Despite everything going on in the current political climate, key elements of the legal superstructure of demographic data analysis remain in place. In this article, Affirmity Principal Business Consultant, Patrick McNiel, PhD, takes a look at these key legal underpinnings—the aspects of law that still require the collection of race, ethnicity, and/or gender demographic data.
Workforce Demographic Data Requirements in Title VII of The Civil Rights Act
At the federal level, the primary impetus for collecting demographic data comes from Title VII of the Civil Rights Act and its amendments. The following text comes from Section 709C of Title VII:
“Every employer, employment agency, and labor organization subject to this subchapter shall (1) make and keep such records relevant to the determinations of whether unlawful employment practices have been or are being committed…”
Later, the 1991 amendment to the Civil Rights Act expanded the definition of discrimination to include disparate impact. Disparate impact occurs when a seemingly neutral policy has a discriminatory effect because it excludes or disadvantages members of a protected group, even if there was no intent to discriminate.
For example, screening individuals out of the hiring process when they have gaps in employment may seem innocuous and perhaps even sensible in some cases. However, this practice screens out women at a higher rate than men because women are far more likely to take time from work to raise kids and care for family. It’s hard to argue that this practice is a business necessity.
Disparate impact liability ensures that organizations and institutions in positions of power do not use proxies for race, gender, or some other protected class to intentionally discriminate through indirect means. Additionally, this form of liability encourages due diligence in ensuring that operations, policies, procedures, and methodologies do not unintentionally discriminate by creating unnecessary barriers, hardships, or requirements that would disproportionately affect people in a protected class.
However, the only way to determine if disparate impact is occurring is to run adverse impact statistics. This tells you if the rate of a certain event (such as getting hired or getting fired) is significantly larger or smaller for a protected group. You’ll get nowhere running those statistics without demographic data, and such data therefore falls within the definition of “records relevant to the determinations of whether unlawful employment practices have been or are being committed” given in the aforementioned Section 709C.
YOU MAY ALSO ENJOY | ‘50 Years of Affirmity: A Video Celebration’
Uniform Guidelines on Employee Selection Procedures (UGESP)
Further reinforcing the legal necessity of your workforce demographic data collection efforts is a set of requirements stemming from the Uniform Guidelines on Employee Selection Procedures (the Guidelines). These guidelines describe what employers must do to comply with various EEO laws and show that their selection procedures are valid, job-related, and consistent with business necessity. Following these guidelines provides a defense for selection procedures that show adverse impact, and can prevent a judgment of disparate impact discrimination.
Organizations should note that while the Guidelines were indeed part of the now-rescinded EO 11246 regulations, they are also part of the EEOC’s regulations for the Civil Rights Act. Therefore, they should still be considered fully in effect. Furthermore, the Guidelines have a history of being treated with great deference by courts. Even with recent Supreme Court cases that might erode their influence, they should still be taken very seriously.
The following text discloses why the Guidelines must be followed:
“The use of any selection procedure which has an adverse impact on the hiring, promotion, or other employment or membership opportunities of members of any race, sex, or ethnic group will be considered to be discriminatory and inconsistent with these guidelines, unless the procedure has been validated in accordance with these guidelines…”
Only when followed can an instance of adverse impact be demonstrated not to be an instance of disparate impact discrimination. You can only legally have different selection rates associated with your selection procedures if the procedure is valid, consistent with business necessity, and related to the job in question.
The Guidelines go on to say:
“Each user should maintain and have available for inspection records or other information which will disclose the impact which its tests and other selection procedures have upon employment opportunities of persons by identifiable race, sex, or ethnic group.”
Consequently, companies need to conduct adverse impact analyses, and to do this for a selection procedure, they must collect demographic information from people prior to that procedure.
UNDERSTAND YOUR CHANGING RESPONSIBILITIES | ‘The Shifting Year in the Life of a Workforce Compliance Professional: 7 Key Non-Discrimination Program Responsibilities’
Relevant Court Interpretations and Decisions
The courts have interpreted many aspects of Title VII law through the years, but there are two key rulings relating to demographic data collection that organizations should be aware of:
- Griggs v. Duke Power Co. (1971): This case was a precursor to the 1991 amendment to the Civil Rights Act and the concept of disparate impact theory. It adopted the view that neutral tests or criteria that disproportionately impact protected classes must be job-related and consistent with business necessity, even if there’s no discriminatory intent. A large number of judgments in subsequent cases have maintained this view.
- EEOC v. Crothall Services Group, Inc. (2015): The EEOC sued Crothall—a national facilities management provider—in the U.S. District Court for the Eastern District of Pennsylvania. The core claim was that Crothall used criminal background checks in hiring without maintaining race, sex, or ethnicity data needed to assess disparate impact. The key decision was that plaintiffs and federal regulators may sue based solely on failure to maintain the required records, regardless of any actual discrimination. This decision rejected Crothall’s argument that recordkeeping was optional.
The New Compliance Landscape
Despite the well-established legal precedents and remaining laws, the current compliance landscape has some organizations considering abandoning their data collection efforts at a time when they should actually be doubling down. A mixture of executive orders, personnel and policy changes at the federal agencies, and new state-level requirements are all factors that are worthy of examination.
Executive Orders
Executive Order 14173 (Ending Illegal Discrimination and Restoring Merit-Based Opportunity) set the tone for things to come, and did a number of things to minimize DEI efforts in the U.S.:
- It rescinded EO 11246, which required race/ethnicity and gender-related affirmative action for federal contractors. For a lot of companies, this was the primary reason for collecting demographic data on these protected groups. But as we have seen above, the legality and rationale for continuing this data collection still remains.
