Talent Risk Assessor, 3: Separation Codes, Termination Patterns, and What the Data Reveals

Announcer: Welcome to the Ogletree Deakins podcast, where we provide listeners with brief discussions about important workplace legal issues. Our podcasts are for informational purposes only and should not be construed as legal advice. You can subscribe through your favorite podcast service. Please consider rating this podcast so we can get your feedback and improve our programs. Please enjoy the podcast.

Scott Kelly: Hey, everyone. Welcome back to Defensible Decisions. I’m your host, Scott Kelly, a shareholder in Ogletree’s Birmingham and Washington, D.C. offices, chair of our Workforce Analytics and Compliance Practice Group, and a co-chair of the Pay Equity and the Government Contract and Compliance Practice Groups.

In Episodes 1 and 2, we introduced the Talent Risk Assessor. We walked through how organizations monitor selection and hiring for EEO risk and talked about how good dispositioning and documentation can help support merit-based decisions. Today, in Episode 3 of the Talent Risk Assessor series, we’re going to turn to the other end of the employment life cycle, separations and termination. Legal risk is often most acute here and it’s exactly where proactive analytics can strengthen a defensible position. Of course, doing that under privilege is the preferred way to go. A quick note before we start: this discussion is informational. Any trends you’re seeing on a dashboard by itself would never establish a legal violation.

It’s a starting point for the kind of inquiry we’re going to walk through today. So, joining me again are Chrissy Blantz and Chris Near. Chrissy is a manager of our data analytics group, and Chris is a shareholder in our Columbia, South Carolina office who has defended systemic discrimination investigations from federal regulators for a number of years. So welcome back, both of you.

Chrissy Blantz: Glad to be here.

Chris Near: Thanks Scott.

Scott Kelly: Before we dive in, a quick connection. Episode 2, we look at hiring, which we described as the front door of the employment relationship. So today I guess we’re going to look at the back door, and that’s who’s leaving and why.

Chris Near: Yeah, Scott, I mean this is a real important section of the employment relationship to examine. Separations are among some of the most scrutinized areas in employment litigation, whether that’s a pattern of involuntary terminations kind of skewing towards a protected characteristic or possibly a more subtle pattern of who is voluntarily leaving and why. So, it’s not uncommon for regulators or plaintiff’s counsel to be looking in this section of the employment relationship for evidence of systemic type issues.

Chrissy Blantz: And the analytical framework carries over from hiring. We compare outcomes across groups within defined analysis groups or business units using statistical testing to sort out whether an observed difference is likely chance or something more systemic.

Scott Kelly: All right. So, we’re going to use the same hypothetical multi-state employer and we’re still talking about the same talent risk assessor, just a different surface or area of analysis and that’s going to be separations today.

Chrissy Blantz: That’s right. It monitors separation rates by legally protected characteristics across each business unit and analysis group flagging any comparison where the difference is statistically significant, so the EEO and legal teams know exactly where to look.

Scott Kelly: All right. Let’s walk through our hypothetical. What does the talent risk assessor show us in this instance, Chrissy?

Chrissy Blantz: Scott, as you mentioned, we’re looking at a specific business unit and department within our multi-state employer. The separations dashboard compares overall selection rates by gender and in this hypothetical case, it’s flagged a statistically significant difference. Males are separating at a higher rate than females in this analysis group.

Chris Near: So, Scott, Chrissy, let’s pause on that for a second. Title VII protects everyone based on sex. That includes men as well. So, a pattern where males are leaving at a disproportionate rate is squarely within an EEO issue, particularly under this current administration that warrants some inquiry as any other protected characteristic flag would. So, like I said, this is kind of one of the main enforcement priorities of the current administration that we need to pay attention to.

Scott Kelly: Totally agree with you, Chris. In our hypothetical today, we’ve got a flag, males are leaving at higher rates in this particular group. What’s our next step, Chrissy?

Chrissy Blantz: Like we saw with hires, our next step is to dig a little bit deeper. In this sense, we’re going to disaggregate. We know that not all separations are alike. We have involuntary terminations, those are employer initiated, and then voluntary terminations, which are employee initiated. And then we look at the likely drivers, risk profile and the path for handling each because they are different in how we would approach them.

Scott Kelly: Let’s go ahead and drill right in. What happens when you split involuntary from voluntary termination decisions?

Chrissy Blantz: This is where it gets interesting. When we isolate the involuntary terminations, those are the discharges, the reductions in force, performance-based separations. We see that the gender difference is not statistically significant. Here we’re seeing male and female rates are essentially comparable in this analysis grouping.

