Bias in hiring has been a topic of discussion for decades, yet our understanding of what actually happens within recruiting processes remains surprisingly limited. Most research focuses on single types of bias in isolation, making it impossible to build a complete picture. Meanwhile, the arrival of AI tools is intensifying the scrutiny of hiring decisions, demanding a level of accountability that many organizations simply aren’t prepared for.
TA leaders often assume they know where bias exists in their processes. But what if those assumptions are wrong? What if the patterns are more complex and counterintuitive than anyone expected?
My guest this week is Bas van de Haterd, Co-founder of the TA Audit Institute. In our conversation, he shares findings from new research that challenges conventional thinking about where bias actually occurs and reveals how much we still don’t know.
In the interview, we discuss:
• Research methodology and sample size
• How this research compares to previous academic research
• What was measured and what was not measured
• What do the results tell us about bias in the hiring process?
• Is bias universal, institutional, or personal?
• Are employers doing better at removing bias than they think?
• Using data to drive targeted change
• Lessons learned and advice to TA Leaders.
• What’s the future direction of the research?
Follow this podcast on Apple Podcasts.
Follow this podcast on Spotify.
00:00
Matt Alder
We talk a lot about bias in recruiting, but how much do we actually know about what’s really happening? With AI related lawsuits now putting hiring under intense scrutiny, how do we close that gap between perception and reality? Keep listening to find out.
00:19
Matt Alder
Support for this podcast comes from Maki. Maki enables businesses to build intelligent, science backed hiring strategies that predict on the job performance and adapt as roles markets change. Companies today face a surge of undifferentiated AI, inflated CVs and often rely on manual screening or inconsistent processes. This leads to missed talent, weaker performance and a poor candidate experience. Maki combines science based assessments, behavioural signals and autonomous AI agents into one integrated engine. It evaluates every candidate, predicts who will succeed, and continuously improves hiring outcomes. Teams hire faster, more fairly and at a lower cost while delivering stronger on the job performance and a better experience for candidates and customers. To find out more you can go to mackypeople.com and Maki is spelled M A K I and that’s makipeople.com.
01:45
Matt Alder
Hi there. Welcome to episode 763 of Recruiting Future with me, Matt Alder. Bias in hiring has been a topic of discussion for decades, yet our understanding of what actually happens within the recruiting process remains surprisingly limited. Most research focuses on single types of bias in isolation, making it impossible to build a complete picture. Meanwhile, the arrival of AI tools is intensifying the scrutiny on hiring decisions, demanding a level of accountability that many organizations simply aren’t prepared for. TA leaders often assume they know where biased exists in their processes. But what if those assumptions are wrong? What if the patterns are more complex and counterintuitive than anyone expected? My guest this week is Bas van der Haterd, co founder of the TA Audit Institute.
02:41
Matt Alder
In our conversation, he shares findings from new research that challenges conventional thinking about where bias actually occurs and reveals how much we still don’t know.
02:53
Matt Alder
Hi Bas, and welcome back to the podcast.
02:55
Bas van der Haterd
Great to be here again, Matt.
02:57
Matt Alder
Now you’ve literally only a couple of weeks since you were last on, but in our chat before we started recording last time, you told me about some more research that you’d done last time. We were talking about career sites, but this is kind of really interesting stuff, so I was really keen to get you back on the show to talk about it before we kind of dive in. Can you quickly introduce yourself and tell everyone what you do in case they’ve not heard you on the show before?
03:22
Bas van der Haterd
Of course. My name is Bas van der Haterd. And for this specific topic, I think it’s best to introduce myself as the co founder of the TA Audit Institute, which audits major corporations on potential bias and labor market discrimination and tells them where it is. And I am also the co host of a new podcast together with Dr. Charles Handler, who’s one of the American experts when it comes to selection. And it’s called the global talent podcast gtp.
03:57
Matt Alder
Fantastic. So tell us about this particular sort of piece of research on bias that you’ve done. Why did you do it and how did you do it?
