EEOC

EEOC to Launch New “Guidance Portal” in Response to Executive Order 13891 – More to Come

February 19, 2020 by Matt Nusbaum

The EEOC will publish a notice in tomorrow’s Federal Register announcing the launch of a new “searchable, indexed database containing all EEOC guidance documents currently in effect.” EEOC guidance documents have not historically been difficult to find, but this new, centralized database is in response to Executive Order 13891 issued by President Trump on October 9, 2019.

The Executive Order takes issue with the practice of Executive Branch agencies imposing legally binding requirements on the public outside of the rulemaking processes contained in the Administrative Procedures Act (APA) and the Paperwork Reduction Act (PRA), arguing that such practices usurp Congress’s Constitutional legislative power. Guidance documents can be used to clarify existing obligations contained in the laws passed by Congress, but cannot be used to effectively create new law. However, the line between clarification and new law is not always clear.

Nonetheless, the Trump Administration has explicitly ordered Executive agencies to treat guidance documents “as non-binding both in law and in practice, except as incorporated into a contract.” Agencies are also instructed to “take public input into account when appropriate in formulating guidance documents,” and to make those documents “readily available to the public.”

The Executive Order directs the Office of Management and Budget (OMB), an Executive Branch agency, to “issue memoranda” implementing the order, which are presumably binding on all Executive agencies. In turn, the Executive agencies have 120 days from the issuance of OMB memoranda to review their guidance documents, rescind any they deem should be no longer be in effect, and to launch their guidance document databases. On October 31, 2019, OMB issued Memorandum M-20-02, giving Executive agencies until February 28, 2020 to comply.

The EEOC and the Department of Defense appear to be the first out of the gate. The EEOC’s guidance portal will go live on the February 28 deadline and will be located at www.eeoc.gov/guidance. The Department of Defense has already launched its “portal” at https://open.defense.gov/Regulatory-Program/Guidance-Documents/, though the announcement is not set to be published in the Federal Register until tomorrow.

The Executive Order also instructs Executive agencies to implement regulations (or revise existing regulations) regarding processes and procedures for issuing new guidance documents. Agencies will be required to clearly state that guidance is non-binding, and provide procedures for the public to petition for withdrawal or modification of a document. If a guidance document is deemed “significant,” agencies must also provide for a period of public notice and comment of at least 30 days and seek approval from the Office of Information and Regulatory Affairs within OMB. This provision effectively makes “significant guidance documents,” as defined in the Executive Order, subject to a mini-PRA.

What does all this mean in practical terms? That remains to be seen, but it should increase transparency. And it should increase stability and predictability for regulated communities. It will also likely lead to a lot of guidance revision and rule-making (or pseudo-rule-making). In other words, there could be a lot of changes coming down the pike.

One practical example jumps to mind regarding federal contractors. Contractors subject to the VEVRAA and/or Section 503 affirmative action regulations must annually prepare a “data collection analysis” that includes total headcounts of all “job openings,” “jobs filled,” and “hires” in the preceding 12 months. The OFCCP’s regulations do not define those terms. Rather, those terms are defined somewhat confusingly and inconsistently in the agency’s “FAQs,” which are guidance documents under the Executive Order. The OFCCP should explicitly treat those definitions as non-binding, meaning contractors do not have to follow them, and contractors should soon have a clear avenue for requesting that the OFCCP either rescind or revise those definitions.

But the potential impact of the Executive Order is myriad. At the very least, lawyers will be sharpening their pencils over any potential enforcement action that isn’t tied to an explicit regulatory citation, which could get very interesting. For instance, what is the OFCCP’s legal basis supporting the popular violation of “insufficient outreach” under the AAP regulations for women and/or people of color? This author has never been able to find it. We will all just need to buckle-up and see where all this leads.

EEO-1 Component 2 Pay Data Analysis

July 23, 2019 by Dan Kuang, Ph.D.

