A Conversation about Building Lawful and Effective Workplace Diversity Practices

[00:00:00] Imagine waking up tomorrow, grabbing your coffee, and pulling up your company's highly celebrated diversity dashboard. Right. You know, you're looking at the exact same metrics your CEO was just praising on an all-hands call last quarter. The numbers are moving in the intended direction. Everyone's getting a pat on the back. And then, bam. Yeah, the other shoe drops. Exactly. You get a notification that the Department of Justice has just targeted your company in a multi-million dollar federal fraud investigation.

[00:00:27] And the crazy part? The evidence they are using against you is that exact same dashboard. I mean, welcome to the new reality of human resources. It's a staggering reversal, truly. I mean, we have gone from an environment where those dashboards were considered the absolute gold standard of corporate citizenship to an environment where they're actively being subpoenaed as evidence of unlawful conduct. And a huge welcome to you, the learner. We are so glad you're joining us today.

[00:00:54] Usually, you know, when we talk about workplace policies on this deep dive, there's some expectation of stability. Sure. But the landscape we are looking at today in late 2026, it has experienced just unprecedented legal tremors. It is a completely different world out there right now. The ground hasn't just shifted beneath employers' feet. In some specific areas of employment law, the ground has completely opened up and just, well, swallowed the old playbooks whole. Which brings us to the mission of today's deep dive.

[00:01:22] We're analyzing a brand new, just incredibly comprehensive piece of source material from the Human Capital Leadership Review. It's a fantastic paper. It really is. The author is Dr. Jonathan H. Westover, and the article is titled Inclusion on Firmer Ground, Building Fair, Lawful, and Effective Practices in the Era of DEI Enforcement. But before we even take our first step into this material, I need to pause and establish some really important ground rules for you, our listener.

[00:01:51] I'm really glad we're doing this. The context here is non-negotiable. Yeah, it has to be said. So before we dive in, a quick note to you, our listener. DEI is a highly debated, highly politically charged topic. People have incredibly strong feelings on all sides of this issue. Absolutely. Our mission today is not to take a side or endorse a viewpoint or play politics.

[00:02:15] We are here strictly as your guides to objectively unpack the original source material, the timeline, the federal lawsuits, the peer-reviewed science, so you can understand exactly what is happening out there and what the data says actually works. All right, because regardless of your personal ideology or where you fall on the political spectrum, the practical reality is that if you are managing a team or leading an organization or honestly just trying to navigate your own career,

[00:02:43] the rules of the road have fundamentally changed. They really have. And Dr. Westover's article sets out to extract an evidence-based toolkit from the wreckage of the old programs. We're going to figure out how you build a workplace that survives this intense new legal scrutiny and, most importantly, how you actually hire and keep the best people. Okay, let's unpack this. We need to start by understanding the timeline because if we rewind the clock to, say, the summer of 2020,

[00:03:11] thousands of organizations poured massive foundations for these programs. Yeah, massive investments. They set explicit representation goals. They launched all these identity-based development tracks. How do we get from that peak to this current era of just massive federal enforcement? Well, to understand the shift, we have to trace a very specific sequence of foundational legal rulings. It's like a domino effect.

[00:03:35] The first major domino fell in 2023 at the Supreme Court with Students for Fair Admissions v. Harvard. But wait, hold on. That was a case about kids getting into college, right? We're talking about adults getting jobs and promotions. How does a university admissions case rewrite corporate employment law? That's a great question. Technically, it didn't change employment law. The Harvard decision was based on the Equal Protection Clause and Title VI, which covers federally funded programs.

[00:04:04] Title VII is the federal law covering employment discrimination. Okay, so different laws entirely. Exactly. But courts in the legal profession, they don't operate in hermetically sealed silos, you know. When the Supreme Court struck down Harvard's race-conscious admissions program, they sent a massive, undeniable signal. They signaled intense judicial skepticism toward any kind of demographic-conscious decision-making. So all the corporate lawyers were basically reading the tea leaves? They absolutely were.

[00:04:33] Employment lawyers across the country immediately started calling their corporate clients, warning them that the logic the Supreme Court applied to universities was inevitably coming for the workplace. Wow. And they were totally correct to sound the alarm. Because the very next year, 2024, the courts fundamentally altered Title VII itself with Muldrow v. City of St. Louis. Okay, so walk me through Muldrow.

[00:04:57] Why was a case about a city employee such a game-changer for massive multinational corporations? So prior to Muldrow, the legal standard in a lot of federal circuits was that if an employee wanted to sue their employer for a discriminatory action, like, say, a forced job transfer or being excluded from a program, they had to prove that the action caused them significant harm. Significant harm. Like what?

[00:05:24] You generally had to show that you suffered a pay cut or a demotion in rank or a tangible loss of benefits. Meaning if my boss transferred me to a different department entirely because of my gender, but, you know, my salary stayed exactly the same, I might not have even had a case. Under the old standard, yeah, that's entirely possible. Many courts would have just thrown that case out. But Muldrow lowered the bar entirely. The Supreme Court ruled that employees no longer need to show significant harm.

[00:05:54] They only need to show some harm to a term or condition of employment. Okay, let's make this visceral for our listeners. If you're a mid-level manager listening to this right now, you're probably mentally scrolling through every mentoring coffee chat you set up last week. Oh, for sure. What does some harm actually look like in the day-to-day corporate grind? Think about the developmental opportunities that companies offer, right? Imagine a prestigious internal leadership retreat. It's fully paid, top executives are going to be there,

[00:06:24] and it's just this great networking opportunity. Sounds like a standard corporate perk. Right. But if a company decides to restrict attendance to that retreat based on race or gender, even with the totally noble intention of uplifting an underrepresented group, any employee who is excluded now has an incredibly low bar to sue. Being denied that networking opportunity, just being left off that calendar invite, is easily argued as some harm to their employment experience.