04:07
Bas van der Haterd
Well, together with my co founders of TA Audit Institute, we figured like, why don’t we look at how much bias is there really? One of the main reasons that we really wanted to do this is to answer the question which bias is most common? So people always complain about age bias, but in the Netherlands there’s been very little real research on that. And is that more common than, for example, ethnic bias? And how does gender bias fall into this? So we’ve never seen an actual research being done on all three biases at once, which makes them comparable. That was my personal reason I really wanted to do this. And then we just needed data to show that there still is bias for our products, to be honest.
05:00
Matt Alder
Cool. So talk us through, how did the research work?
05:03
Bas van der Haterd
Well, the research is actually quite simple. We looked at about 250of the major organizations, both government as big corporate enterprises, and we applied with similar resumes. So resumes that were equally qualified with the only difference being age. So usually 30 and above 50, 55, gender, men and women or ethnicity, which is signaled by the name. So a typical Dutch name versus an Arabic name.
05:38
Matt Alder
And how do you make sure that’s sort of robust? Because it sounds like there’s kind of lots of variables there. What did you do in the design to make sure were isolating the right biases, as it were?
05:49
Bas van der Haterd
Well, those are the only things which really differed. And for the rest, the level of experience was different, only with maximum of one or sometimes two years, if it’s over 20 years of experience. So with age, what we did was you go up faster as a younger candidate and we made sure that everybody fit all the requirements in the job description.
06:21
Matt Alder
And obviously a lot of research comes out of universities, academic bodies, institutes around bias and things like that. What was different about the way that you did this?
06:33
Bas van der Haterd
Well, first of all, we did it so in a similar way for all three biases at once. And especially here in the Netherlands, the academic research only looks at One bias. And because every research is done a little bit differently, you can’t really compare them. The other main difference is that went to the corporate careers websites of all these companies and we didn’t go to indeed or another job board. So we only took the biggest of organizations, the most well known organizations, and went to the corporate career sites. Which also means that if there’s jobs on there which aren’t advertised, and we all know that some corporates have, oh, it would be really amazing if somebody ever applied to this job, but we’re not going to spend money on it, on them.
07:23
Bas van der Haterd
So we had a little bit of a different set of jobs on which we applied.
07:30
Matt Alder
So talk us through the results because I know that you found some of the findings quite surprising.
07:36
Bas van der Haterd
So the main thing is the question that has been unanswered for ages is which bias is most common? It’s the age bias. So 48% of our age related, paired applies, as they’re called in academia, came back not equally treated. So an equally treated means both are invited, both are rejected, both got a phone call. And there’s a certain hierarchy of course, in that gender was second with 45% and ethnicity was 38% which were unequally treated. So that was the first result. Age is the most relevant bias. But the weirdest thing was when we looked at the direction of the bias on both ethnic and gender, there was almost zero net bias. Which means that those 45% of gender difference CVs which weren’t equally treated, literally half of them the men, got preferential treatment and half of them the woman got preferential treatment.
08:49
Bas van der Haterd
Almost the same for ethnic bias. But the weirdest one was actually the age bias where in two thirds of the sample the younger candidate got a lesser response than the older candidate. So the older candidate got invited. Well, the younger candidate didn’t.
09:10
Matt Alder
Let’s start with the age thing. That is interesting because there is obviously, you know, so many people are experiencing age related bias from the older perspective, as it were, and we often hear about it from the kind of entry level perspective. But do you have any kind of insights into what’s going on here with, with those kind of results or what might be happening that’s different to the perception of what we think is happening?
09:36
Bas van der Haterd
The question is it different than what we think is happening or is this specific to the way we did the research? Because we only looked at jobs where the older candidate would be perfect also to apply for? So older CVs are usually sent to people with more senior levels. So a minimum of, I don’t know, 10 years of experience is required, for example. The other thing where I think we differ and I would love to do another research on this, is we made sure that the younger candidates just grew into a certain position. So for example, if we’re applying at manager jobs, the younger candidate took 10 years to become a team lead and the older candidate took 30 years to become a team lead.
