Under federal affirmative action regulations, all covered federal contractors are required annually to perform an evaluation of compensation practices to ensure minorities and women are being fairly treated. Since inception, the published regulations, 41 CFR 60.2.17(b)(3), have been silent as to the methodology contractors should employ to analyze their workforces for pay equity purposes. On June 16, 2006 the Office of Federal Contract Compliance Programs published its final guidelines on performing compensation self evaluations, entitled "Voluntary Guidelines for Self Evaluation with Compensation Practices for Compliance with Non Discrimination Requirements of Executive Order 11246 With Respect to Systemic Compensation Discrimination". The Final Guidelines follow the November 16, 2004 proposed self-evaluation guidelines, which drew vigorous public comment.
The Final Guidelines provide only suggested compliance techniques and are voluntary, and the OFCCP has taken great pains to emphasize there are other ways to comply with the obligation to ensure pay equity. However, to create an incentive for contractors to use the Final Guidelines, the OFCCP will forego independent review of a contractor's raw compensation data and instead evaluate the adequacy of the contractor's compliance with the Final Guidelines themselves. This approach is called "compliance coordination."
While compliance coordination could make things easier for contractors, a principal component of the Final Guidelines requires contractors to provide monetary relief to remedy compensation disparities found through the self-evaluation. In other words, the OFCCP expects use of the Final Guidelines to result in "pay fixes" for affected minorities and women. There also is the danger that compliance with the Self Evaluation Guidelines will expose contractors to potential liability outside the context of OFCCP review, and employers can expect compliance coordination to be a routine part of discovery requests.
The principal changes to the Proposed Guidelines and the resulting Final Guidelines are discussed below.
Brief Synopsis of the Proposed and Final Guidelines
According to the OFCCP, its Proposed Guidelines had four principal components. First, the Proposed Guidelines were voluntary. Therefore, contractors could choose any method to comply with the annual self-evaluation regulatory obligation. In their final form, the OFCCP has reinforced the voluntary nature of the Guidelines.
Second, to encourage contractors to use the Proposed Guidelines, the OFCCP offered to forego independent analysis of the contractor's compensation data and practices and, instead, assess whether the contractor's self evaluation reasonably adhered to the Proposed Guidelines. This approach was labeled "compliance coordination." Part and parcel to an acceptable self evaluation was the expectation that a contactor would create and keep (and disclose, if necessary), adequate records of: (1) how it conducted the evaluation and formed its comparison groups, (2) the results of any statistical or non-statistical analyses performed and the data used to perform the analyses, and (3) the investigations into disparities and resulting remedies undertaken. Such information was to be kept at least two years, but longer if the self evaluation was marginally acceptable and required re-examination during a subsequent compliance review. This component has not materially changed.
Third, the Proposed Guidelines set forth minimum standards of acceptability for self evaluation. For example, a compliant self-evaluation must:
- cover at least 80% of the workforce in the facility/plan;
- be performed annually;
- be based on similarly situated employee groupings (SSEGs) of employees (those performing similar work and possessing similar levels of responsibility, skills and qualifications) containing at least 30 employees, including at least 5 comparisons from each protected group;
- contain some form of non-statistical analysis to cover employees who do not fall into SSEGs;
- for facilities or plans of less than 250 employees, utilize a form of statistical analysis that allows evaluation of SSEGs, and accounts for factors that influence compensation (i.e., experience, education, performance, location);
- use multiple regression analysis for facility or plan workforces of 250 employees or more;
- test for "statistical significance" (at the OFCCP's two standard deviation threshold); and,
- resolve statistically significant pay disparities that can not be explained by legitimate factors through make whole relief.
This component has been materially changed in the Final Guidelines.