Despite resistance from the employer community, EEO-1 “Component 2” compensation reporting requirements will most likely be upheld by federal courts. As employers scramble to assemble their reports, many are also concerned with the potential risk exposure these reports may introduce. First and foremost, how will the OFCCP and EEOC use this information in their pay equity investigations?

Strategically, the OFCCP has largely shared their two-part game plan:

Internal Pay Equity: The OFCCP will measure pay equity within a company to ensure that there is internal equity; in other words, that employees within the same EEO Category are paid equitably.
External Pay Equity: The OFCCP will measure pay equity between companies; specifically, they will compare a company’s EEO-1 Pay Data against others in the relevant business sector.

Measuring Internal Pay Equity

To measure internal pay equity, the OFCCP is clear¹:

Mann-Whitney U test will be used to measure pay equity for gender and 2-group comparisons.
Kruskal-Wallis test will be used to measure pay equity for race where more than two groups are compared.

The mechanics of these statistical tests are not important for the average Human Resources (HR) professional. What is important, however, is for HR professionals to conceptually understand what these statistical tests are evaluating.

In concept, these analyses can determine whether there is a bias in the distribution of men and women along the 12 pay bands (ranging from lowest 1 = less than $19,239, to highest 12 = $208,000 or greater). For example, if women tend to dominate lower-paying salary bands while men are in the higher-paying salary bands, then women are being paid less than men. The Mann-Whitney U and Kruskal-Wallis can determine if there are differences in the distribution of gender or race across the 12 bands, respectively.

To help make sense of all of this, here is a sample report by gender for EEO Category professional:

Per the EEOC reporting requirements, the counts of male and female employees were reported by the pay bands. The statistical analyses will examine the actual counts. In concept, the analyses are looking for patterns in the distribution of males and females across the pay bands. In this example, note that females dominate the lower 1-5 pay bands, while men dominate the higher 7-12 pay bands. The statistical methods will identify this group difference and estimate its statistical significance.

Measuring External Pay Equity

The OFCCP is less clear on their strategy to measure external pay equity with the EEO-1 pay data. This is understandable as Title VII is primarily concerned with internal pay equity, and the law is less concerned with compensation philosophy where a company may choose to pay below-labor market standards as long as everyone is impacted equally.

Contractors need to be aware that there is an analytical approach that could assist the OFCCP and EEOC to identify employers with potential pay equity problems. Consider the following analysis example of EEO Category 1.1:

To simplify this analytical method, pay bands 1-6 are collapsed into “1-6” and pay bands 7-12 are collapsed into “7-12”. Note the interesting pattern when comparing internal composition (%) against external composition (%):

Females have higher internal representation in pay bands 1-6 (72%) than external data (60%).
Females have lower internal representation in pay bands 7-12 (28%) than external data (40%).
Males have lower internal representation in pay bands 1-6 (40%) than external data (57%).
Males have higher internal representation in pay bands 7-12 (60%) than external data (43%).

These results draw three important conclusions:

Women tend to hold the lower pay band positions (1-6) while men hold the higher pay band positions (7-12). This pattern maps onto the external pay data and, therefore, is not too concerning.
Women have higher representation in lower pay bands than external data, while the opposite is true for men, who have lower representation in the lower pay bands than external data. This pattern is problematic.
Women have lower representation in higher pay bands than external data, while the opposite is true for men, who have higher representation in higher pay bands than external data. This pattern is problematic.

In effect, women do worse than expected among lower pay bands while men do better than expected among higher pay bands. In and of itself, this pattern is sufficient to raise concerns for the OFCCP and EEOC. Additionally, if this pattern is observed across multiple EEO categories, concerns for systemic pay discrimination will naturally be raised.

Conclusions

One of the most compelling arguments against EEO-1 Component 2 reporting has been the concern that no meaningful conclusions can be drawn from analysis of broad EEO categories; the results will primarily be a mix of false positives and false negatives. Analyses that do not make proper apples-to-apples comparisons among similarly situated employees will most likely suffer this very likely consequence. Having said that, it is important to keep in mind that the OFCCP and EEOC will most likely use the EEO-1 pay data to flag employers who may have pay equity problems rather than to establish prima-facia claims of pay disparity. The methods discussed in this paper serve this purpose well.

