Federal Contractor's Pay Practices Will Now Undergo Statistical Analysis

Federal contractors' and subcontractors' compensation practices will be subject to statistical review using 'multiple regression analysis' under final standards issued by the Department of Labor's Office of Federal Contract Compliance Programs (OFCCP) on June 16, 2006. The statistical analysis is part of the compliance review process the OFCCP uses to determine whether a contractor or subcontractor has engaged in systemic compensation discrimination in violation of Executive Order 11246.

EO 11246, which requires federal contractors and subcontractors to implement affirmative action programs and to guarantee equal employment opportunity in the workplace without regard to race or gender, is enforced by the OFCCP. Under OFCCP regulations adopted in 2000 (41 CFR Part 60-2.17(b)(3)), covered contractors must evaluate their compensation systems to determine whether there are disparities based on gender, race or ethnicity. As part of its compliance review process, the OFCCP investigates whether contractors' pay practices are discriminatory.

Compensation Discrimination Standards

The agency's approach has two major components:

The determination of 'Similarly Situated Employee Groups' (SSEGs) - employees who are 'similarly situated,' for purposes of comparing contractor pay decisions - will focus on the similarity of the work performed, the levels of responsibility, and the skills and qualifications involved in the positions.

The use of multiple regression analysis, which is a statistical tool for understanding the relationship between two or more variables, will determine whether there are any significant differences in compensation by gender and/or race that are not explained by legitimate factors.

OFCCP will investigate whether employees in an employer's pre-existing groupings, such as pay grades or Affirmative Action Plan (AAP) job groups, are in fact similarly situated by looking at job descriptions and conducting employee interviews. In addition, the agency may consider other factors, including department or other functional unit of the employer, employment status ( e.g., full-time or part-time), and compensation status ( e.g., union, nonunion, hourly, salaried, or commission).

Voluntary Self-Evaluation Guidelines

The same day the agency issued the final standards, it also issued final guidelines for contractors' voluntary self-evaluation of compensation practices (71 FR 35114-35122). The guidelines are OFCCP's first effort at providing guidance to contractors and OFCCP personnel on suggested techniques for complying with the compensation self-evaluation regulatory requirement (41 CFR Part 60-2.17(b)(3)). Even so, the agency stresses that the guidelines are entirely voluntary.

If a contractor decides to follow the guidelines, it must make a reasonable attempt to produce SSEGs that are large enough for meaningful statistical analysis (at least 30 employees with five or more incumbents who are members of either a male/female pair or a minority/non-minority pair). Other than in establishments or AAPs with 500 or more employees (in which case multiple regression analysis is required), contractors may use any statistical analysis that accounts for factors that legitimately affect employees' compensation, such as experience, education, performance, productivity, location, etc.

Contractors must investigate statistically significant compensation disparities to determine whether they are explained by legitimate factors. Any disparities not explained by legitimate factors must be remedied.

If the contractor's compensation self-evaluation system reasonably meets the general standards outlined in the guidelines, the OFCCP will consider the contractor's compensation practices to be in compliance with EO 11246. At that point, the compensation portion of the audit will end.

Documentation Issues

All relevant statistical and non-statistical documentation must be made available to OFCCP during a compliance review. OFCCP may review personnel records and conduct employee interviews to determine the accuracy of the data and documents.

In addition, the OFCCP will review data and documents explaining the results of the non-statistical methods used to evaluate pay decisions of those employees who were eliminated from the statistical evaluation process.

The agency will also review documentation regarding any follow-up investigation into statistically significant disparities, the conclusions of such investigation, and any pay adjustments made to remedy disparities.

Alternative Compliance Certification

In the alternative, a contractor may certify that it has performed a compensation self-evaluation, which it asserts is subject to attorney-client privilege and/or attorney work product doctrine, in lieu of producing the methodology or results during a compliance review. For contractors who elect this route, OFCCP will evaluate their compensation practices without regard to their compensation self-evaluation.

More Aggressive Focus On Compensation

The final standards signal that OFCCP will now focus more aggressively on compensation during compliance reviews, Doyle observed. 'As the agency trains compliance officers and implements the final standards, contractors should expect more aggressive investigation of compensation during OFCCP audits.'

Most contractors are not prepared to perform multiple regression analysis and getting to that point is costly. Compensation data are often incomplete and not available electronically. Moreover, because the OFCCP is now following legally recognized standards and utilizing sophisticated statistical techniques, contractors risk greater exposure to liability, not only as a result of an OFCCP investigation but also in the context of class action litigation.