Fourth, the Proposed Guidelines sought to address concerns that disclosing the self evaluation to the OFCCP would interfere with a contractor's subsequent exercise of the attorney-client or attorney-work product privileges. Consequently, the OFCCP proposed a "Compliance Certification Alternative" allowing the contractor to avoid submitting its self-evaluation for review by the OFCCP if the Contractor certified under penalty of perjury that its self evaluation was covered by either or both of the aforementioned privilege doctrines. However, contractors who chose this method could not take advantage of compliance coordination and, instead, would be required to submit raw data for analysis by the OFCCP. This component also contained reassurances about confidentiality of data. While this component has not materially changed, the OFCCP made modifications to language intended to bolster the sanctity of the legal privilege and increase confidentiality assurances.
"Relaxed" Standards for Creating SSEGs
Many contractors complained that the Proposed Guidelines compelled contractors to force non-similarly situated employees into SSEGs where they did not belong. These concerns arose from 1) language in the Proposed Guidelines that the OFCCP would "carefully scrutinize" any analysis that did not encompass at least 80% of the workforce at the facility or AAP, 2) the limited number of factors a contractor was permitted to consider when creating SSEGs, and 3) the requirement that SSEGs contain at least 30 employees with 5 comparisons from each protected group (the "size requirement").
The OFCCP addressed the first two concerns, but not the third (the size requirement). To address the first concern, the Final Guidelines state that contractors should not include non-similarly situated employees in SSEGs and lowered from 80% to 70% the proportion of the workforce that must be included in any acceptable analysis. To afford additional flexibility in creating SSEGs, the Agency also added a provision that a self evaluation could cover multiple facilities as long as it covers at least all of the employees within an affirmative action plan or establishment.
To address the second concern, the Final Guidelines add new factors to consider when determining whether employees are similarly situated for proper inclusion in a SSEG. These new factors include department or functional unit, location, full or part time status, and compensation status, such as whether an employee belongs to a union or is paid on an hourly, salaried or commission basis.
While the foregoing changes are helpful, the ramifications of submitting a self evaluation that covers more than just the plan or facility being audited are unclear. To the extent a self evaluation covers more than the facility undergoing audit, a contractor may unintentionally alert the OFCCP to problems at other facilities. Until the contractor community gains a greater sense about how the OFCCP will apportion liability and consider results outside of the context of an individual audit, contractors should approach expansion of self evaluation beyond individual facilities with caution.
Some Relief from Multiple Regression Analysis
The Proposed Guidelines required use of multiple regression analysis at facilities with 250 or more employees. Multiple regression analyses are complex and may require the involvement of a statistician. An even greater burden than performing the analysis is the task of gathering the data necessary to perform the analysis. The utility and reliability of a multiple regression analysis is directly influenced by its ability to account for all of the legitimate factors that influence compensation. Therefore, there are tremendous costs associated with investigating, collecting and auditing the data that goes into the analysis. To reduce this burden, some comments proposed 1) significantly raising the 250 employee threshold, 2) requiring self evaluations less frequently that annually, or 3) allowing use of a multi-tiered approach similar to the method used to conduct compliance evaluations.
Threshold Raised from 250 to 500, Annual Evaluation Still Required
In responding to contractor concerns over the burdens associated with multiple regression analysis, in the Final Guidelines the OFCCP raised the threshold from 250 to 500 employees at a worksite, but declined to allow use of a multi-tiered approach to self evaluation or allow evaluations to be conducted less frequently than annually. The OFCCP explained that self evaluation must be done annually to prevent the development of patterns of discrimination. The OFCCP also rejected the multi-tiered approach because it reasoned that any burden placed on contractors by performance of the more rigorous multiple regression analysis was outweighed by the fact that performing such an analysis is voluntary and doing so allows the contractor to take advantage of the compliance coordination option.
