Affirmative Action Alert
PHYSICIAN, HEAL THYSELF! Issues Final Guidance on Standards for Investigating and “Self-Evaluating” Pay Discrimination

June 20, 2006

Back in November 2004, we reported that the federal Office of Federal Contract Compliance Programs had issued a proposed guidance regarding methods for investigating systemic compensation discrimination, employer self-evaluation of its compensation practices, and a “safe harbor” for contractors.  The agency has just now issued its Final Interpretative Standards, as well as guidelines for contractors and a Fact Sheet relating to both. 

We are addressing them here in “FAQ” format.

In summary, the OFCCP has become serious about investigating compensation discrimination.  The agency will be using the sophisticated multiple regression analysis to analyze contractors’ data, and contractors are well-advised to do the same.  The OFCCP recently filed two compensation cases, both of which resulted in significant settlements – one for $5.5 million!

FAQs

BURNING QUESTIONS
What is the “safe harbor”?
Is there a catch?
Does Constangy recommend that we go for the optional “safe harbor”?
If I don’t care about the optional “safe harbor,” do I need to read further?

THE GUIDANCE, AND CONSTANGY’S RECOMMENDATION
Which rules, if any, apply to all contractors?
If we should use the OFCCP’s method in any event, then why not go for the “safe harbor” while we’re at it?   
If we do a self-evaluation but want the protection of attorney-client privilege or work product, what should we do?

CONDUCTING THE SELF-EVALUATION
So, Constangy recommends that we try to use the OFCCP’s recommended methodology in conducting our self-evaluation, even if we decide not to disclose the data or results to the OFCCP?
What should we do?

EVALUATION METHODS
What is a “multiple regression analysis”?
What if my company can’t do a multiple regression analysis?
What if I can’t or don’t want to use a statistical method of analysis?

THE BOTTOM LINE
What happens with the self-evaluation information during a compliance review?
What is the result under each of my options?

BURNING QUESTIONS

What is the “safe harbor”?

The details regarding how to qualify for the safe harbor are set forth below.  But the “safe harbor” means very generally that, if a contractor conducts a self-evaluation consistent with the guidance, and if the evaluation shows that there is no compensation discrimination, then the OFCCP will accept the contractor’s finding.

Contractors are not required to qualify for the safe harbor – it is completely optional.

Is there a catch?

Of course!  The catch is that contractors will have to comply with the OFCCP’s guidelines about conducting the evaluation and, of most concern, will have to make the evaluation and underlying documentation available to the OFCCP upon request during a compliance review. 

For this reason, Constangy does not recommend that contractors attempt to qualify for the optional “safe harbor.”

Does Constangy recommend that we go for the optional “safe harbor”?

No, for the reason given above.  However, the OFCCP offers a “middle ground” that Constangy does recommend.  (See below.)  Moreover, Constangy recommends that contractors, to the extent possible, perform a multiple regression analysis to analyze their pay data for the following reasons:  First, it is essential to know what your data will look like to the OFCCP.  Second, the multiple regression analysis, unlike the discredited DuBray Analysis, is statistically and logically sound.

If I don’t care about the optional “safe harbor,” do I need to read further?

Of course you do!  There are some provisions that apply to all contractors, which will be discussed below.  It’s also a good idea to be generally familiar with the guidelines so you can make an informed decision about what is best for your company.

THE GUIDANCE, AND CONSTANGY’S RECOMMENDATION

Which rules, if any, apply to all contractors?

Whether or not a contractor seeks the optional “safe harbor,” it must conduct some sort of self-evaluation of its compensation practices.  Failure to conduct any self-evaluation is a violation of the regulations regarding compensation analysis.

In other words, the “safe harbor” is optional; the self-evaluation is not. 

If we should use the OFCCP’s method in any event, then why not go for the “safe harbor” while we’re at it?  

The problem with the optional “safe harbor” is confidentiality.  To take advantage of the safe harbor, the contractor has to be willing to share its evaluation, as well as the underlying data, with the OFCCP.  Moreover, the OFCCP will conduct employee interviews and use other means to verify that the evaluation is proper.  The information provided by the contractor will be protected to the fullest extent allowed by the Freedom of Information Act (“FOIA”), which means something less than 100 percent protection. 

