Preliminary Statement and General Terms and Conditions
As a preliminary matter, there are a number of new provisions in the Model CA that address the contractual nature of the agreement. In the Model CA a new Part I, entitled "Preliminary Statement" sets out the parties and the relevant laws at issue, and ends with the following statement, "In the interest of resolving the violations without engaging in further legal proceedings and in exchange for the good and valuable consideration described in this document, OFCCP and [CONTRACTOR] enter this contract ("Conciliation Agreement" or "Agreement") and agree to all the terms stated below." Part II entitled "General Terms and Conditions" picks up with what exactly OFCCP is promising in exchange for the contractor's agreement, "In exchange for [CONTRACTOR]'s fulfillment of all obligations in Parts III and IV of the Agreement, OFCCP agrees not to institute administrative or judicial enforcement proceedings under E.O. 11246 [Section 503, and/or VEVRAA] based on the violations described in more detail in Part III below." Paragraph 5 of the Model states, "The parties understand the terms of this Agreement and enter into it voluntarily." Paragraph 6 of the Model CA states, "This document and its attachments contain the complete and final understanding of the parties with respect to the matters referenced herein. This Agreement contains all terms by which the parties are bound, and it supersedes all prior written or oral negotiations and agreements. There will be no modifications or amendments to this Agreement unless they are in writing, signed by all parties." Paragraph 7 of the Model CA states, "If one or more provisions of this Agreement are rendered unlawful or unenforceable, the remaining provisions will remain in full force and effect." These additional provisions appear to be designed to make it very explicit that this is a contract in the traditional sense of promises entered into in writing for valuable consideration. OFCCP does not explain the impetus for these changes.
OFCCP has, also, made changes to clarify when the agreement starts and when it ends. The date of acceptance of the agreement by the government has been changed to be the "effective date" to clarify the date the contract will be deemed to have begun. The Model CA also includes a provision for the expiration of the agreement 60 days after submission of the final report unless OFCCP informs the contractor in writing prior to that date that all of the obligations under the agreement have not been met.
The other provisions from the standard text CA General Provisions are retained or slightly modified in the Model CA's Part II. For example, OFCCP has enhanced the non-admissions clause by adding "nor has there been an adjudicated finding that [CONTRACTOR] violated any laws" to the familiar non-admission language. The mandatory provisions from paragraph 9 of the standard text are retained with minor edits. The mandatory clause sets out the process for challenging violations of the agreement and makes it clear that OFCCP does not have to prove the underlying violations in order to enforce the agreement. OFCCP's right to review compliance with the agreement including requiring written reports, interviewing witnesses, etc., is retained. The only change here is that the Model CA deletes the limiting phrase, "as may be relevant to the matter under investigation and pertinent to [name of contractor]'s compliance." The contractor's continuing obligation to comply with the laws enforced by OFCCP is also retained.
(Addendum B, Model CA for Hiring, Recordkeeping, Access, and other issues tracks the language discussed above in Part I and II of that document.)
Specific Violations and Remedies
It is in Part III of the Model CA that it becomes specifically a Model CA for Compensation. This section begins with a brief statement of violations which covers when the violation is believed to have started, which specific race/ethnic, gender or other group was affected, what position or job group was involved, and what provision of the regulations was violated. This is followed by a summary of the specific findings in the compliance evaluation that led to the CA. Based on the sample language provided, contractors can reasonably expect that the specific violations will indicate the methodology used to identify the violations (e.g. multiple regression analysis), the date from which the violation is said to have commenced (the snapshot date), the specific groups favored and disfavored, the level of statistical significance of the disparity in pay, the basis for finding the comparators similarly situated, and the specific policy or practice that is alleged to have caused the disparity. Contractors should make sure that they have all of this information prior to entering into negotiations with OFCCP.
Contractors should be particularly concerned with Part III, Section C, Subsection 4, Impact Analysis and Prospective Salary Adjustments. This sample section provides that within a specified number of days from the effective date of the agreement the contractor will perform a regression analysis using methodology prescribed in an attachment to the CA. If the analysis results in a statistically significant disparity adverse to the disfavored group (presumably the same disfavored group as in the discrimination finding), the contractor agrees to increase the salaries of the disfavored group using the formula set forth in an attachment to the CA within a specified time period after conducting the analysis. The compliance officer is directed to consult with the OFCCP Statistician for the methodology to be used for both the statistical analyses and the formula for salary increases.