[00:06:53] So suddenly, all those exclusive, specialized programs, the women in tech retreats, the minority leadership accelerators, those transformed overnight from celebrated HR initiatives into massive legal liabilities. That is the new reality. And the Supreme Court compounded that liability in 2025 with another massive decision. Ames v. Ohio Department of Youth Services. This was a unanimous ruling, which is quite rare,

[00:07:22] and it completely dismantled an old legal hurdle called the background circumstances rule. Okay, I feel like you need a law degree just to keep up with some of these terms. I read the summary of the Ames case in the source material, but break down this background circumstances concept for me. I know, the legal jargon gets super dense, but the mechanism is actually quite simple once you see it. Historically, in some federal circuits, courts created a double standard. A double standard how?

[00:07:49] Well, if a plaintiff from a minority group sued for discrimination, they went through the standard legal process. But if a plaintiff from a majority group, for instance, a white male employee, wanted to sue for discrimination, the courts actually made it harder for them. Harder how? What did they have to prove? They had to prove background circumstances. Basically, they had to show that their specific employer was the highly unusual type of employer who actively discriminates against the majority.

[00:08:16] It was this extra hurdle just to get their case heard in court. Oh, wow. But in Ames, the Supreme Court unanimously said absolutely not. They ruled that Title VII protects every single individual equally. Full stop. The legal standard is now identical regardless of your demographic background. So it's a total leveling of the playing field for plaintiffs. Completely. So we have this rapid fire series of Supreme Court decisions making it incredibly easy to sue over employment programs.

[00:08:43] But the executive branch didn't just sit on the sidelines, did they? The source material gets into the False Claims Act, which I got to be honest, sounds like something out of a defense contractor thriller, not an HR manual. Yeah, the executive branch's involvement in January 2025 is really where the financial stakes became existential. We saw the issuance of Executive Order 14173. And what did that do?

[00:09:07] This order revoked old affirmative action requirements for federal contractors that had been in place since the 1960s. But the truly revolutionary part was tying DEI compliance directly to federal funding via the False Claims Act or the FCA. I have to admit, my understanding of the False Claims Act is mostly from reading about like health care fraud or aerospace companies that were charging the Pentagon for spare parts. Right. The classic fraud cases. Exactly. How on earth does an HR diversity initiative trigger a fraud statute?

[00:09:37] This is where we really need to dive deep into the mechanics of government contracting. When a company does business with the federal government, whether they are providing software or consulting services or manufacturing, they submit invoices to get paid. Okay. Standard vendor stuff. Right. But under this new executive order, federal contractors and grantees are required to legally certify that they do not operate employment programs that violate federal anti-discrimination law.

[00:10:04] So there's an actual piece of paperwork where some executive signs on the dotted line saying our HR practices are completely lawful. Yes, exactly. And the False Claims Act makes it a severe federal crime to knowingly submit false claims for government payment. If an executive signs that certification, but the company is actively running, say, an unlawful demographic quota system for hiring in their HR department, they haven't just committed a standard employment violation. They've committed fraud.

[00:10:32] They have committed fraud against the United States government. They essentially lied to get federal money. Wow. The DOJ isn't going to care about your good intentions if you lied on a federal invoice. The DOJ sees it purely as a compliance and fraud issue. And the SCA has a unique, frankly terrifying provision for employers. It allows for private whistleblowers to file lawsuits on behalf of the government. The legal term is a kwaitam relator. Oh, bounty hunters.

[00:11:00] That is a very accurate description. If a whistleblower brings an FCA case and the government recovers money from the company, the whistleblower gets a percentage of that settlement. Okay. I want to make sure I'm grasping the scale of this.

[00:11:14] You're telling me that a mid-level data analyst in human resources who just happens to have access to the company's hiring goals and demographic targets could look at those numbers, realize they violate the new legal standards, report their own employer to the federal government, and potentially walk away a millionaire? That is the exact mechanism the False Claims Act created. Basically, weaponized internal HR compliance overnight.

[00:11:41] Your biggest legal threat isn't necessarily an external auditor anymore. It's the employee sitting three desks down who knows exactly how the promotion slate was finalized last week. That completely changes the psychology of middle management. You aren't just worried about a routine EEOC complaint. You are actively worried about triggering a federal fraud investigation. And the agencies themselves, the DOJ and the EEOC, really fueled this shift in 2025 and 2026.

[00:12:08] They released highly specific guidance targeting what they call proxies. Proxies. Like using a stand-in variable to get the demographic result you want without explicitly stating it? That's the exact definition. The guidance explicitly warned employers against using facially neutral terms to unlawfully engineer demographic outcomes. The classic example they cited was requiring candidates to have a specific type of lived experience.

[00:12:37] Right, because lived experience sounds like a neutral qualification on paper, but it's often a covert filter. Exactly. If an employer uses the phrase lived experience in a job description, but the DOJ uncovers internal emails showing the hiring manager was actually using that phrase as a mechanism to filter out certain racial or gender groups, the agency considers that an unlawful proxy for intentional discrimination. So they're looking right past the jargon. They are.

[00:13:04] And then the final massive tremor hit in June 2026. This was the rescinding of the 1979 guidance, right? Yes. The EEOC voted to officially rescind its 1979 guidance on voluntary affirmative action. Employers have been relying on that specific document for nearly 50 years. Okay, let me offer an analogy here. For half a century, that 1979 guidance was basically a corporate immunity idol. I like that.

[00:13:32] If a company got sued over an affirmative action program, they could hold up that EEOC document in court and say, hey, we followed the blueprint you gave us. We acted in good faith. You can't punish us. And by rescinding it, the EEOC didn't just change the rules of the game. They confiscated the immunity idol entirely. The immunity idol is completely gone. Title VII has a statutory good faith defense, meaning if you follow written EEOC guidance, you are generally shielded from liability.

[00:14:01] By ripping up that 1979 document, the EEOC stripped away the protective armor that thousands of voluntary corporate programs were built on. Man, if the immunity idol is gone in terms like lived experience or being flagged as illegal proxies, how is a manager supposed to know what the boundaries even are? I mean, if the text of Title VII doesn't even mention the acronym DEI, what is the actual legal standard?

[00:14:29] This raises an important question, and it's one that Dr. Westover addresses very bluntly in the source material. DEI is not a legal term. It is a piece of corporate jargon invented by the consulting industry. The law, specifically Title VII, only cares about one fundamental concept, and that is disparate treatment. Define disparate treatment for me in plain English. Disparate treatment means intentional discrimination.