OFCCP Declines to Allow Contractors to Use Multi-Tiered Approach as Substitute for Self-Evaluation
In contrast to a contractor's obligations under the Final Guidelines to perform a comprehensive self-evaluation, the OFCCP uses a multi-tiered approach. The OFCCP first conducts a rudimentary analysis of the very basic compensation data it receives as part of each routine plan submission. If this preliminary analysis reveals problems, the OFCCP will require additional fields of data upon which it will perform a "cluster" regression analysis. If this second level of analysis reveals problems, the OFCCP will schedule an on-site to investigate and collect data sufficient to perform a full blown multiple regression analysis. However, there is no guarantee the OFCCP will refrain from performing a multiple regression analysis at a facility under review even if the usual triggers are absent.
The OFCCP further explained that while a multi-tiered approach is not appropriate for contractors, the OFCCP needs to use such an approach because it cannot possibly perform a full blown multiple regression analysis for each of the approximately 100,000 covered federal contractor establishments or during each of the 6,000+ reviews it conducts each year. In other words, performing multiple regression analysis during each review would be too burdensome for the OFCCP. Ironically, the OFCCP does not deem it too burdensome for contractors. The OFCCP also declined to make available to contractors the compensation analysis programs the Agency itself uses to review compliance, stating that the commercially available SAS software used by the Agency can be purchased by contractors as well.
While the OFCCP did not accept tiered evaluation as an alternative for self-evaluation under the Final Guidelines, contractors still may consider using an "OFCCP-like" approach for identifying issues and focusing resources. In other words, a contractor may review compensation data using the OFCCP's multi-tiered approach and, if no problems are found, turn its focus to the next facility.
Contractors Conducting Self Evaluations Must Remedy Compensation Disparities with Back-Pay and Other Make Whole Relief
The Proposed Guidelines provided that remedial action is required where there are statistically significant compensation disparities within an SSEG that cannot be accounted for by legitimate non-discriminatory factors. Responding to requests for clarification about under what circumstances a remedy would be required and how to determine the remedy, the OFCCP states in the Final Guidelines that the "remedial action that is appropriate will depend on the facts of the case, but should include back-pay and other make whole relief." (Emphasis added). The OFCCP expects that contractors will use back-pay as a remedy during the first year of "fixing" an employee's salary and, thereafter, in subsequent "iterations" of the annual self evaluation make prospective "fixes."
Contractors Who Comply with the Guidelines Do Not Have to Submit the Compensation Data Required by Item 11 of the OFCCP's Scheduling Notice
Item 11 of the OFCCP's standard scheduling letter requires contractors to submit annualized compensation (wages, salaries, commission and bonuses) by salary range, rate, grade, or level showing total number of employees by race and gender and total compensation by race and gender. Contractors requested guidance about how to respond to Item 11 if they followed the Self Evaluation Guidelines. Consequently, the OFCCP responded that contractors who choose to comply with the Self Evaluation Guidelines and to take advantage of the compliance coordination incentive do not have to submit data in response to Item 11. Rather, simply stating words to the effect of "contractor seeks compliance coordination under the OFCCP's voluntary compensation self-evaluation guidelines" relieves the contractor from the submission obligation. According to the OFCCP, making this statement will trigger a telephone inquiry from the OFCCP regarding what documents and information the contractor must submit to establish that its self evaluation is compliant.
Language Regarding Confidentiality of Submissions Is Strengthened
In response to contractor concerns about disclosure of compensation data, the OFCCP has added stronger confidentiality provisions. Specifically, the Final Guidelines state the "OFCCP will treat compensation and other personnel information… as confidential to the maximum extent the information is exempt from public disclosure under the Freedom of Information Act." This language is intended to give an additional level of comfort. To take full advantage of the protections attached, contractors should ensure they clearly mark all data submitted to the OFCCP as "Privileged and Confidential."
Alternative Compliance Certification Remains Largely the Same
A number of concerns were raised about the section of the Proposed Guidelines addressing the Alternative Compliance Certification. Using the ACC allows contractors to submit a "certificate of privilege" to avoid the actual self evaluation. However, by using the ACC, contractors lose the opportunity to take advantage of the OFCCP's compliance coordination initiative and instead, must submit raw compensation data for analysis by the OFCCP. While many comments offered suggested modifications to this component of the Proposed Guidelines, most were rejected. For example, relief that was sought and rejected by the OFCCP included requests to 1) allow ACC without submission of any data, 2) perform random "checks" of contractors who choose AAC (rather than review them all), and 3) require the OFCCP to return compensation data and self-evaluations when its analysis was complete.
