already been selected and received desk audit notices. In an "Alice through the looking glass"-like approach to determining whether the EO Survey can be used effectively as a predictor of possible discrimination, an outside consultant hired by the federal government will analyze the validity of the surveys as a predictor based on, for example, the actual dollars collected by the government from those contractors who were randomly selected for audit following their submission of the survey. While such a result oriented "after the fact" approach might well have social scientists pulling out their hair, we are stuck with it.
Lesson for Employers
As a practical matter, the OFCCP will be overwhelmed with the number of affirmative action programs it receives for desk audit purposes in response to this test program. The OFCCP District Offices around the country will have to engage in a triage-like process of weeding through the hiring and discharge data and, most importantly, the compensation data provided by each contractor in response to the scheduling letter to determine who are the best candidates to pursue beyond the desk audit stage in order to obtain the big bucks.
While the OFCCP most certainly will pay attention to impact ratio analyses indicating that minorities and/or women were rejected for hire, not promoted or fired at disproportionately high rates compared with their white and/or male counterparts, the OFCCP has learned that most employers strongly resist OFCCP's attempts to use this type of statistical analysis concerning hiring, promotion and/or discharge to identify an affected class of minorities and/or women and to then demand large amounts of back pay for that "affected class." Instead, rather than roll over, employers typically resist the OFCCP's demands by defending with the standard "non-discriminatory legitimate business justification" arguments those employers would make before the EEOC or a state anti-discrimination agency when faced with an individual charge of discrimination in hiring, promotion or discharge.
But the OFCCP also has learned that employers are less prepared and less willing to fight over determinations that minorities and/or women have been paid less than their white and/or male counterparts. That most likely is because to defend such litigation, the employer may be forced to spend significant amounts of time and money fighting the OFCCP's statistical case and may need to hire an expert statistician to rebut it. Of course, the flip side of the coin is that the OFCCP would have to spend the same amounts of time and money pursuing such cases and hiring its own experts, time and money that simply is not there for OFCCP. But no employer wins if it has to litigate a compensation case against the OFCCP (or, more precisely, against the Department of Labor's Solicitor's Office), because it costs time and money even to win those cases. A win for an employer occurs only when its affirmative action program is sufficiently well prepared so that the OFCCP, after reviewing it, decides not to pursue the employer beyond the desk audit stage and instead closes out the review with a Letter of Acceptance.
Q&B Comment
In many cases, as we have seen when assisting our clients prepare for desk audits in the last few months, the best way to receive a Letter of Acceptance is to give significant time and attention to the method the employer will use to present the compensation information to the OFCCP in response to the initial desk audit scheduling letter. As described in that scheduling letter, the employer has a number of options as to how the compensation information will be presented. Employers should consider testing various approaches through the use of legal counsel (internal or external), so that the preliminary testing may be protected by the attorney-client privilege and only the final analysis presented to the government may be discoverable in the event of future legal action.
If significant differences do appear between female and male salaries and/or between minority and non-minority salaries, even where the employer has presented the compensation data in the most favorable way possible, an employer should give strong consideration to addressing and explaining the legitimate business reasons for those differences in the Identification of Potential Problem Areas portion of its AAP narrative analysis. Experience has shown that where care and attention is paid to this area, the chance of the OFCCP accepting the employer's explanation for the differences, without the need for an on-site visit, goes up dramatically. Obviously, the flip side of the coin is that by providing the detailed analysis to the OFCCP, the OFCCP's attention will be called to the disparity. A judgment call must be made as to which is the better approach under the particular facts and circumstances of each employer's situation.
If you have any questions about the topics covered in this alert or if you need assistance with the preparation of your affirmative action program or in responding to a desk audit or other request from the OFCCP, please contact Ely Leichtling 414-277-5681, who chairs the Milwaukee District OFCCP Industry Liaison Group, Pam Ploor 414-277-5661 in our Milwaukee office, Tom Mandler in our Chicago office 312-715-5222 or your Quarles & Brady attorney.