[00:14:55] It means treating an individual differently regarding any aspect of their employment because of their race, color, religion, sex, or national origin. So the label on the program doesn't matter? Not at all. It does not matter if your HR program is called the Inclusive Leadership Academy or the Diversity Accelerator. A judge won't care about their branding. If the actual mechanics of the program involve treating individuals differently based on their demographics,

[00:15:21] if a black employee gets a different opportunity than a white employee, or a female employee gets a different track than a male employee, it is disparate treatment under the law. So the legal theory is totally established. We know the immunity idol is gone. The bounty hunters are incentivized. And the Supreme Court has completely leveled the playing field. What does this actually look like when it hits a Fortune 500 company?

[00:15:46] Dr. Westover's research highlights some absolutely staggering financial settlements from 2025 and 2026. And we need to look at the mechanics of what these companies were actually doing. The fallout is sobering, to say the least. Let's start by looking at IBM in April of 2026. They entered into a $17 million settlement with the Department of Justice over False Claims Act allegations. $17 million? Yeah. That is not a slap on the wrist.

[00:16:16] What specific HR practices triggered a federal fraud settlement of that magnitude? Well, the DOJ alleged that IBM was signing those federal contracting certifications, claiming non-discrimination, while simultaneously running internal practices that violated the law. Now, specifically, the government cited the use of a diversity modifier in executive pay and bonus decisions, as well as mandating diverse interview slates and operating development programs that were limited by race or sex.

[00:16:44] A diversity modifier in pay. So they were explicitly altering compensation based on demographic outcomes? According to the allegations, yes. Now, IBM denied the allegations. They admitted no liability in the settlement, but they paid $17 million to close the investigation. And they agreed to end or heavily modify those targeted programs. A $17 million warning shot to the entire tech industry. But the DOJ didn't stop there, did they? They absolutely did not.

[00:17:13] Just a few months later, in August 2026, Deloitte was hit even harder. They agreed to a $21.5 million DOJ settlement, plus another $1.2 million to the state of Indiana. And remember those Quatam relaters, the whistleblowers we discussed? The ones who get a cut of the settlement? How much did they walk away with in the Deloitte case? There were later in this case was a group called the American Alliance for Equal Rights. And they walked away with $4.3 million of that federal settlement.

[00:17:43] Over $4 million just for pointing out an unlawful HR practice. Exactly. That is a massive financial incentive to scrutinize your employer's policies. What was Deloitte allegedly doing to warrant a payout like that? The DOJ's allegations against Deloitte were highly specific and highly structural. According to the government, Deloitte's business units were receiving monthly, color-coded summaries that track progress toward race and sex goals.

[00:18:09] Senior leaders were allegedly evaluated and their compensation was partly determined by whether they met these specific workforce composition targets. Okay. Imagine being a vice president, right? You're staring at a dashboard that is glowing red because you haven't hired enough of a specific demographic. And knowing your year-end bonus is tied to turning that dashboard green. The psychological pressure to make an unlawful race-based hiring decision must have been immense.

[00:18:36] That pressure is exactly what the DOJ views as unlawful disparate treatment. Furthermore, the DOJ cited specific internal development programs at Deloitte. They were called Springboard and Compass, I believe, which allegedly limited eligibility based on race and sex. Just like IBM, Deloitte denied wrongdoing. But the sheer size of the settlement speaks volumes about how aggressively the government is pursuing these mechanics. It's not just the DOJ and the False Claims Act, though, is it?

[00:19:05] The EEOC itself has been aggressively filing lawsuits. And the examples in the source material really show how widespread this enforcement is. It's not just tech giants or massive consulting firms. The EEOC has cast an incredibly wide net across every sector of the economy. Take Coca-Cola Beverages Northeast, a regional distributor. In February 2026, the EEOC sued them over a networking event. Over a networking event? Yeah.

[00:19:33] The company hosted a two-day event at a resort in Connecticut. It included paid time off and featured executive speakers. Sounds like a pretty standard corporate perk. What was the issue? The issue was that they only invited female employees. Approximately 250 women attended. The EEOC sued the company on the grounds that male employees were denied an equivalent employment opportunity, which violates Title VII. Which perfectly aligns with that Muldrow ruling we broke down earlier.

[00:20:01] Being excluded from a valuable paid networking event at a resort is absolutely some harm to your employment conditions. The legal pieces fit together perfectly. Then you have the media sector. The EEOC filed a lawsuit against the New York Times company. The core allegation there is that a white male editor was passed over for a deputy editor promotion. But passing someone over for a promotion happens every day in every company. What made this a federal discrimination lawsuit?

[00:20:28] The EEOC claims that the hiring manager actively bypassed the standard internal process to bring in an external candidate for the final interview panel. They allege that this external candidate ultimately got the job, despite rating lower on the evaluation panel than the other finalists, including the internal white male candidate. Okay, so it wasn't just a subjective choice. The allegation is that they actively manipulated their own structured process to engineer a specific outcome. I assume the Times fought this.

[00:20:57] They completely denied the allegations and publicly called the lawsuit politically motivated. But regardless of the outcome, it shows that the EEOC is closely scrutinizing whether established hiring processes are being manipulated. And then there is higher education, which brings us to the Washington University in St. Louis case. This one involves a training session, right? Yes. The EEOC sued Washington University over a mandatory DEI training session.

[00:21:24] The core allegation is that the university forced employees into race-separated breakout rooms. Unbelievable. We are going to talk much more about the human element of that specific case in a moment. But it highlights the extreme legal danger of physically segregating employees by race, even for a supposedly temporary training exercise. And didn't the legal industry itself get swept up in this enforcement wave? They did, which is perhaps the most telling indicator of all.

[00:21:49] The EEOC sent targeted enforcement letters to some of the biggest, most prestigious law firms in the world. As a result, four massive firms, Kirkland & Ellis, Latham & Watkins, Simpson-Thatcher, and A&O Sherman, entered into a settlement with the agency. Wow. They didn't admit liability, but they formally committed to strict merit-based employment decisions and ongoing compliance monitoring. Think about what that means.