10:28
Matt Alder
Yeah, interesting.
10:29
Bas van der Haterd
But they still grew throughout their career. The people I hear complaining about it a lot are people who’ve been at the same job level for 20 or 30 years. And we didn’t really measure that. So there might be some form of hiatus there. Like if you’ve been. We only had a few of them, so it’s not representative. And we saw positive and negative results on those as well. But if you’ve been like, I don’t know, a newspaper editor for 25 years does that, and you and you apply it for a newspaper editor position, how does age discrimination look then? That’s, that’s something we, all of our people kept on growing in their career.
11:22
Matt Alder
Yeah, yeah. And I suppose that, you know, with all the talk that’s out there about skills based hiring in this particular sort of framing of it is kind of really indicating that those employers are valuing either consciously or subconsciously experience over anything else as a sort of defining factor.
11:44
Bas van der Haterd
Yeah. And it’s, but it’s also the way and the sort of experience, I guess, you know, it’s, you can be very experienced, but if you’re only very experienced in one specific thing, are you really that experienced if you haven’t shown any growth? I think they’re looking at it from that kind of perspective. But were extremely surprised with these results, to be honest.
12:15
Matt Alder
And talk us through the gender and the ethnicity part of it as well. What do you kind of think is going on there? Or how have you sort of rationalized the results that you got?
12:25
Bas van der Haterd
Honestly, I can’t rationalize them. It’s just what we see is what we see and we see that the net discrimination was almost none. I actually made at some point a hypothesis, especially with ethnic discrimination, that it might be that the profit companies might be discriminating and the non profits might be positively discriminating, as they used to call it here, the ethnic minorities. But that wasn’t the case either. So I think it, I know that in Fleming, in Flanders, in Belgium, there was a research where it showed that. But with Every doubling of a company size, the discrimination dropped by I think 6% or something.
13:14
Bas van der Haterd
Since we only took the biggest companies and there’s a decent sized non profit in there, it might be that we just got lucky or it might be that for only these specific companies, they’re just not discriminating anymore in net terms because they’re still discriminating. And I, I still think that positive discrimination is as much discrimination, of course, and you should have an equal treatment wherever you apply. But for the entire labor market, we didn’t see this right now.
13:53
Matt Alder
And does it sort of tell us.
13:54
Matt Alder
That we tend to sort of look at bias as, you know, something that’s institutionalized? It’s like, you know, across the whole of a company’s recruiting process. But could this perhaps tell us the influence of individual hiring managers in terms of what happened?
14:11
Bas van der Haterd
The thing is, I don’t know who made the decision if it’s the hiring manager or the recruiter. And the way it’s usually structured, a lot of it is also on the recruiter, I think. Does it have to do with the individuals? Without a doubt. Is it institutionalized? Well, let’s be very Honest, Matt, eventually 44% of all the equal resumes were not equally treated. So there is an institutionalized bias. It just goes everywhere. You know, it’s not a one directional bias. It’s not anymore that women get discriminated against. Apparently men get discriminated against equally. I would love to see, for example, at some point, I know in Switzerland there was a really great research on this where if it’s typical male jobs, women now have decent chances.
15:12
Bas van der Haterd
But in typical female jobs, so the five jobs which are most occupied by women, like childcare, men, don’t stand a chance. We haven’t seen that in the Netherlands. I don’t know how our research would pin that out. We did try to get some in there, like male secretaries and stuff like that, but not enough to say with any certainty was it correct. But the few ones we tried, both of them either got invited or rejected.
15:43
Matt Alder
What are the practical implications of the findings that you’ve got here? I mean, how would you sort of take them back to the employers in terms of advising how they can improve their hiring or what they can do differently?