Because the information is shared with the contractor’s “adversary” (that is, the OFCCP), the contractor would also have waived (given up) any claim of attorney-client privilege or attorney work product that it might otherwise have had.

Not only is the OFCCP entitled to view “unprivileged” self-evaluations, but also adversaries in employment discrimination lawsuits might be entitled to obtain the evaluations in discovery  Any unexplained and unremedied compensation disparities could support an award of compensatory and punitive damages.

If we do a self-evaluation but want the protection of attorney-client privilege or work product, what should we do?

The new guidelines say that you may provide an “Alternative Compliance Certification” (RECOMMENDED) stating that (1) you have performed a compensation self-evaluation with respect to the affirmative action program or facility at issue; (2) that the evaluation was conducted at the direction of your counsel; and (3) that your counsel has advised you that the evaluation and results are subject to attorney-client privilege or the work product doctrine.  This alternative certification must be in writing and signed “by a duly authorized officer of the contractor under penalty of perjury.”

If you choose the alternative certification, you will not be able to take advantage of the safe harbor, and the OFCCP will conduct its own independent investigation of your compensation practices.  However, Constangy considers this a worthwhile trade-off, considering the risks associated with disclosure of your self-evaluation.

CONDUCTING THE SELF-EVALUATION

So, Constangy recommends that we try to use the OFCCP’s recommended methodology in conducting our self-evaluation, even if we decide not to disclose the data or results to the OFCCP?

Yes.

What should we do?

Step One.  First, the contractor should divide employees into “Similarly Situated Employee Groupings,” or “SSEGs,” based on actual job duties and responsibilities, and skills and qualifications required.  Ideally, the SSEGs will contain at least 30 employees, and at least five who are members of either of the following pairs:  (1) female/male; or (2) minority/non-minority.  The SSEGs should encompass 70% of the employees in the AAP or establishment, and the OFCCP will allow contractors to group employees from more than one establishment to satisfy this criteria.  (The OFCCP recognizes that satisfying these criteria may not always be possible and provides for exceptions.)

Because the OFCCP has completely disavowed the DuBray Analysis, which considered employees in the same pay grade as “similarly situated,” contractors must avoid using SSEGs based on “artificial,” or formal, criteria, such as pay grades.   The agency has made it very clear that it will be looking at actual, not formalistic, similarities.  Inappropriate “formalistic” divisions may include job groups for Affirmative Action Plan purposes.  (You can and should still use job groups in your AAPs – just don’t count on them for your compensation analysis.)

Step Two.  Once the SSEGs are created, the contractor should determine all factors that affect compensation for each SSEG.  This is very important, and very difficult, for most contractors.

Step Three.  Once a year, the contractor should perform a multiple regression or other  analysis on the SSEGs.  If the facility or AAP has 500 employees or more, it is required to use a multiple regression analysis to qualify for the safe harbor.  (If, as Constangy recommends, the large contractor is not seeking to qualify for the “safe harbor,” the contractor may use another method.)  For those employees not included in an SSEG, the contractor may perform a non-statistical analysis for their compensation.

Step Four.  If the analysis (whether multiple regression or some other method) reveals a pay disparity that is statistically significant, the contractor must determine whether there is a legitimate explanation.  If not, the contractor must take appropriate remedial action, which means providing back pay and other “make whole” relief.

Step Five.  For the “safe harbor,” the contractor must create and maintain the following documents and make them available to the agency during a compliance review:

(1) Documents to explain the rationale for the SSEGs, including the exclusion of any employees, and the factors used in the analysis, and the form of the analysis.

(2)  Data used in the “current” statistical analysis, as well as the results of statistical analyses conducted in the two years preceding the current analysis.

(3)  In the case of employees who were excluded from the statistical analysis, any data or documents that explain the results of the non-statistical analysis of the excluded employees.

(4) Documents related to follow-up as a result of the discovery of statistically significant disparities in compensation, conclusions reached, and (if applicable) remedial action taken.

Even if a contractor chooses not to seek qualification for the “safe harbor,” this type of documentation is very helpful to create and maintain, provided that an attorney is involved in the process to allow the information to be protected by the attorney-client privilege.

EVALUATION METHODS

What is a “multiple regression analysis”?