[00:22:14] When the top employment lawyers in the country, the people whose literal job it is to defend corporations in court, when they decide it's safer to settle with the EEOC over their own internal HR practices, you know the game has permanently changed. Without a doubt. So seeing all of this, the $20 million settlements, the bounty hunters, the EEOC subpoenas, how is corporate America reacting? I have to imagine some companies are just running for the hills.

[00:22:42] There has absolutely been a major retreat among certain corporations. Based on reporting from the Associated Press cited in the article, Meta completely ended its diverse slate approach to hiring and actually dissolved its dedicated DEI team. But McDonald's retired its diversity goals for senior leadership. Target ended its three-year diversity goals. Google rescinded a highly publicized 2020 goal regarding leadership representation. But it's not a universal retreat, right?

[00:23:09] Dr. Westover points out that some massive companies are digging their heels in and holding the line. That is the fascinating dichotomy we're seeing in 2026. While some retreat, others double down. Look at Costco and Apple. In early 2025, conservative think tanks put forward aggressive shareholder proposals, demanding that these companies roll back their DEI practices and formally assess their legal risks. And how did the shareholders actually vote? It was a landslide.

[00:23:37] At Costco, 98% of shareholders voted against the proposal. At Apple, over 97% rejected it. The executive leadership at those companies argued that their practices are entirely lawful, strictly merit-based, and absolutely essential for attracting and retaining top talent in a highly competitive market. Okay, let me play devil's advocate here. I'm a CEO. 98% of my shareholders are perfectly happy with my inclusion programs. My board believes these programs help the bottom line.

[00:24:05] Why should I care if the DOJ is angry? Isn't a $20 million settlement just the cost of doing business for a multibillion-dollar corporation? It's a pragmatic question, but the financial settlements are just the tip of the iceberg. The hidden costs of these investigations are staggering. Even if you ultimately never pay a dime in a settlement, dealing with a federal agency investigation is incredibly disruptive. Because of the audits. Yes. Take Nike, for example.

[00:24:32] They were hit with a massive subpoena seeking internal data on 16 different DEI-related programs. Compiling data on 16 different programs across a global workforce? I mean, that's not a weekend project. Responding to a subpoena like that requires thousands of hours of legal, HR, and data science work. It completely paralyzes the department. And that's not even mentioning the reputational damage of being publicly sued for discrimination by the federal government. But perhaps the most existential threat to a company comes from the new executive orders.

[00:25:02] You're talking about EO 14398? Yes. Under EO 14398, signed in March 2026, if you are found to violate these standards, you don't just pay a fine. The government has the authority to terminate your federal contracts entirely and declare your company ineligible for any future federal awards. A total debarment. For companies in defense, aerospace, healthcare, enterprise software, losing federal contracts isn't the cost of doing business. That is a death sentence for the company. Exactly.

[00:25:31] The macro perspective is crystal clear. The government is cracking down. Millions of dollars are changing hands. And boardrooms are deeply divided on how to proceed. But these aren't just faceless corporate entities bleeding money on a spreadsheet. Right. These lawsuits are triggered by individual human beings who felt alienated, overlooked, or actively harmed by the very programs that were supposed to promote inclusion. We need to look at the psychology of the workers on the ground. We really do.

[00:25:56] Because if these programs are triggering lawsuits, they must be alienating the very people they were supposed to help. We have to zoom in on the human toll of bad program design. And the most striking example in Dr. Westover's research is returning to that Washington University case we mentioned earlier. Right. The training seminar with the race-separated breakout rooms. Let's look closely at the charging party in that case. She was a black senior program manager.

[00:26:21] According to the federal complaint, before the training session even started, she proactively warned the organizers that she would be the only BIPOC, black, indigenous, and person of color, participant on her specific team. So she saw the structural flaw before it even happened. She did. But despite formally raising that concern, she was forced into a race-separated room anyway. And the intent of that room was supposedly to provide a safe space for shared racial identity, right? That was undoubtedly the facilitator's intent.

[00:26:51] But her actual lived experience was the exact opposite. She described the experience in her complaint as profoundly dehumanizing. And the most alarming part of the EEOC allegation is that after she formally complained to the agency about this treatment, the university allegedly fired her. It is a tragic, staggering irony.

[00:27:12] A program designed in the name of diversity and inclusion ends up dehumanizing a black female employee and allegedly results in her termination for having the courage to speak up about it. It really shows how abstract academic theory can completely break down and cause real harm when applied to actual human beings in a workplace. It perfectly illustrates why employees on the ground are starting to rebel against the clumsy execution of these programs.

[00:27:39] And we see this human reaction reflected broadly in the macro data. The Pew Research Center has been tracking worker sentiment for years. And the shift between 2023 and late 2024 is highly revealing. I'm looking at these Pew numbers from the article right now. In February 2023, 56 percent of U.S. workers said focusing on DEI at work was a good thing. But by October 2024, that had dropped to 52 percent.

[00:28:04] Meanwhile, the percentage of workers explicitly calling it a bad thing rose from 16 percent to 21 percent. And if you break those numbers down by demographic groups, the narrative gets even more complex. Unsurprisingly, given the political climate, the share of Republican workers calling it a bad thing jumped from 30 percent to 42 percent. But the most fascinating and perhaps unexpected data point involves black workers. Yeah, this statistic really stood out to me.

[00:28:32] The share of black workers who said their employers paid too little attention to DEI actually fell dramatically, dropping from 28 percent down to 18 percent. What that suggests to researchers is that a growing number of employees across various demographics are experiencing profound fatigue, skepticism, or outright alienation regarding how these programs are actually being executed in their daily lives. Let's dig into the why of that.

[00:28:58] Why does a program meant to foster inclusion trigger feelings of exclusion? There is fascinating psychological research that explains this exact phenomenon. Dr. Westover cites a seminal 2011 study by Plout and colleagues. They ran a series of five detailed studies looking at implicit associations, what people subconsciously believe rather than what they say out loud. They found that white participants implicitly associated the concept of multiculturalism with exclusion rather than inclusion.