It is not clear why the regression analysis already conducted on the snapshot date is not sufficient to establish the appropriate prospective salary adjustments for the already identified affected class members. Since the violation itself is based on the snapshot date regression analysis, it would appear that the remedy should be based on the same data. Otherwise the remedy is not tailored to the specific violation. If no prospective adjustments have been made since the regression that supported the violation finding (the violation regression), a regression showing discrimination should be expected. If a post conciliation regression is run within days of the agreement, any employee hired into the position after the date of the violation regression but before the post-CA regression, who happens to share disfavored class characteristics would automatically be included for a pay increase. This would be so even if it is demonstrable that the individual could not have been impacted by the identified discriminatory policy or practice. Once the prospective salary increases have been determined from the violation regression, OFCCP does not need another regression to check if prospective salary adjustments have been made. It need only check to make sure the current pay reflects those increases rather than requiring another regression and a list of new affected class members.
Compensation compliance evaluations can take a considerable amount of time from start to finish. If the contractor has long since discontinued the policy or practice identified by OFCCP as causing the discrimination, the post-CA regression analysis is even further removed from the violation that the CA is meant to remedy. It appears from the sample language that the obligation to correct pay disparities would be linked to the particular disfavored group covered by the CA. If any other group exhibits pay disparities in the regression, OFCCP should open a new compliance evaluation. OFCCP would need to determine if this new affected class is entitled not only to prospective pay increase but also to back pay or other remedies.
It would appear from the Model CA that the contractor is expected to sign a CA before it knows how much it will be required to pay in prospective salary increases. If the salary increases were based on the violation regression, the contractor would have the information on its potential financial exposure before signing.
If the contractor has the capacity to conduct regression analysis, it may be appropriate to require it to do so as part of its monitoring responsibilities, not immediately after the CA signing to set prospective salaries as part of the remedy.
Contractors should also be concerned with the samples under D. Non-Monetary Remedies, in Part III of the Model CA. OFCCP, understandably, starts with a requirement to eliminate the identified discriminatory pay practices and policies alleged to have caused the identified discrimination. The second action is rather curious in light of the first. The second action is labeled "evaluation" and the sample language would require the contractor to conduct a study to determine whether promotion decisions, performance evaluation ratings, and a long list of other things, have a disproportionately negative impact on compensation for the disfavored group. The compliance officer is advised to consider requiring the contractor to retain an expert consultant to design and/or conduct required studies and develop new compensation policies. This is curious because OFCCP's analysis and OFCCP's experts were supposed to identify which contractor practices had a disproportionately negative effect on the compensation of the disfavored group and to identify remedies for the identified violation as part of the compliance evaluation. A study would appear to be duplicative of the original compliance evaluation. There is no guarantee that OFCCP would agree with the results of such a study even if it were conducted.
After this study, the sample says that the contractor is to develop and write new policies to eliminate all practices that had an adverse effect on the compensation of the disfavored group. It would appear, however, that this is what the first action was to have been - the elimination of the specific policy or practice responsible for the identified disparity.
OFCCP elaborates on non-monetary remedies with specific examples of possible corrective actions. The first sample has to do with performance evaluations and ratings. The sample posits the following:
- [CONTRACTOR]'s infrequent or inconsistent ratings of employees' performance had an adverse effect on the compensation of [DISFAVORED GROUP] [POSITION]. [CONTRACTOR] will write and implement a policy requiring annual performance evaluations based on standardized criteria. [CONTRACTOR] will conduct an annual performance evaluation for each employee working as a [POSITION(s)]. Each employee's direct supervisor will evaluate the employee's performance for the purpose of providing constructive feedback and guidance to the employee, and for documenting, inter alia, the extent to which the employee has achieved his/her job goals and objectives. Supervisors are expected to identify specific examples of performance behaviors to support high or low ratings. [CONTRACTOR] will provide training to supervisors explaining rating levels and how best to document and record employees' performance. Each employee must be provided an electronic or written copy of his/her completed performance evaluation form. [CONTRACTOR] must retain an electronic or written copy of each completed evaluation form for the length of time required by 41 C.F.R. § 60- 1.12.
