The Science Of Success with Iris Bohnet

0:00Women would get nines, but they just don't get tens because in our minds perfection is male connotated.

0:12What would happen if we just told the truth? Welcome to TruthWorks, where we dig into the nitty-gritty of leadership and work. And what needs to change? I'm Jessica Neal. And I'm Patti McCord. Our journey together started in HR, but trust us, it's evolved into something wild, honest, and well, a bit rebellious. So throw out the handbook. We're here to redefine rules to work for us, not against us. Let's dive into another episode.

0:48Hi, everyone. Welcome to another episode of TruthWorks. I'm Jessica Neal. I'm here with my lovely co-host, Patti McCourt. Hello, hello. And today we have the honor of talking to a very special woman named Iris Bonnet. And she is the professor of business and government and the co-director of the Women and Public Policy Program at Harvard Kennedy School, combining all of her insights from economics and psychology to improve decision-making in organizations, which we all need. So thank you, Iris, for being a guest. We're so excited to have you. I can't help but mention have you at this time. As we talked about before we started recording, we don't want to give a lot of air to what's out in the system, talking about gender equality and DEI.

1:41But this is something that I think you will give a good perspective on no matter what's happening out there in the world, because you have data You know, that suggests and proves likely that these things work and are good, no matter what's happening in the world. So that's what I'm excited to talk about today. So thank you for coming on. Yeah, thank you. Well, thank you for having me, Jessica and Patty. I'm delighted to be here. And yes, data absolutely are my passion. Plus, I love the combination that you have of economists and, you know, putting that perspective on business because Jessica and I are from the HR world.

2:25And sometimes we use data effectively and sometimes we don't. But that beautiful, as HR professionals, and that beautiful combination allows us to be true business people, you know, in the workplace so that we're actually contributing to the business and not just to the people, soft people issues. So thanks for all of that. Yeah. So tell us a little bit about how this all started for you and became your passion. I think it started actually with a man. Doesn't it always? True. Well, so with the dean of the Kennedy School at the time, and he asked me whether I could direct one of the Kennedy School's research centers, the Women in Public Policy program that you mentioned beforehand.

3:16And at the time, I had done a little work on gender, not very much. I had mostly focused on just helping organizations make better decisions, full stop, you know, on negotiation, decision making, trust, other topics. And I would say I've always been a feminist at heart, but I had used gender more as a control variable in my other kind of analyses. And I wasn't quite sure why he asked me and whether I should do this, but I did. And then it took me a little bit, not very long, but a little bit to realize that my old self could actually speak to my new self. That the way I thought about helping organizations make better decisions, in fact, was exactly how I should think about gender equity and that we should make the best possible decisions under all circumstances.

4:06And we could use data to inform those types of decisions. It's fascinating. So was there anything in your early research that surprised you or confirmed what you instinctively knew? Luxury of being a researcher is that I'm constantly surprised. And I think that curiosity is almost a precondition for doing the kind of job that I'm doing. Because if you're not curious and not ready to be surprised, then why do the research? So much of my research looks at what works and what doesn't work. So we're trying to evaluate, really, whether the new hiring procedure or the new promotion protocol really works or not.

4:51And a study, so let me actually give you not one of my studies, but another study that just is one of those studies that you won't forget. So that was a study done by Lauren Rivera and colleagues. weeks, they looked at the scales that we use in our performance appraisal processes. So you're both from HR, so you probably have used a version of this. So many, many organizations evaluate people on a scale from 1 to 10 or 1 to 5 or 1 to 100 or, you know, whatever it might be, A, B, D, C, in any case. So she was nervous that a common scale that focuses on evaluating people from 1 to 10 might lead to gender gaps.

5:35Why? Because her suspicion was that 10 or 100, that in our minds, they feel a bit like perfection. So, you know, 100 % or 10 out of 10 feels like really an A plus, plus, plus kind of thing. Right. Like it's amazing. It's hard to get. Exactly. So she worked with an organization that was willing to do an experiment where they compared the 1 to 10 scale that they had been using to a 1 to 6 scale. So there's nothing magic about the 6, but she wanted to choose a number that in our minds didn't mean very much, actually. And she found that it was much less likely for women to get a 10, but the gender gap basically disappeared when the best grade, so to speak, that you could get was a 6.

6:26Fascinating. so right so so i think um that is it's an interesting example and i'm glad you're surprised too um not all of the research is quite as surprising for women i mean that's fascinating no no women would get nines but they just don't get tens because in our minds perfection is male connotated right but if six is the best yeah so you you can get a six and you can get a It's not, to your point, I guess it's the association with the number. Yeah, it's absolutely the association with the number. And I think that's, you know, I'm a behavioral economist, as you said, and that's the type of design question that we ask.

7:07So we always look at hiring performance evaluations, et cetera, and go deep into the design choices that we all have made, sometimes inadvertently, you know, not really intentionally. And then we're trying out different types of designs that might level the playing field between men and women, but also more generally. Yeah. When I was researching you, I think it was a talk that you were giving, but you were talking about how you have all this data behind how diversity programs really don't work and don't move the needle. I think there is some fairness around what works and what doesn't. It's my experience in HR that we can pretty effectively and make a lot of changes around the hiring process.