[00:29:28] Wait, so when they heard the word multiculturalism, their brains subconsciously thought, this means I am being excluded from the group. Precisely. They felt that multiculturalism was basically a code word for everyone but me. However, Plout's team also discovered a critical nuance.

[00:29:45] When diversity messages were explicitly framed to include everyone, when the messaging made it abundantly clear that the majority group was also an integral part of the diversity equation, support for the initiatives went up significantly. That makes total sense from a human nature perspective. People don't just judge these programs by their noble stated purpose on the corporate website. They judge them by a very basic primal question. Am I welcome in this room? Right.

[00:30:12] If you just slap a generic diversity and inclusion label on something, you might accidentally be triggering defensive exclusionary feelings in a massive segment of your workforce. And it gets even more complicated when we introduce the concept of merit into the psychological equation. I'm flipping through my notes here to a study by Emilio Castilla from 2010. This is arguably one of the most important concepts in the entire deep dive, the paradox of meritocracy. This study completely blew my mind.

[00:30:40] Walk us through exactly how Castilla set up this experiment. Castilla ran highly controlled experiments with hundreds of participants, all of whom had real-world managerial experience. He had these managers evaluate fictional employees for performance bonuses. He presented two identical employee profiles. They had the exact same performance metrics, the exact same tenure, the same output. The only difference was that one profile was male and one was female. Okay, that's a standard setup for a bias test.

[00:31:09] You want to see if they pay the man more than the identically performing woman. But here is the brilliant twist Castilla added. For half the managers, the fictional company they were pretending to work for was explicitly described as a company that strongly values merit and performance. For the other half of the managers, the company was just described generically, without any special emphasis on being a meritocracy. So the first group is psychologically primed. They are explicitly told, we are a strict meritocracy.

[00:31:39] We only care about objective performance. Yes. And what do you think happened? In the company that was explicitly described as valuing merit, the managers actually awarded the male employee a larger monetary bonus than the identically performing female employee. That is wild. Yes. The managers who were explicitly told they were operating in a fair meritocracy were the ones who exhibited the highest levels of gender bias. That is the paradox of meritocracy.

[00:32:06] Just declaring that your system is fair or writing, we are a merit-based organization in huge letters on the lobby wall doesn't make it true. In fact, it does the exact opposite. It gives human decision makers a false sense of moral license. A moral license. Yes. They subconsciously think, well, the company is a meritocracy and I'm a good fair manager. Therefore, whatever decision I make must naturally be based on merit. It completely blinds them to their own implicit biases.

[00:32:36] So what does this all mean? It's like slapping a bright green healthy label on a giant thousand calorie candy bar. Exactly. You eat it and because the wrapper says healthy, you tell yourself you're making a good dietary choice, even while you're literally consuming massive amounts of sugar. Just declaring that you are fair gives you the moral cover to act on your hidden biases without feeling guilty about it. What's fascinating here is that this research perfectly explains why the massive corporate response in 2020 eventually fractured.

[00:33:04] The response in 2020 was largely symbolic. It was about making loud declarations, setting aspirational demographic goals, and rolling out mandatory awareness training. But symbolic fairness fails because it doesn't actually change the underlying mechanics of how decisions are made in the room. You can't just change the symbols. You have to change the structure. We need structural fairness, not symbolic fairness.

[00:33:28] Structural fairness means changing the architecture of the decision itself so that bias cannot easily enter the equation, regardless of the manager's intentions. If symbols don't work and declaring yourself a meritocracy backfires and setting demographic quotas gets you sued by the DOJ for $21.5 million, what does the science say actually does work? That is the big question. Right. If we can't rely on symbols, how do we actually hire people fairly without getting sued? Let's talk about the gold standard for structural fairness, structured selection.

[00:33:59] If a listener takes only one actionable piece of advice from this entire deep dive, it should be this. Implement structured interviews. It is the ultimate tool for fair, lawful, and highly effective hiring. Now, I know you were geeking out over the data on this when we were prepping. Tell me about this massive meta-analysis by Sackett and his team in 2022, because this fundamentally rewrites how HR should operate.

[00:34:22] I am literally highlighting this section right now because it is a monumental shift in industrial organizational psychology. For decades, HR professionals relied on older meta-analyses that suggested cognitive ability tests, essentially standardized IQ tests for the workplace, were the absolute best way to predict a candidate's future job performance. Right. But Sackett and his team realized that those older studies had a massive methodological flaw built into their math.

[00:34:52] What was the flaw? I know it involves a concept called range restriction, but how does that actually distort the data? Well, think about it like this. Imagine you're trying to study whether a person's height matters in professional basketball, but you only study players who are already on the active roster in the NBA. Well, they're all going to be incredibly tall.

[00:35:13] So if you only look at NBA players, height might not seem like a distinguishing factor among that specific elite group, even though it's obviously crucial for the sport generally. Your data is restricted to a narrow range. The older hiring studies overcorrected for this range restriction phenomenon in ways that artificially inflated the predictive power of cognitive tests and significantly undervalued other methods.

[00:35:37] Sackett's team re-crunched the data across decades of research, removing those massive statistical overcorrections. And when they finally cleared the statistical noise, what was the new reality? The new reality is that structured interviews are the single most valid predictor of job performance. They have ability score of 0.42. That beats job knowledge tests, which scored 0.40. And it absolutely crushes cognitive ability tests, which plummeted to 0.31 in the new analysis.

[00:36:05] Think about what that means practically. Sitting down and asking structured questions is objectively better at finding top performers than forcing candidates to take a standardized cognitive test. Yes. And crucially, Sackett's research also showed that structured interviews result in significantly smaller demographic subgroup differences compared to cognitive tests. So you get a more accurate prediction of who will be a great employee with much less adverse impact on diverse candidates. It is the rare genuine win-win in HR.

[00:36:34] Okay, but what exactly makes an interview structured? Because I guarantee every manager listening to this thinks they are already asking good structured questions. Dr. Westover relies on a comprehensive 2014 review by Leveshina and colleagues to define this. An interview is only truly structured if it meets very strict criteria. First, you have to do a formal job analysis. You must define the exact skills required for the role before you ever post a job description. Okay, that makes sense.