Two themes are suggested by this example. First, OFCCP appears to be trying to extend its reach beyond those parameters previously applicable in its remedy formulation. OFCCP remedies, especially non-monetary remedies, traditionally hewed very close to the regulatory requirements. In essence, they required the contractor to do what they were already obligated to do under the regulations. If training was ordered it was specifically tied to EEO compliance. In this sample, OFCCP is devising performance evaluation processes and making them part of the remedy. A consequence of agreeing to this kind of remedy is that the employer could violate the CA simply by not providing a specific example to support a particular rating even if no discrimination did or could result from this failure. OFCCP's non-monetary remedies, like its monetary remedies, should be tailored to eliminate discrimination, not to implement general reforms in the contractor's business practices. A second theme has to do with increasing the contractor's recordkeeping obligations. This sample suggests several records that the employer would have to create, retain, and/or provide to employees. The heart of OFCCP compliance evaluations is the collection of records. As is well known and well documented, often the records the agency wants either do not exist or are too vague to use for analysis. The crafting of more specific recordkeeping obligations as a part of the remedy appears to be another way OFCCP hopes to make its job easier. Contractors, on the other hand, would likely aim to limit the agreement to records that are already required under the regulations rather than to take on additional recordkeeping requirements voluntarily.
The next specific example is the "Setting Starting Salary Sample":
- [CONTRACTOR] will develop and write improved guidelines for determining the starting salaries to be assigned to newly hired employees. Factors included in setting starting salary may include information regarding the employee's certifications, previous experience, and the salaries of similarly situated [CONTRACTOR] peers. Employees' pre-hire salary at his or her former employer should not be the sole factor considered because prior salary can reflect race or sex-based compensation discrimination.
The most interesting part of this example is the last sentence. OFCCP appears to suggest that the contractor is somehow responsible for correcting the cumulative impact of pay discrimination over the applicant's earning life. How is a contractor to know whether or to what extent the applicant's prior salaries were impacted by compensation discrimination as opposed to a host of other factors? How and why would a contractor be held responsible for a previous employer's discriminatory salary practices, especially where it had no knowledge of or role in those practices?
Direct salary discrimination is not the only kind of discrimination that might lead to a lower previous salary. For example, discrimination could have impacted the individual's opportunities to gain the experience, certifications, or other criteria that OFCCP indicates can be included in salary setting. In other words, any and all of the factors that go into salary setting could have been impacted in some way, shape, or form by discrimination. Why single out prior salary as the only factor that should not be the sole factor? If an employer uses it as "a factor" but not the "sole factor" will that be a safe harbor or will they still be viewed as using a potentially discriminatory factor in setting starting salary? From a disparate treatment perspective, the question is whether anyone showing up with the same background and previous salary would have received the same starting salary regardless of race, gender, or other protected status. If the answer is yes, the employer is treating all applicants and employees the same with respect to starting salary. They are not making a discriminatory salary decision. From a disparate impact perspective, there would have to be statistically significant evidence showing that using prior salary to determine starting salary has a disproportionately negative impact on an affected class and that the factor is not job related and consistent with business necessity or that there was a less discriminatory means to meet the employer's job related need. This is assuming, as OFCCP and EEOC do, that disparate impact is applicable in Title VII pay cases. It is not evident from the example, what OFCCP's legal basis is for the remedy suggested in its sample. While the sample goes on to require implementation of the improved salary setting system, it is not evident what OFCCP would accept as an "improved" salary setting system.
The Model CA refers the reader to a Memo entitled, "Model Equitable Relief Provisions for Conciliation Agreements and Consent Decrees" for additional suggestions and model language. Unfortunately, I was unable to find a copy of this in the revised FCCM and did not get a hit when I put this title in OFCCP's search engine.
It is important to note that the samples in the Model CA are just suggestions. As OFCCP takes pains to point out a CA is a contract. The terms are negotiable. Contractors should be particularly attuned to proposed CA language that exceeds regulatory requirements, that extends beyond the EEO arena, that potentially enlarges the affected class, and that is not narrowly tailored to the identified discriminatory practice. As I have noted before, if the statistical significance drops out of the case, as it so often does in compensation cases, the violation should identify a specific compensation decision that affected the remaining individuals in any cohort or small group analysis. The Model CA only addresses statistically significant pay disparities and some sample language may not be appropriate for small cohorts or one-on-one cases.