7:53But the part you talked about in terms of internal mobility, my experience is those are the hardest things to change because there's the combination between systems and decision makers. and the systems, even if you go to a six evaluation, if I'm going to promote Jessica to a vice president, that's probably going to happen behind closed doors at an executive level, right? And then the decision makers are often men. So, yeah, so talk to us about what you discovered in terms of those kinds of programs and what worked and what didn't. Yeah. Yes. So you both ask two very big questions. So let me try to do this.

8:39Sorry, we get excited. Yeah, right. Yeah, no, I'm excited too. I share your enthusiasm. And I am. So one thing that I sometimes say, Jessica, to your question is that fairness is not a program, but a way of doing things. And so that's really kind of a very important insight that I've come across these last dozen years or so that I've really, really focused on this. Yeah. Well, let's unpack that a little bit because, you know, I think how that plays out in HR organizations and just companies in general is everything has to be fair. And that's the program. And I think what you're saying is a little different.

9:22So walk us through that. First of all, I think we have to make it very clear that currently we don't have a meritocracy. You know, we have about 300 studies. That's why actually I start with hiring, because the evidence is so strong. We have about 300 studies that did the following. And you too probably were subjects in one of these experiments, because what the researchers did, they sent two identical resumes to employers. So they were, in fact, responding to a job ad, and then they sent a resume with the name Susie Smith and another resume with the name of John Schmoe. Or they might have names that connotate ethnic or racial background, or they might indicate someone's religious background in the resume, you know, an extracurricular activity, for example, sexual orientation.

10:14So researchers really have done this type of work on all kinds of dimensions that you can imagine. Age is another interesting one. Parental status is another interesting one. And then they look at who gets a call back and who is invited to an interview by the company. And if we were unbiased, then the likelihood should be about the same because everything else is identical other than maybe your religious background. And we have found, still are finding today, that that's not the case. It's a heroic aspiration to think about meritocracies, but there really isn't a meritocracy without fairness. So first, we really have to level the playing field.

10:57And I just wanted to give some data suggesting that that's not the world we live in right now. And then Jessica, to your question, what is different between a program and a process maybe? Right. So you asked me about diversity training. Why am I so pessimistic? Not because I wouldn't want them to work. They're very attractive. And I am a trainer. I'm a professor, after all. So I'm a teacher. But we actually do have about 500 studies suggesting that they don't change behavior. They can raise short-term awareness. So if you quiz people right after the training about unconscious bias, they are able to tell you what unconscious bias is.

11:39But the real question, of course, is whether that changes behavior down the line. However, and if I may, I'm going to add one more thing here, because I have actually new news, really new news that you haven't heard me talk about, because it's just come out in science. Ooh, exciting. So while I had kind of given up on kind of diversity training and focused much more on systemic change, recently a number of us got together and said, okay, they're still quite popular, and can't we really not make them better? Shouldn't we just, you know, really redesign them? So we tried to throw kind of a lot of behavioral science at a training that we evaluated.

12:21And when I tell you what we did, this will totally resonate with you. So one thing I think that is not working so well with diversity trainings is that they happen at a random moment in time, maybe during onboarding or maybe every September or, you know, so, okay, fine. We checked our box. We did the diversity training. Yeah. Exactly. And secondly, the trainings tend to be relatively generic. So they're not really targeting the types of responsibilities that I have for the job that I have tomorrow. And so these were honestly the two core insights. We did a number of other things, but these were the two core insights that led to the design of a very short training.

12:59So first of all, we asked ourselves, who is our target group? And then we focused on hiring managers. And then we asked ourselves, okay, how can we find a task that they are involved in tomorrow that we could try to impact? And so we chose the moment when these hiring managers move people from the long list to the short list. So they move them from, you know, kind of being screened initially and then decide whether they want to invite them for an interview, for example. And so literally that was the moment when we offered the training. The training was a seven-minute video delivered by a senior manager in the firm.

13:38It was a large 100 ,000 employees, engineering company. And what we then evaluated was whether reminding them of the values of this firm, which included diversity and inclusion, and of the purpose of hiring, which focused on skills, on complementarities of skills and how do you create a well-functioning team, whether that could actually lead them to make less biased choices. And that's exactly what we found. And this company was interested in attracting more women and also more people from outside of the hiring country, which is a little unusual. But this is a multinational firm. They are active in 100 different countries.

14:22And they found that, you know, Swiss originally, so that the Swiss were hiring the Swiss and the Malaysians were hiring the Malaysians. And we really were benefiting from the global talent pool that the firm theoretically had available. So these were the two goals, and that's what we looked at. And what we found, particularly women from outside of the country ended up more likely than being hired. So I think the takeaway here is, you know, trainings do serve a purpose, but we have to think very hard about how we can then give people the types of information and the types of tools that are actually useful for them, ideally tomorrow or next week.