[00:37:04] Second, you must ask every single candidate the exact same core questions. No winging it based on their resume. Third, use anchored rubrics. Let's pause on anchored rubrics, meaning you don't just rate a candidate one to five based on their general vibe. An anchored rubric means you have written descriptions of what a poor, adequate, and excellent answer actually sounds like for that specific question, established before the interview begins.

[00:37:31] Fourth, interviewers must record their ratings independently before they discuss the candidate with the rest of the hiring panel. That's to prevent groupthink, right? That's correct. So the loudest, most senior person in the room doesn't just dictate how everyone else votes before they've formed their own opinion. That's the core of it. And finally, absolutely no off-process exceptions. Every candidate, whether they are an internal referral or a cold applicant, goes through the exact same gauntlet.

[00:38:00] This sounds incredibly rigorous, but I can already hear the objections from hiring managers. Oh, I hear them all the time. Right. If I'm a manager and you're telling me I have to read from a standardized script and score people on a rigid rubric, aren't we just turning hiring managers into robots? What happened to trusting your gut? Isn't hiring supposed to be an art? What's fascinating here is that trusting your gut is exactly where bias lives and breathes. Your gut feeling is not some mystical radar for talent.

[00:38:28] It is just a rapid cognitive synthesis of your past experiences, your comfort levels, and your unconscious preferences. When managers hire based on gut or cultural fit, they almost invariably hire people who look, talk, and think exactly like they do. And that's how you end up with a completely homogenous team and a massive cognitive blind spot. More than that, in this new era of federal DEI enforcement, trusting your gut is a massive legal liability.

[00:38:56] Let's look back at the EEOC lawsuit against the New York Times we discussed earlier. The entire basis of that federal lawsuit is that a hiring manager allegedly bypassed the structured process to bring in an external candidate who hadn't gone through the standard steps and then hired that person even though they scored lower on the formal panel. I see where you're going. The plaintiff lawyers aren't looking for a smoking gun email that says, don't hire this person because of their race. Those emails rarely exist.

[00:39:25] They are looking for deviations from the established process. An off-process finalist who gets the job anyway is exactly what plaintiffs' lawyers look for as proof of disparate treatment. Structure protects the candidate from bias, but it also protects the company from liability. If you can show a documented rubric-based rationale for why candidate A was hired over candidate B, you are on incredibly firm legal ground. Google actually uses this methodology, right?

[00:39:53] The Google read.work data is cited in this study. Yes, Google's internal data strongly supports this approach. They mandate pre-made questions and standardize rubrics. And they found that providing this structure actually saves interviewers an average of 40 minutes of prep time per interview. So it's not some bureaucratic time suck. It's actually an efficiency multiplier. And here is the most incredible statistic.

[00:40:16] Candidates who were rejected by Google but who went through a structured interview were 35% happier with their overall experience than rejected candidates who had unstructured interviews. That is amazing. Even when you tell them no, they feel respected because they know they were evaluated fairly on their actual skills. They didn't just lose out because they didn't know the secret handshake or didn't go to the same college as the hiring manager. It destroys the vague exclusionary cultural fit proxy.

[00:40:46] Now, we also need to address the legal danger of diverse slates because structure applies to the pool as well. Right. This is the Rooney Rule concept, mandating that a hiring manager must interview at least one woman or one minority candidate for every open leadership role. For years, that was considered the absolute best practice in HR. But look at the DOJ settlement with IBM. Diverse interview slates were explicitly cited in the government's allegations. Meta completely dropped their diverse slate approach.

[00:41:16] Why? Because if you force hiring manager to add a candidate to an interview slate solely because of their race or sex to meet a corporate quota, you are making an employment decision based on a protected characteristic. Which under the new strict interpretation of Title VII is disparate treatment. So what is the lawful alternative? If we want a diverse workforce, how do we get there legally? Dr. Westover points out that you can and absolutely should widen the sourcing pool.

[00:41:44] Break down the difference between sourcing and evaluation for me. Sourcing is how you find candidates. It's the top of the funnel. It is perfectly legal and highly encouraged to expand your recruiting efforts to historically black colleges or new geographic regions or different professional networks to ensure a diverse applicant pool. That's sourcing. Right. But once those candidates enter the pipeline, the evaluation process must be identical for everyone. No quotas for who makes it to the final round.

[00:42:14] Widen the top of the funnel as much as possible, but keep the filter completely uniform. Sourcing gets them in the door fairly and structure neutralizes bias at the threshold. But hiring is just the beginning. If your new diverse hires are just sitting at their desks without access to the inner circle, we've failed. How do we develop these employees once they are inside the building? Without triggering a DOJ lawsuit over an identity-restricted leadership program.

[00:42:40] Development is the area where companies are most legally exposed right now. Because identity-restricted programs are so easy to spot on a corporate calendar. The EEOC has explicitly stated that access to mentoring, employee resource groups, and specialized training are considered privileges of employment. And going back to the Muldrow decision, if you deny someone a privilege of employment based on their race or gender, you have caused them some harm, which means they can sue you.

[00:43:09] That is the exact legal theory behind the Deloitte allegations regarding their springboard and compass programs and the Coca-Cola beverages Northeast lawsuit over the women-only networking event at the resort. If you have a high-potential leadership accelerator that is only open to Hispanic employees, you are legally exposed. But wait a minute. The science actually shows that mentoring works, doesn't it? We shouldn't just scrap these programs out of fear. We absolutely should not scrap them.

[00:43:37] Dr. West oversight's foundational research by Kalev and Dobbin, analyzing over 800 firms over several decades. They found that structured mentoring programs are incredibly effective. Over a five-year period, mentoring programs resulted in a 9% to 24% increase in the representation of minority women and men in management positions. A 24% bump in management representation? That's huge. It's massive.

[00:44:05] Mentoring works because it gives people access to informal organizational knowledge, visibility to senior leadership, and powerful advocacy behind closed doors. But here is the critical insight for 2026. The mechanism of mentoring does not require restricting eligibility based on identity. Ah. So you don't need a Black mentoring program to effectively help Black employees. You just need a robust structural mentoring program.