Even where a multiple regression analysis does not reveal any pay disparities, plaintiffs' lawyers who bring class actions typically are astute enough to tweak the model in search of pay disparities revealed by the addition or elimination of a factor or two, Doyle observed.

* * *

Suggestions For Contractors

What should contractors be doing? According to Doyle, contractors need to develop specific compliance and defense strategies in response to the new standards. He recommends that contractors focus on four areas.

1. Construct careful pay divisions. First, explore different options for submitting the 'Pay Divisions,' which are the groupings the contractor uses to submit compensation data in response to Item 11 of the Scheduling Letter contractors receive from the OFCCP. 'Contractors should attempt to identify Pay Divisions that pass OFCCP's desk audit test and that are likely to be acceptable to OFCCP,' Doyle advised.

Pay Divisions should be consistent across a contractor's facilities within the same OFCCP district, unless there are compelling explanations, like a unionized workforce in one facility but not the other, Doyle added.

2. Do privileged internal assessment. Second, consider doing an internal privileged risk assessment analysis to assess whether there are compensation disparities that OFCCP would discover if it did a full-blown investigation using multiple regression analysis. This is also important for large employers in industries that have been targeted for class action litigation, said Doyle.

According to Doyle, the objectives of the risk assessment are:

to explore the likely regression models that the OFCCP, EEOC and plaintiffs' lawyers would use;

to develop compelling defense models, and

to take steps to secure the factual foundation for a defense model.

'You can't meet any of these objectives unless you really understand how the courts look at these analyses and the types of arguments advanced by plaintiffs' lawyers and government agencies,' he said.

'The risk assessment allows the contractor to identify and correct misclassifications, data discrepancies, and other problems that would make it easier for OFCCP or a plaintiff's lawyer to demonstrate statistically significant pay disparities. Data changes are often greeted with suspicion by OFCCP during an investigation, so waiting until an audit has been scheduled may limit the contractor's defense options,' explained Doyle.

Doyle advises contractors not to disclose the fact that they have conducted the risk assessment for two reasons. 'First, they should take the position that the fact that they conducted the risk assessment is confidential. Second, they did not conduct the risk assessment to comply with the AAP regulations, so it is irrelevant to any discussion of their OFCCP compliance status,' Doyle suggested.

Contractors should also understand that the attorney-client privilege and the attorney work product doctrine only apply if the risk assessment was conducted for the purposes of obtaining legal advice or developing a defense to potential litigation, Doyle cautioned. 'A risk assessment conducted to meet a regulatory obligation probably will not be privileged.'

3. Pass on the voluntary guidelines. Doyle advises contractors not to attempt to take advantage of the voluntary self-audit guidelines without first consulting counsel and conducting a confidential risk assessment. Although the guidelines afford contractors some latitude to develop a statistical analysis other than a full multiple regression analysis, these statistical analyses are less sophisticated than multiple regression methods and are likely to be less accurate because they do not account for the joint impact of the legitimate factors that explain pay disparities, Doyle observed.

'Less sophisticated methods are likely to generate pay disparities more frequently and to generate larger pay disparities. It may be difficult to convince OFCCP compliance officers that large pay disparities are entirely justified and that back pay and pay adjustments are not warranted,' said Doyle. 'A contractor who makes pay adjustments based on these less rigorous statistical methods may also have a harder time defending against reverse discrimination claims,' he added.

4. Keep it simple. Finally, OFCCP regulations require contractors to do some kind of monitoring of compensation practices as part of their affirmative action programs. The voluntary guidelines make clear that this monitoring obligation does not require contractors to conduct any data analysis or statistical analysis. For example, a contractor could comply by having its HR and compensation staff monitor compensation decisions on a regular basis to ensure that all decisions are based on legitimate nondiscriminatory factors, Doyle remarked.

Contractors have ultimate discretion to determine what they will do; however, during a compliance review, OFCCP will ask contractors to identify what they have done to comply, Doyle said. He recommends that contractors not develop multiple regression analyses outside of a risk assessment study. To conduct a risk assessment study that serves the purposes of assessing risk and identifying risk mitigation steps, contractors will need statistical and legal expertise, noted Doyle.

Contractors can contain costs by conducting much of the risk assessment in-house after utilizing legal and statistical experts to assist with the initial setup of the analysis, Doyle suggested. However, he stressed that corporate counsel must remain intimately involved as the overall manager of the internal risk assessment project in order to protect the analysis under the attorney-client privilege and work product doctrine.

Published November 1, 2006.