Scott Kelly: So, the employer’s own term decisions aren’t driving the overall flag here.

Chrissy Blantz: That’s right. But when we isolate voluntary terminations, the statistically significant flag remains. So, this tells us that males in this group are voluntarily separating at rates that exceed what we’d expect, and the statistical test confirms that it’s unlikely to be chance alone. That tells us where to do some digging, not what’s causing it.

Chris Near: And kind of outside of the numbers and talking more about the actual kind of digging in of the way a company might look at this, keep in mind that organizations often can assume that, well, it’s a voluntary termination, so it must be low risk because the employee chose to leave us instead of us making some decision to cause the employee to leave. But a persistent pattern by a protected characteristic, involuntary separations might still raise questions that are worth investigating. That why can be just as important as the action itself. So, things to be looking at, when employees are leaving, are they saying it’s because maybe there’s better opportunity somewhere else? Why might that be? Maybe they’re leaving because of some type of problem at work. I need to dig in. What is it? Who does it concern? Is it due to some manager behavior? Is there no clear path for advancement or is there a concern that’s raised about low or unfair pay?

When was the last time the company did a pay analysis? Is there some other…it could be a number of other potential issues that the company may need to look at that really aren’t specific numbers driven.

Scott Kelly: In this hypothetical, even though nobody was fired here, there still can be legal exposure, Chris?

Chris Near: Yeah. I mean, potentially. Depending on the facts, the patterns like this could implicate theories ranging from maybe constructive discharge to some broader systemic disparate treatment claim. So, it becomes important to look at the totality of the evidence. That’s what matters. And practically, it’s often harder to defend because organizations sometimes don’t document voluntary departures the same way that they might document an involuntary termination decision. So not to mention it could also signal other risks happening in this group that might have other legal implications. So, all of this is something that you need to be looking at.

Scott Kelly: All right. So voluntary terms are what’s driving the flag here. The next question I suppose is why. The separation reasons are probably where we need to go to. Are those actually coded in the system?

Chrissy Blantz: The talent risk assessor lets us drill into separation reason codes generally. Like we saw on the hire side, we were able to dig into dispositions. We think of separation reasons in a very similar way. These would be reasons like someone leaving for another opportunity or compensation, maybe they’re choosing to retire, maybe there was misconduct or attendance policies that they violated, reduction in force and temporary employment, meaning a role that was time limited by design. And in our hypothetical, that last category turns out to matter here.

Scott Kelly: What we’re really dealing with here is we’re going to run the reason level analysis on the employer’s data. So, in this hypothetical, Chrissy, when we did that, what happens?

Chrissy Blantz: One category stands out, temporary employment. So, we’re seeing that a disproportionate share of the male voluntary separations are coded that way, meaning these individuals were in time-based contract roles that ended on a predetermined date. Statistically, that category is doing most of the work behind this overall flag.

Scott Kelly: All right. So, this isn’t really a story about dissatisfaction. It looks more like a story about who ends up in roles with a built-in end date.

Chrissy Blantz: That’s what the numbers are telling us. When you look at temporary employment and you pull those out of the analysis, and then we see the statistical difference of the gender difference in voluntary separations drops substantially, that tells us it’s a strong indicator, though on its own, it still doesn’t tell us why more men land in those roles in the first place.

Chris Near: Why? That’s the big question, right? And it’s precisely why we dig in here, why we ask questions, and we don’t immediately jump to a conclusion when a statistical flag pops up. For our listeners, if you’ve listened to either of our first two podcasts, you’ve heard us say this already, but it bears repeating again, a statistical flag is not a conclusion. It’s a prompt for a legally-informed inquiry. So, when we get that indicator, the next question based on the information we have so far seems simple. In our hypothetical, is there something in how people are assigned to these time-based or contract roles that warrant a closer look at policy or practice that the company has, or is there some kind of legitimate non-discriminatory explanation for who ends up in those roles? We don’t know that yet, and so that’s why we need to keep looking at this more closely and asking these questions.

Scott Kelly: All right. Thanks, Chris. So practically, what might the organization actually go ask then?

Chris Near: In this scenario, a couple of things possibly, and none of them assume an answer, right? It’s just we continue to ask these questions and kind of continue to go down those paths. One question might be, are temporary role assignments made on objective job-related criteria or is there enough manager discretion to let bias in unintentionally? Other things you could ask, do people in these temporary roles, do they get the same access to training, development, conversion to permanent positions, things like that, regardless of sex? And is the organization’s use of temporary employment structured in a way that could create some form of adverse impact even without anyone intending it to be so?