In the context of a compensation review, multiple regression analysis examines pay (the “dependent variable”) against all the factors that may affect pay (the “explanatory variables”) – for example, experience with prior employers and relevant education. 

What if my company can’t do a multiple regression analysis?

First, there is no requirement that you perform a multiple regression analysis unless your facility or AAP has 500 or more employees AND you are seeking the safe harbor. 

That said, a multiple regression analysis is a statistically rigorous method of analysis if used under the proper conditions and with proper attention paid to explanatory variables. 

If you cannot perform your own multiple regression analysis, Constangy’s Affirmative Action Practice Group can do it for you. 

Contractors should be cautioned that the OFCCP will perform the multiple regression analysis using its own software.  For obvious reasons, Constangy recommends that contractors conduct their own analysis before the OFCCP does.

What if I can’t or don’t want to use a statistical method of analysis?

If your facility or AAP has 500 or more employees AND you are seeking the safe harbor, you have no choice – you are required to use a multiple regression analysis.  But otherwise, you are free to use non-statistical methods in assessing your compensation. 

THE BOTTOM LINE

What happens with the self-evaluation information during a compliance review?

If you seek to qualify for the safe harbor (NOT RECOMMENDED), you would provide it to the OFCCP during a compliance review.  The agency may also review personnel files and interview employees to verify the accuracy of your data and analysis.

If you do not seek to qualify for the safe harbor but use the Alternative Compliance Certification to preserve attorney-client privilege and work product protections (RECOMMENDED), the OFCCP would still have the right to review personnel files and interview employees, but it would not have a right to any of the above documentation. 

What is the result under each of my options?

If you seek the safe harbor (NOT RECOMMENDED), you will have to disclose all of your self-evaluation documentation to the OFCCP during a compliance review.  The OFCCP has the right to review personnel records and interview employees to verify that your data and analysis are sound.  But if your self-evaluation showed either (1) no statistically significant pay disparities, or (2) pay disparities that had a good explanation, or (3) pay disparities that were corrected, and if your evaluation documents “pass inspection,” the OFCCP will accept your evaluation.

(As always, there are some exceptions to this.)

If you do not seek the safe harbor but use the Alternative Compliance Certification to preserve attorney-client privilege and work product (RECOMMENDED), you still are required to conduct a self-evaluation of your compensation practices.  In the event of a compliance review, you will not have to provide the self-evaluation or underlying documentation to the OFCCP.  However, the OFCCP may conduct its own analysis, and it may or may not find a violation.  (As with any OFCCP review, you would have the opportunity to provide unprivileged information in rebuttal.)

Constangy does not recommend that you forgo both the safe harbor and the Alternative Compliance Certification.

Constangy does not recommend that you forgo a self-evaluation.  The regulations specifically provide that this is a violation.

Please note that the above is only a summary of the new guidelines, which took effect on June 16, 2006. 
   
If you have any questions relating to the OFCCP’s Final Rule or contractor guidelines, need assistance with a self-evaluation, want to conduct a multiple regression analysis, or need any other affirmative action compliance assistance, please contact any member of the Affirmative Action Practice Group, or the Constangy attorney of your choice.   

  1. JVA increased the coverage threshold from a contract of $25,000 or more to a contract of $100,000 or more;
  2. JVA changed the categories of covered veterans. The “Vietnam era” category was eliminated and replaced with a new category of “veterans who, while serving on active duty in the Armed Forces, participated in a United States military operation for which an Armed Forces service medal was awarded.” Vietnam era veterans may continue to be covered in other categories. Further, JVA expanded the coverage of “veterans with disabilities” to include all veterans with service-connected disabilities. Finally, the coverage of recently separated veterans was expanded to three years after discharge or release from active duty.
  3. JVA modified the mandatory job listing requirement. Listing positions solely with America’s Job Bank is no longer sufficient. Rather, all positions must be listed with the “appropriate employment service delivery system.” As a practical matter, this means that contractors should continue to list their open positions with America’s Job Bank but also with the State Employment Service. The job listing requirement applies to all vacancies for positions other than executive/senior management that will last more than three days.
If you have any questions relating to the OFCCP’s proposed regulations, please contact any member of the Affirmative Action Practice Group, or the Constangy attorney of your choice.