Affirmative Action Services Conducts Nationwide Study On Best Practices In Compensation Analysis - By Dr. Jim Higgins, Ed.D.

At the 2010 National Industry Liaison Group (NILG) meeting held in Las Vegas Nevada, Dr. Patricia Shiu, Director of the US Office of Federal Contract Compliance Programs (OFCCP) reported that it was the OFCCP’s intention to rescind the 2006 Voluntary Interpretive Standards for Compensation Discrimination (Standards) that were implemented under the George W. Bush Administration. These Guidelines relied on statistical analyses of federal Contractor compensation practices. Specifically, the guidelines focused on three components.

First, the OFCCP expected federal Contractors to take steps to ensure that employees were appropriately grouped. This requirement was based on the assumption that employees who are employed in jobs requiring similar qualifications, require workers to perform similar duties, and to complete their work under similar conditions (e.g., level of independence, supervision received, etc.) should be paid at a similar level. If this assumption is valid, then it would be safe to assume that any differences in average pay received by groups (e.g., males and females) will be due to one of two factors: a) legitimate differences in job related factors such as experience, training, and performance, or b) factors that are not legitimate (i.e., discrimination).

The 2006 Guidelines state that once employees are grouped into Similarly Situated Employee Groups (SSEGs), the appropriate statistical tool for distinguishing between legitimate differences in pay and potential discrimination is Multiple Linear Regression Analysis (MLR). MLR is a tool that allows analysts to “model” the factors that impact an outcome (in this case, compensation). When used correctly, MLR would allow federal Contractors and the OFCCP determine whether differences in employee compensation are due to legitimate, job related factors, or if they are instead at least partially related to an employee’s protected characteristics such as gender or racial status.

Given that every statistical analysis has some likelihood of identifying “spurious results” or differences that are due to chance rather than real, repeatable differences, the OFCCP added a third stage in the analysis. This third stage, which was to be used whenever statistically significant differences were observed or if the SSEG had too few employees to permit a valid MLR analysis, was a non-statistical cohort analysis. This technique provided a qualitative analysis comparing the highest and lowest paid members of one group with the highest and lowest paid workers of the comparison group to determine whether there were other factors that might account for why one groups highest or lowest paid members were paid differently than the comparison group’s highest or lowest paid workers.

Together, these three components are like the legs of a stool. Jointly, they form a stable and viable approach to understanding Contractor compensation. Take away any one component and the methodology will become unstable and the likelihood of realizing invalid analytical results dramatically increases. Further, by establishing the Guidelines, the OFCCP provided federal Contractors with a clear set of criteria against which they will be evaluated. The idea was that, by providing this framework of analysis, the OFCCP could encourage contractors to use a standardized methodology and drive the entire contractor community toward ensuring that employee compensation is based on valid, job-related factors rather than characteristics that are irrelevant to job performance such as gender or minority status.

From the OFCCP’s perspective, however, the Guidelines have placed them in a difficult situation. In a budgetary environment where program effectiveness is judged by the size and scope of conciliation agreements, the Guidelines may have backfired. A review of recent compensation-related conciliation agreements between federal Contractors and the OFCCP will reveal that few compensation-related cases have been settled. Those responsible for the OFCCP’s budget may look at these results and decide that it indicates the OFCCP is not effectively performing its role. The assumption could be made that the economy is rife with pay discrimination and the OFCCP is failing to find it.

There is an alternative explanation, however. If could be that, now that federal Contractors possess a clear standard against which they will be judged and having a strategy that was carefully crafted to find and address compensation disparities, they are taking proactive steps to eliminate disparities. To the extent that Contractors are finding and addressing compensation disparities proactively—rather than only when audited—there are fewer situations with unexplained compensation disparities left for the OFCCP to find.

This is especially true for large Contractors. These organizations tend to have the greatest resources and therefore can either afford to hire the qualified statistical staff or pay for competent consultants who can conduct the analyses. These larger contractors also tend to have the resources to make the pay adjustments necessary eliminate pay disparities. As a result, it may become increasingly less likely that the OFCCP will find large disparities.

As a result, the ultimate question is which of these two reasons for a lack of compensation-related conciliation agreements is correct? Is the OFCCP failing to perform its job? Or rather, is the fact that it is finding fewer pay disparities evidence that the OFCCP is actually performing its job better than ever?

On January 3, 2011, the OFCCP announced that it is proposing to rescind Interpretive Standards for Systemic Compensation Discrimination (see 41 CRF Parts 60-1 and 60-2). They are accepting public comments regarding this proposal through March 4, 2011.

From December 18, 2010 through February 18, 2011, Affirmative Action Services, a full-service affirmative action consulting firm founded by Dr. Jim Higgins, Ed.D. and Robert Newland, a former OFCCP Compliance Officer began conducting a non-partisan and objective study to collect feedback from federal Contractors and to provide this information to the OFCCP as one part of their wider public comment collection effort. Our goal is to provide the information collected with no bias and with the intention to improve the process of identifying and addressing pay disparities. In subsequent posts, I will discuss many of the findings from this study and the implications for fostering a diverse and equitable workforce.

[To Be Continued]