By measuring internal and external pay equity with a lens for patterns, the OFCCP and EEOC can more efficiently identify employers who may be considered for more thorough and complex Title VII-oriented pay equity investigations. Employers should not dismiss the very likely danger that lurks once their EEO-1 Component 2 pay data is reported.

1. Federal Register, Vol 80 (20), p. 5118, footnote 47

Repost: July 2019 – LocalJobNetwork The OFCCP Digest, Volume 8 September Edition

EEOC Officially Reinstates “Component 2” Reporting and Sets Deadlines

May 2, 2019 by Matt Nusbaum

As anticipated, the Equal Employment Opportunity Commission (EEOC) has published a notice in the Federal Register announcing the “immediate reinstatement of the Revised EEO-1: Pay Data Collection,” commonly known as “Component 2.” EEOC’s initial announcement stated that the agency intends to collect 2018 pay data by September 30, 2019 (opening the survey some time in mid-July), but had not yet determined whether or not the agency would collect 2017 pay data.

Now the EEOC has scheduled the publication of a follow-up announcement in tomorrow’s Federal Register stating that the agency has determined that they will, in fact, be collecting 2017 and 2018 pay data. More importantly, they are going to collect both between mid-July and September 30 of this year.

By the time the EEOC re-opens the survey in July to collect the “Component 2” data, employers will have already filed their “regular” (Component 1) 2018 EEO-1 reports. And, of course, 2017 reports were filed last year. It remains to be seen how the EEOC will require employers to report the 2017 and 2018 Component 2 data, particularly those that file via the data upload option. Will employers be required to essentially re-file 2017 and 2018 reports with Component 2 data added in, or will employers be allowed to upload just the Component 2 portions? We won’t know until the EEOC releases more information.

In the meantime, employers are encouraged to start preparing now. That requires first getting your hands on the employee snapshot data used for the 2017 and 2018 reporting cycles. The employee ID numbers in those snapshots represent the universe of employees for the Component 2 data.

Using the 2017 and 2018 snapshots, the next step is to query the payroll system for total-year W-2 wages and total-year hours worked. For the 2017 snapshot population, wages and hours should reflect the totals for calendar year 2017. For the 2018 snapshot population, the data should reflect the totals for calendar year 2018.

The total-year W-2 wages should be whatever was reported in “Box 1” on the employee’s W-2 form. The EEOC expects that this information is readily available in most payroll systems.

Total hours worked could present more of a challenge. For FLSA non-exempt employees, the EEOC expects employers to use the actual total number of hours worked for each employee (because the FLSA requires employers to track that information). For FLSA exempt employees, employers are expected to use actual total hours worked if that information is captured; otherwise, employers are allowed to calculate an estimated total hours worked for each employee using 40 hours per week for full-time exempt employees, or 20 hours per week for part-time exempt employees. Be mindful of employees who switched from part- to full-time or vice versa during the calendar year and adjust their estimated total-hours-worked calculation accordingly.

Do not get distracted by whether or not any particular employee was in the same job the entire year or employed by the company the entire year. You are not required to collect pay and hours-worked data only for the period each employee occupied the job they were in when the snapshot was taken. You need total W-2 compensation earned and total hours worked in any capacity for the employees in each snapshot.

With total W-2 earnings and total-hours-worked added to your 2017 and 2018 EEO-1 snapshot data, you will be well poised to prepare your upload files when the EEOC releases those specifications sometime before the survey re-opens in mid-July.

Employers will likely want to evaluate the compensation data they are about to submit to the federal government, but the EEOC has not published any useful information yet as to whether, and to what extent, that data will be analyzed by the government, much less how. Stay tuned for an upcoming blog post by Dr. Dan Kuang with suggestions for how to approach those statistical analyses.

In the meantime, Biddle consultants are available to help you navigate these new waters. Contact a Biddle EEO/AA specialist at staff@biddle.com.