Chrissy Blantz: And from the analytics side, you look upstream. Who’s being assigned to these roles in the first place, what the temporary pipeline into temporary positions even looks like, and whether the conversion rates from temporary to permanent employee status differs by sex? That’s how you turn an open question into an answerable one.

Scott Kelly: So, a separations analysis has led us back to an assignment and to a placement question without us even concluding anyone did anything wrong here. That is something that federal enforcement agencies like the EEOC and the OFCCP, when they were doing these types of full audits, would focus in on. I actually saw a recent case announced by EEOC impacting women as victims regarding assignments, placement, and compensation. There’s a lot of monitoring that the employer that settled that case has agreed to with a lot of implications for down the road, both in hiring and compensation.

Chris Near: Yeah. And Scott, I think everything you’re saying there just kind of leads us to what is the value of drilling down and asking these questions? Why is it important for us to do this? Because really if we stop at that top level number, we might overreact to an issue that turns out to be a non-issue, but also if we stop at just one area of separations and ignore the others, for example, we’re just going to ignore these voluntary separations, we might wave it off as just an employee’s choice and never get to the why the employees making that choice. Drilling down into the reasons for these actions and the results gets us to a place where we can come up with a story, come up with our answers and figure out what, or if anything, needs to be an appropriate plan of action.

Scott Kelly: Okay. So, before we wrap up, can we zoom out for a second here? Because temporary employment seems to be the driver in the hypothetical, but the model isn’t limited to that one category. What else might organizations watch out for?

Chris Near: Well, really, probably all categories, right? Every reason or category of reasons really carries its own risk profile. Misconduct and attendance driven separations might look one way, but they might also mask a manager issue or an accommodation or a scheduling issue. Reductions in force. Those are heavily litigated and demand very careful adverse impact review of the selection criteria for those lists. Even patterns of who leaves for another opportunity, for compensation reasons, or even for retirement are worth a look. Temporary employment happened to be the story in our hypothetical today, but another employer’s flag might trace back to a completely different category and tell a completely different story.

Chrissy Blantz: Exactly. And the talent risk assessor monitors all of these changes at once, so the organization has real-time visibility into emerging patterns instead of waiting for a complaint or a lawsuit to force a retrospective look back.

Scott Kelly: Which is really the defensibility theme of this whole series, right? Having the data, asking the right questions, documenting the review is what creates a defensible position and the talent risk assessor really kind of helps us help clients get there.

Chris Near: Yeah, that’s right. And I mean, whatever the inquiry ultimately finds, whether it finds a legitimate business explanation or it finds a practice worth changing, it illustrates that an organization took proactive action. It looked at these issues; it identified if there was a flag. And if there was a flag, didn’t just stick their head in the sand, but investigated it under privilege, documented the review, and took appropriate next steps. So that’s a far stronger defensible position than if somebody had just never looked at anything at all and first learns of it when it’s raised by some third party.

Scott Kelly: All right. Very helpful. Let’s recap here. We started with the separation flag. Males were separating at a statistically higher rate in a specific group. We drilled into involuntary versus voluntary reasons and found the driver was on the voluntary side. Then we examined separation reasons and found temporary employment was doing most of the work behind the flag. The takeaway here was that separation analysis isn’t just about who gets fired, it’s about the full spectrum of how and why people leave and whether a pattern points to a legitimate explanation or to a practice worth revisiting. As we said earlier, that’s an EEO and a legal risk question, not a verdict. The same discipline applies whether the driver turns out to be the temporary employment, misconduct, attendance, reduction in force, or something else entirely, connecting it back to the selection and documentation practices we covered in episodes one and two of this series.

In Episode 4, the talent and progression and retention episode, we’ll follow the same employer into promotions, tenure, and the patterns that determine who advances and who stays. It’s going to be the same framework, the same rigor, just a different point in an employment life cycle. So, thank you, Chrissy. Thank you, Chris. Really appreciate you both being here for these first three episodes of the Talent Risk Assessor. And for our listeners today, thank you for sticking with us on what we hope you’re finding to be an interesting topic. Until next time, I’m Scott Kelly at Ogletree. Thank you.

Announcer: Thank you for joining us on the Ogletree Deakins podcast. You can subscribe to our podcast on Apple Podcasts or through your favorite podcast service. Please consider rating and reviewing so that we may continue to provide the content that covers your needs. And remember, the information in this podcast is for informational purposes only and is not to be construed as legal advice.