15:55
Bas van der Haterd
The first thing I want to mention as well is that 47% of all the organizations which we researched did not have any bias. So all of this bias comes close from half of the organizations, almost. The other half is bias free, which could of course because they just all kept rejecting Us, I was recently talking to one company and she said, yeah, you guys, we mentioned the jobs because were evaluating the results with her. And she says, yeah, you only took jobs with 500 applicants each. So, yeah, every one of us got rejected, but then they still, both of them got rejected. So it’s not biased, but over half of them actually are doing a pretty good job, which surprised the hell out of me.
16:53
Bas van der Haterd
And I can tell you, every TA leader I spoke about was surprised as well that they passed the test, so.
17:01
Matt Alder
To say, which also kind of speaks volumes in terms of people having a proper kind of grip on this, if that. If that makes sense. If people don’t know that they’re actually good at not being biased, you know, that sort of indicates that perhaps we just don’t have enough of a kind of a grasp on what’s kind of happening in recruiting when it comes to bias.
17:21
Bas van der Haterd
I think, at least in the Netherlands, the TA leaders I talk to, which is most of them, they are actually so aware that their processes aren’t perfect or aren’t great yet that they are unaware of how much they already improved and how much small improvements can make a difference.
17:42
Matt Alder
It’s kind of a really positive message, isn’t it? Because I think that a lot of the talk of bias in recruitment is always on that other 50%, you know, the process is biased and all these kind of things, but if companies are doing good work on it, then it should be celebrated.
17:56
Bas van der Haterd
Exactly. And I think that especially. I don’t know about other countries, but the TA leaders I know in Britain are pretty much the same. They have been so focused on, we need to do better. We know weren’t great, that they’re now in a mode where they’re like, we still need to do better because I, I haven’t achieved all of my goals. And let’s be honest, 44% of all the applications we sent weren’t equally treated. So there’s still a lot of room for improvement as well. But the companies that are really screwing up, to be honest, only 2% of all the organizations we researched had all three types of bias. Right.
18:42
Matt Alder
Okay.
18:42
Bas van der Haterd
So most of them are doing, or almost half of them are doing, really. Okay, now back to your other question. How do we go about this? Well, start by simply implementing the basic things where we all know a structured system of doing job requirements. We know that actually with a structured approach to looking at CVs already helps a lot.
19:13
Matt Alder
Final question to you. Where does this go next? Where are you taking the research next? How is it going to develop?
19:19
Bas van der Haterd
Well, right now, Matt, we are talking to a lot of companies where we’re saying, like, okay, do you want your own personal audit? And then I can tell you’ve got, I don’t know, gender bias in sales or ethnic bias in it or age bias in marketing. I don’t know what pops up for you. And then you can look at it more specifically. You’ve got data. I have one client, I’m a potential client I’m talking to who said, like, listen, I know I’ve got a hiring manager over there who’s ethnically very biased. And I’ve tried to escalate it a few times to C suite, but every time he has a story why that candidate shouldn’t be hired. And I, as a TA lead, I’m now getting problems with not delivering enough quality candidates because he keeps rejecting everybody with an Arabic name.
20:16
Bas van der Haterd
So with your data, I can finally get this done internally. But, yeah, there’s a lot of companies that are just really interested now to know, like, where are we and how can we now have targeted interventions?
20:35
Matt Alder
Absolutely. I mean, I think it makes perfect sense from that perspective. Bas, thank you very much for talking to me.
20:41
Bas van der Haterd
Always a pleasure, Matt.
20:43
Matt Alder
My thanks to Bas. You can follow this podcast on Apple Podcasts on Spotify or wherever you listen to your podcasts. You can search all the past episodes at recruitingfuture.com on that site. You can also subscribe to our weekly newsletter, Recruiting Future Feast, and get the inside track on everything that’s coming up on the show. Thanks very much for listening. I’ll be back next time, and I hope you’ll join me.
Transcribed by https://fireflies.ai/