- It prohibited “illegal DEI” for federal contractors and required certification that “illegal DEI” has not occurred, as a condition of the contract.
- It establishes a process whereby federal agencies are expected to inform on companies conducting “illegal DEI”, and make it so whistleblowers within a company can do the same.
The latter two points here are understandably a source of anxiety for contractors—a chief concern is whether the False Claims Act could be brought into play if they certify that “illegal DEI” has not taken place, only for this to be challenged by a later investigation.
To the extent that data collection is related to DEI, this initial executive order has made companies nervous about data collection. However, no decision is made by the act of collecting data alone, and “illegal DEI” as has been subsequently described, largely appears to require that some decision is made about people (training is a potential notable exception, but this is currently hotly debated).
This, coupled with the fact that data collection is required elsewhere, means that in order to follow the letter of the law, companies must continue to collect the data.
Executive Order 14281 (Restoring Equality of Opportunity and Meritocracy), on the other hand, attempts to countermand the disparate impact theory of liability that is part of the amended Civil Rights Act (and has been upheld by courts for over 50 years.)
The primary effect of this EO is to discourage states from enforcing disparate impact liability while also instructing federal agencies to disregard it. This prompts the question: “if the EEOC will no longer care about our data on demographics, should we be collecting it?” Simply put, yes, you should because:
- Disparate impact liability remains the law and plaintiffs can still sue.
- The EEOC will likely continue to use this data against organizations in other cases, such as those seeking to prove intentional discrimination rather than disparate impact.
- A number of states are not going to react by capitulating. For example, California recently came out with a law that requires adverse impact statistics to be run for AI systems used in the employment context.
GET INSIGHT ON THE DOJ’S CURRENT POSITION | ‘Justice Department Releases Guidance for Federal Contractors Regarding Unlawful Discrimination and “Illegal DEI”’
Federal Agencies and Remaining Requirements
Recent personnel changes at the EEOC and OFCCP are also shaping federal activity and focus. The EEOC currently lacks a quorum after the firing of incumbent, Democrat-appointed commissioners Charlotte Burrows and Jocelyn Samuels. At the time of writing, neither has been replaced, so EEOC activity has been curtailed. Prior directives are in place that allow President Trump’s acting EEOC general counsel, Andrew Rogers, to file suit and deal with prior resolutions. However, guidance changes, rulemaking, and policy changes are on hold.
If this situation were to change, it may have implications for the Uniform Guidelines on Employee Selection Procedures (the Guidelines), which are embedded within the EEOC’s regulations. If the EEOC gains back its quorum, we could see the Guidelines altered or eliminated—or perhaps simply left alone.
Federal contractors and employers more generally should be conscious of the following federal requirements:
- EEO-1 Reporting
- VETS-4212 Reporting
- Vietnam Era Veterans’ Readjustment Assistance Act (VEVRAA) AAPs
- Section 503 of the Rehabilitation Act of 1973 AAPs
State-Level Affirmative Action
Federal-level affirmative action planning previously provided a framework for complying with state-level affirmative action requirements, but the rescission of Executive Order 11246 has created something of a vacuum at the state level.
Organizations are now finding that many states—21 at our last count—have some level of reporting requirement that they may be subject to. This picture could easily change in the coming months and years, as states fall in line with the current administration’s stance or double down on the necessity of running such programs.
LEARN MORE ABOUT REMAINING FORMS OF AFFIRMATIVE ACTION | ‘Why Affirmative Action Should Still Be On Your Agenda: When Is an AAP Required at the Federal and State Level?’
Conclusion: An Opportunity to Not Only Continue, But Evolve Your Approach
So, if you agree with our argument that you should continue to collect demographic data, what does this look like in practice? And specifically, from a workforce planning and talent management standpoint, how are you going to strategically use demographic data if you decide to continue collecting it, and what needs to change in how you report it?
Before the recission of the executive order, contractors were preparing affirmative action plans and primarily looking at females versus males, and minorities versus non-minorities. There’s an argument that we’ve actually been given an opportunity to be much more flexible and intentional in the way that we prepare this kind of reporting, perhaps even to the level of other business intelligence within the organization.
After all, in most organizations, you know what your sales performance has been for the last quarter, allowing you to identify operational efficiencies and market trends. So why wouldn’t you want to also understand what your workforce looks like and how it evolves over time?
Continue to Explore the Strategic Case for Workforce Demographic Data
This blog post is an extract from a larger piece—our ebook, “The Strategic Case for Collecting and Analyzing Workforce Demographic Data.” In addition to the information above, you can also expect to learn:
- How demographic data collection allows you to better control the story around your people processes, and to understand your strategic heading
- The potential long-term impact of halting data collection and analysis
- The opportunity presented by the removal of affirmative action’s rigid structures
Continue getting invaluable insights about this topic: download the full ebook. For assistance with all aspects of workforce demographic data collection and analysis, contact us today.
About the Author
Patrick McNiel, PhD is a Principal Business Consultant for Affirmity. Dr. McNiel specializes in workforce analytics using both qualitative and quantitative methods to analyze employment practices and inform employment decisions. Dr. McNiel holds a PhD in Industrial and Organizational Psychology, and is licensed to practice psychology by the State of Texas. Connect with him on LinkedIn.