[00:44:31] To fix the legal liability, you simply open the program to everyone. You shift from an identity-based design to a needs-based design. Give me an example of needs-based design. Instead of running a women in leadership accelerator, you launch an accelerator for first-time managers or employees transitioning from technical roles to leadership or returning parents. Those are distinct developmental needs that cross all demographic lines. That's a great pivot.

[00:44:58] You state explicitly in the charter that the program is open to all employees. If demand exceeds capacity, you use transparent job-related criteria or a randomized lottery to select participants. I have to ask, though, and I know a lot of HR professionals listening are thinking this. If we make employee resource groups or ERGs open to everyone, don't we destroy the safe space aspect? If a women in tech group is forced to let men join, doesn't that completely defeat the purpose of the group?

[00:45:27] It's a very common concern, and it gets to the heart of the tension between inclusion theory and hard employment law. The blunt answer is the law, as interpreted by the Supreme Court and the current enforcement agencies, says that employment privileges cannot be segregated. But from an organizational design perspective, inclusive ERG charters actually lead to better outcomes. How so? When you open a women in tech ERG to male allies, you aren't changing the topical focus of the group.

[00:45:57] The group can still focus entirely on issues facing women in the technology sector, advocating for better family leave or analyzing gender pay dynamics. But by opening the door, you foster cross-pollination. You educate the majority group on the challenges the minority group is facing. I see. You get the male executives in the room listening to the concerns, which is how you actually get the buy-in necessary to change the culture.

[00:46:20] Furthermore, Plout's research on multiculturalism, which we discussed earlier, shows that when majority groups feel excluded from diversity initiatives, they become resentful and unsupportive. Open access ERGs reduce that resentment. You keep the topical focus, but the door must be open to all. What are the pitfalls here? What shouldn't companies do when transitioning these programs? The biggest pitfall is the fake-out.

[00:46:45] Do not change the policy on paper to say open to all, but keep it secretly restricted in practice. Don't hide the meeting invites or manipulate the nomination process so only certain demographics get in. The DOJ and EEOC look at practice, not just policy. And the second pitfall is panic-canceling. Don't cancel your mentoring programs entirely out of fear of a lawsuit.

[00:47:09] That hurts everyone in the organization, particularly those who lack natural informal networks. Okay. We are mentoring them legally. We are hiring them structurally. But what happens when it's bonus season? This brings us to the quietest, most dangerous, and most expensive area for bias. It's time to talk about money, compensation, and accountability. This is where we look at pay audits and how we measure success without using illegal demographic quotas.

[00:47:38] Let's go back to Emilio Castilla, the researcher who gave us the paradox of meritocracy. He has another study from 2008 that is just brutal to read. His 2008 study on a large service organization documented something he calls performance reward bias. He found that even when minority employees and women received the exact same performance rating as their white male counterparts, they received smaller salary increases. Let me make sure I'm hearing that right. A manager sits down and rates two employees. One is a white man.

[00:48:08] One is a minority woman. The manager gives them both a perfect five out of five on their annual performance review. But when it comes time to hand out the actual cash bonuses, the white man gets a bigger check. The math proved it. The bias doesn't necessarily happen at the evaluation stage. It creeps in at the translation stage, when performance ratings are translated into financial rewards. Why? Because the translation process was highly discretionary, lacked transparency, and managers weren't required to justify their math to anyone.

[00:48:38] That is infuriating. But Castilla didn't just point out the problem. He actually found the solution in a follow-up study in 2015, didn't he? He followed a company with nearly 9,000 employees as they introduced strict accountability and transparency into their compensation system. What does that accountability look like in practice? It looks like calibrated ratings and mandatory explanations. Before ratings were tied to pay, managers had to sit in calibration sessions and justify their ratings against defined criteria to their peers.

[00:49:07] More importantly, if a manager wanted to give a reward that diverged from what the performance rating dictated, say, giving a slightly larger raise to employee A than employee B, despite them both having a score of 4, the manager had to provide a written, documented explanation for the variance. And what was the result? The unexplained pay gaps shrank dramatically. The crucial point here is that the company didn't give anyone a demographic preference or a diversity bump.

[00:49:34] They simply made it harder for factors other than actual performance to shape pay. This sounds like a strong argument for rigorous pay equity audits. It is the strongest argument. Pay equity audits are incredibly legally defensible. You run a statistical analysis controlling for legitimate factors like role, location, and performance. If you find an unexplained gap, you fix it. But here's the key to remaining legally compliant under the new Title VII interpretation.

[00:50:02] You fix the gap for anyone, regardless of their demographic group. If the audit shows a white male employee is being underpaid relative to his peers and performance, you adjust his salary too. The standard is absolute structural fairness. Now compare that scientific audit-based approach to what gets companies sued. We're talking about responsibility structures versus quotas. This is a critical distinction.

[00:50:27] The massive research by Caleb and Dobbin found that responsibility structures like diversity task forces or fair process councils were incredibly effective in improving managerial diversity. Just having a group of people whose job it is to ask, are our processes fair, changes behavior. But didn't Deloitte have an accountability structure? And didn't the DOJ just hit them with a $21.5 million settlement over it?

[00:50:54] If we connect this to the bigger picture, Deloitte's alleged accountability structure crossed the line into demographic quotas. According to the DOJ, Deloitte wasn't just asking, are our processes fair? They were reportedly generating color-coded dashboards tracking race and sex targets and tying executive compensation to hitting those specific demographic numbers. Like, if you don't hire 15% more women in your division, you lose your executive bonus?

[00:51:23] When you tie an executive's personal financial compensation to hitting a demographic target, you place immense pressure on that executive to make individual hiring and promotion decisions based on race or sex. And that is disparate treatment. Here's where it gets really interesting. Dr. Westover cites researcher Lisa Leslie and this concept of false progress. It's like trying to lose weight by just tampering with the numbers on your bathroom scale instead of actually changing your diet and exercising.

[00:51:52] That is a perfect analogy. When you set demographic quotas, you incentivize managers to achieve the number by any means necessary. Rather than fixing the underlying structural issues. Lisa Leslie's research details how managers game the system. You might hit your demographic target by poaching talent from a competitor or manipulating job titles to reclassify someone. But you haven't actually created a fair organization or expanded the talent pool. That is false progress. So what is the alternative?

