COMPENSATION DATA COLLECTION TOOL: EO SURVEY 2.0?
The new memorandum from President Obama, signed on Equal Pay Day, directs the Secretary of Labor to publish a Notice of Proposed Rulemaking (NPRM) within 120 days of the date of this memorandum. The rule will require Federal contractors to submit summary compensation data by sex and race to DOL. According to the memorandum, the rule should allow for evaluation, such that DOL can identify those more likely to be engaging in compensation discrimination, while also minimizing, to the extent feasible, the burden on the contractor community.
A new compensation data collection tool has been in the works for a few years, with the Advanced Notice of Proposed Rulemaking (ANPRM) having been published August 10, 2011. The ANPRM mentioned that the previous attempt at compensation data collection, through the EO Survey, had been discontinued in light of questionable utility for its intended purposes. The ANPRM asked for comments on a series of questions relating to what types of data should be collected to allow OFCCP to focus compliance reviews on those more likely to be discriminating in pay and to what extent additional information might help in furthering other objectives (e.g., potentially initiating multi-establishment reviews). There were 2,441 comments received in response to the ANPRM, which should be addressed, to the extent feasible, in the NPRM.
Is history repeating itself? In 1998, OFCCP began developing and releasing the Equal Opportunity (EO) Survey with three objectives: 1) to increase compliance with equal opportunity requirements by improving contractor self-awareness and encouraging self-evaluations; 2) to improve the deployment of scarce federal government resources to those contractors most likely to be out of compliance; and 3) to increase agency efficiency by building on the tiered-review process already accomplished by OFCCP’s regulatory reform efforts, resulting in better resource allocation. The survey was piloted with just under 7,000 contractors in early 2000, sent to approximately 50,000 contractors in late 2000, and sent to 10,000 contractors in late 2002, 2003, and 2004. The compensation information was the last of three parts to the survey. Pay data was requested by EEO Category or AAP job group (contractors had the option to pick one) for full time employees only and with three salary data points (average, minimum, maximum) and average tenure reported for each sex/race category (e.g., White male, Asian female).
Two separate reports were conducted to analyze the utility of the EO survey. Both reports were referenced in the final rule rescinding the EO Survey in 2006. See this previous blog for a thorough overview of the key aspects of that rule and an overview of the reports. In short, the EO Survey was deemed not valid for serving its stated objectives, partially because the data collected was too aggregated (i.e., an average/minimum/maximum salary by EEO category, not similarly situated jobs) to allow for meaningful analysis. Note: Statistical analyses such as t-tests, Fishers Exact Tests and multiple regression analyses cannot be conducted on aggregated data.
The EO Survey is the perfect example to demonstrate the need for thoughtful, considered steps on this new data collection tool. Over 85,000 contractor establishments were required to complete the survey over a four year period, though it was never actually used for its intended purpose. When it was rescinded, the OFCCP noted that the compensation data collected was inappropriate for the intended purpose. The study concluded that an unacceptable number of non-discriminating contractor establishments would be “flagged” by the tool while an equally unacceptable number of discriminating contractor establishments would not be “flagged” for more detailed screening. However, it took six years, over 1,800,000 contractor hours, and over $5,000,000 estimated tax payer dollars to make that determination. (Note: Contractor hours were determined by the 21 hours stated to complete the survey, multiplied by an approximation of the number of surveys completed. The monetary information was obtained by the Center for Corporate Equality in 2006 via a FOIA request.)
While it is still not clear that this additional data collection tool is necessary, the Presidential memorandum puts the contractor community on notice that the tool will be forthcoming anyway. The often-repeated figure of 77 cents to the dollar that women earn compared to men is presented prominently in both the memorandum and the ANPRM. This is despite a report, commissioned by the DOL, exploring the cause of this gap and concluding that it is not possible to determine if the difference is really due to discrimination, as opposed to the myriad of other relevant factors involved in determining pay.
The tool is coming, and contractors should be prepared to provide input and comments on the NPRM. Things to watch for and consider include:
- The required level of grouping (e.g. EEO-1 category, AAP job group, salary grade, SSEG, PAG)
- What type of data will be collected (e.g. Base salary, bonus, commission)
- What percentage of contractors will have to submit this data every year, and at what cost (labor hours and monetary) to contractors.
- Additionally, it remains to be clarified exactly how this data will be used to identify contractors who are more likely to be engaging in pay discrimination. See the Center for Corporate Equality’s comment to the ANPRM for a more thorough analysis of these concerns, including estimates from contractors about the extent to which some compensation data is even available in HRIS systems and hour estimates for compiling and submitting the data.
Finally, there has been no clear explanation as to how this data will be collected and stored. With electronic data breaches on the rise, there is a very real reason for contractors to be wary of how their data will be protected and held confidential. It is also likely that some smaller contractor establishments will be reporting data at a level that makes personal employee compensation information readily apparent to anyone with access (i.e., the compensation data of one Asian Male in a category), and any attempt to aggregate that data further would also make analyzing it meaningless.
Sometime within the next 120 days there will be an NPRM for the new compensation data collection tool. Many questions remain about how the ANPRM comments will be incorporated into the NPRM, what data OFCCP will collect, how that data will be collected and protected, what will be done with it, and whether precautions are being taken to avoid a repeat of the original EO Survey’s failure. How the OFCCP will meet the President’s mandate to minimize the burden on contractors in light of the flurry of other changes to regulations and reporting also remains to be seen. Stay tuned!
By Kristen Pryor, M.S., Associate Consultant, and David Cohen, President, DCI Consulting Group