The primary modification made by the OFCCP in this section focused on changing language in the Proposed Guidelines that appeared to suggest only quantitative or statistical analyses would satisfy the general obligation to annually perform "some kind" of compensation self evaluation under 41 CFR 60-2.17 (b)(3). The OFCCP reaffirmed that contractors may choose any method of self-evaluation to comply with this section of the regulations and removed the words "analysis" or "analyses" after the phrase "self-evaluation" to reinforce that there are many types of self evaluations that could suffice; not only statistical or quantitive analyses.
No Adverse Inferences Drawn from Not Using Guidelines
Many contractors expressed concern that the OFCCP would negatively view the failure to follow the self evaluation guidelines. To address these concerns, OFCCP revised its regulations to explicitly state that the guidelines are voluntary and that a contractor's decision not to follow the Guidelines "will not be used as a basis for any negative or adverse inference about the contractor's compliance status."
OFCCP Declines Grace Period for Implementation
Because of the complexity of the type of analysis required and the potential enormity of the resulting burden, many contractors requested a grace period during which to implement the new guidelines. However, the OFCCP declined to so. According to the OFCCP, since the self evaluation guidelines are voluntary, contractors are not required to use them. Consequently, the OFCCP does not believe contractors need a grace period.
Contractors opting to follow these guidelines should be fully aware of the implications of doing so. The methods embraced by the Self Evaluation Guidelines have been characterized by the OFCCP as a recognized form of proof of discrimination under Title VII. Therefore, the principal danger in using these guidelines involves a possible concession that the salary disparities identified by a company are evidence of discrimination. Also, once a contractor has conceded the appropriateness of a SSEG for purposes of self evaluation (whether or not you use Compliance Coordination or try to cloak your efforts in privilege by using the Alternative Compliance Certification) it may be difficult, if not impossible, to argue they are not suitable for comparison in other contexts. Contactors also should be concerned about the discoverability of evaluations performed in compliance with these guidelines. In the context of discrimination lawsuits, plaintiffs may seek discovery of any self evaluation performed and any resulting remedial actions. Disclosure may be unavoidable, and plaintiffs' attorneys will become aware that federal contractors have this obligation.
Consequently, we recommend that contractors perform their self evaluations under the direction of and with the participation of counsel, taking full advantage of the attorney-client or attorney-work product privilege doctrines. How employees are grouped and the factors used to account for compensation will be crucial, no matter what form of evaluation your company chooses. Discussions about how to arrive at these groupings and factors must be made under the cloak of privilege. Attorneys in our Jackson Lewis Affirmative Action Practice Group are available to provide advice about these new guidelines and all other related areas.
The Final Guidelines can be found at http://www.dol.gov/esa/regs/fedreg/notices/2006005457.htm. The Self Evaluation Guidelines were published at the same time as the OFCCP's final Guidelines on Systemic Compensation discrimination, which can be found at http://www.dol.gov/esa/regs/fedreg/notices/2006005458.htm.
 There also are other challenges to understanding the OFCCP's approach to analyzing compensation. For example, the OFCCP does not use a uniform approach for the first level of analysis. Some OFCCP Regions use a straight 5% disparity in average salaries by pay groupings (e.g., titles) of minorities vs. non-minorities and women vs. men, while others use the 30-30-3 rule. The 30-30-3 rule is triggered if there are at least 30 affected protected group members, 1) making at least 2% less on average than their non-protected group counterparts in the same categories, 2) who comprise at least 30% of that protected groups representation in the workforce, and 3) who are affected at least 3 times worse than non-protected group members making on average at least 2 % less than protected group members.