[00:52:19] If we can't measure demographic quotas, what can we measure to hold leaders accountable? You measure the process, not the demographics. You track process integrity metrics. Give me some examples of process integrity metrics. You ask your analytics team. Are structured interview rubrics being used in 100% of our hires? Is the promotion rationale documented for every single advancement? You run barrier analyses.

[00:52:45] If you notice that attrition is spiking disproportionately in one specific business unit, you investigate the management practices in that unit. You decouple executive pay from demographic quotas. And you tie it to process integrity and team climate scores. Speaking of climate, we have to talk about training because for years the default corporate response to any HR crisis was to force everyone into a three-hour mandatory bias training seminar. What does the science say about that? The science is devastating for mandatory diversity training.

[00:53:14] A massive 2016 meta-analysis by Betz Rukova and colleagues found that attitude-focused training has very weak effects on actual behavior change. And whatever small effects it does have fade almost entirely over time. And there was a massive field study in 2019, Chang and colleagues. They ran a randomized field experiment with over 3,000 employees at a global organization. They tested online diversity training against a placebo.

[00:53:43] The training changed some attitudes among groups that were already somewhat unsupportive, but it resulted in zero meaningful behavioral change for those groups. The only people who changed their behavior were the ones who were already highly supportive before the training even started. So mandatory training is basically just HR theater? Worse than theater. Dobin and Kalev found that mandatory training can actually provoke resistance and backlash, leading to declines in representation. So if training doesn't work, how do you fix the culture?

[00:54:13] You focus on inclusive climate, based on research by Nishi in 2013. Climate isn't built by a seminar. It's built by daily leadership behaviors. It's how a manager runs a Tuesday morning meeting. Do they interrupt people? Do they give specific actionable feedback to everyone or just their favorites? Leaders need to be trained on concrete management skills, not abstract ideological concepts.

[00:54:38] And most importantly, as the Washington University case showed us, you must never segregate your training by race or sex. We've covered so much ground today. We have the tools. Structured hiring, open access mentoring, rigorous pay audits, and process accountability. But how do we ensure this doesn't all fall apart the next time a new executive order drops or the political winds change? Dr. Westover concludes by offering a blueprint for long-term sustainability.

[00:55:05] It requires building a governance structure that unites three specific pillars in your organization. What are the three pillars? Okay. First, employment law. You need legal counsel integrated into HR to assess risk and keep up with these rapid court decisions. Second, human resources. HR professionals know how the programs actually operate on the ground beyond what the policy manual says. And third, data science and analytics.

[00:55:30] You need people who understand statistics to evaluate the evidence, correct range restrictions, and run the pay audits. This sounds great in a textbook, but getting legal, HR, and data analytics all in the same room, that sounds like a recipe for a bureaucratic nightmare where nothing ever gets done. Lawyers say no. HR says we have to. And data analytics says we need a larger sample size. It definitely requires upfront work and strong executive leadership to align them. But think about the alternative.

[00:56:00] An internal cross-functional audit is infinitely cheaper and less bureaucratic than responding to a DOJ subpoena or paying a $21.5 million settlement. Fair point. This United Governance team needs to run every single people program through four rigorous design tests. Okay. Let's list these four tests so our listeners can write them down. Test number one, open access. Is this program, ERG, or event, explicitly open to every employee regardless of protected characteristics?

[00:56:31] Test number two, job-related criteria. Are we selecting participants and making hires based solely on validated, predefined skills needed for the role? Test number three, consistent process. Is every candidate or employee going through the exact same evaluation steps without off-process exceptions? And test number four, outcome monitoring. Are we tracking the integrity of our processes without imposing demographic targets that pressure individual decisions?

[00:57:00] If a program fails any of those tests, it needs to be redesigned. Immediately. Furthermore, leaders need to adopt a new narrative. Based on Plough's research, leaders must explicitly communicate a credible narrative of fairness for everyone. This isn't a retreat. It aligns perfectly with the Supreme Court's stance and aims that Title VII protects all individuals. It tells your workforce that you're building systems immune to bias, which benefits everyone.

[00:57:28] Treat every HR policy as a hypothesis. Continuous learning is the goal. You pilot a new structured interview format. You evaluate the data and you adapt. You build a system that is immune to political pendulum swings because it is anchored in objective fairness and strict legal compliance. So, bring this all together for you or listener. The era of identity-restricted programs and demographic quotas is functionally over. The legal enforcement from the DOJ, the EEOC, and the Supreme Court is too heavy.

[00:57:57] And the financial risks under the False Claims Act are just too massive to ignore. But that does not mean you are abandoning inclusion. By focusing on structural fairness, structured interviews, open access mentoring, rigorous pay audits, and process accountability, you are actually adopting the practices that peer-reviewed science proves are significantly more effective at reducing bias. You aren't retreating. You are putting inclusion on firmer, lawful ground.

[00:58:24] You are rebuilding the foundation so it doesn't wash away in the next storm. Before we sign off, I know you had one final provocative thought about where all this is heading. We've spent this entire deep dive talking about human processes, how human managers use rubrics and calibration sessions to eliminate bias and avoid unlawful proxies. But what happens when we hand these newly designed, strictly merit-based criteria over to AI hiring algorithms? Oh, wow.

[00:58:52] That is a terrifying thought. If we train AI on this new paradigm of proxy-free merit, will the algorithms truly be blind to bias? Or will they just invent entirely new invisible proxies that humans can't even detect in a pay equity audit? As we fix the human system, we might be feeding a machine learning system that creates an entirely new set of legal and ethical liabilities that we haven't even begun to legislate.

[00:59:19] That is a fascinating and somewhat chilling thought to leave you with. The ground might be firming up under our human processes, but the AI tectonic plates are just starting to rumble. Thank you all for joining us on this deep dive. Tomorrow morning, when you look at your company's hiring rubrics or mentoring programs, I hope you question everything. Until next time. Parsons is 뭐 do you want to look at your company's hiring örnis über collecting a new怪step