15:02Yeah. And so in this training, it was a training, but it was short. And it really was a reminder of values that this is important. Diversity is important. We're going for that. And that was it. It wasn't like, you know, here's how to look at a resume versus here's not how to look at a resume. It was just stating what was important. Right. Or deeply understand your unconscious bias. What I love about it is, and it's been my beef with training forever, is, you know, it doesn't matter if it's not in the moment, right? Like conflict management only matters when you're in conflict. I mean, it really, you don't go, I'm really mad at you.

15:47Let me run back and look at my binder from the training I took last year. And I, I love that in the moment thing. It's really interesting. The other thing you said that was so important is meritocracy is a myth, right? I mean, that, that with that's something that everybody's talking about now, right? Well, we really don't, We want to be completely biased. We're colorblind. We just, we're gender biased, blind. We just want to look at skills. And that's just not what happens. I mean, again, it's an aspiration. And we should absolutely de-bias our systems, right? We should, getting now beyond programs, my real passion really is in the systems.

16:30We should absolutely get into the systems and de-bias the systems so that people have equal opportunity. And maybe, Patty, that's your question from before. Yes. There's just so much happening in career advancement that is hard for many reasons. I'll give you one example that not many people talk about. And a colleague of mine coined this performance support bias. And that is way before you do a performance appraisal or even a promotion decision, kind of the question, do we all get the same type of support to succeed in our jobs? And I came across this when I worked with a law firm a couple of years ago, and they actually had an interesting term that I then used to write a paper on.

17:13They called this the thin file problem. And they said, look, some of us just have thin files when they come up for promotions to partner. And with the thin file, they meant it looked like these people just hadn't done a lot of work when they were associates in the law firm. And we actually then unpacked the data and could show that this difference started in the very first year, in the first year of associates' life, where partners chose people they wanted to work with who basically looked like them. Yep. So given that the leadership was male and pale and mostly American, mostly straight, so this was an American headquartered law firm, but a global one, that led to those replications.

17:55And some people just were never given a chance to really succeed. They weren't included in the cool deals, didn't meet the important clients, and didn't have the rainmaker as a mentor. Right. So anyway, so that's, I think, an important one. This, how do we make sure people get the same kind of support? So what this law firm actually ended up doing was its centralized work allocation in the first year. That seemed like strange initially, but then was actually accepted quite easily. So initially, partners just would, you know, look at the resumes and choose people. But then they centralize it. They actually take turns.

18:28It's not the same partner every year. And they, you know, they say like, you know, we have 100 new first year associates. And here's how we will allocate them to the different partners. And then the second thing they do, they keep track of the opportunities that people are given over those eight years. So who gets to speak at the law conference and who gets to, you know, work with the rainmaker, et cetera. It's not perfect by any stretch of imagination, but I think it gives us a bit of a sense of what we have to do to equalize the playing field on support. And it's a beautiful example of the best laid plans kind of thing, right?

19:06Which is like one of the beefs I have with HR-based DEI programs is they're always just starts. They're never reflections on. In the software engineering space, you do this thing called code review. and it's when you invent something or develop something, develop a product. And after you launch it, you get together and say, did it work? And it's not emotional. It's not blaming. It's just, you know, what did we try that worked? And what did we try that didn't work? And how do we learn from that? And I think what you said is so important, which is, do we stop and reflect and say, okay, what's the actual data?

19:46Right? It was a great effort. So did it work? Did it move the needle? Did it not move the needle? And then why? And then how do you reflect back on that? That's really interesting. So a lot of our listeners are CEOs, founders, a lot of leaders just in general, and then, of course, a lot of HR leaders. What do you think if they're trying to have a more inclusive workplace and they do have all these programs that aren't maybe moving the needle, where should they focus? Where should they start? So first, I would say, actually building on what Patty mentioned before, I do think a good start would be to measure what works.

20:28Right to A, B test, the way we test, as you just described, Patty, new products, but also the way even non-tech firms test a new shampoo before it is unleashed on the world. So we just need to do more of that just to get a better sense of what works and what doesn't work. So that would be maybe my major appeal. And then, I mean, we can start with kind of unpacking what are the biggest challenges that we're facing today. So I'll give you one that is still a challenge, but maybe not as big as it used to be. And that's the gender pay gap. Still a challenge, particularly in the United States for women of color who make much less than white men, particularly African-American and Hispanic women.

21:12something. I don't want to say this is a small problem, but I just want to tell you we have actually made progress in closing the gap between, for example, a woman and a man for equal work. And that's kind of important to say, right? So that means we're controlling for everything else and we're really comparing apples with apples. And that has shrunk over time. Now, where the gaps are today, if you look at the difference in what men and women earn in this country, It has more to do with seniority, that some of us are in more senior roles, right? Men tend to be in the more senior roles. Women tend to be in the less senior roles.

21:50That's a big driver of these earnings differences. And the second driver is what economists call occupational segregation. And what we mean with that is that it's still true today that men are more likely to choose STEM jobs and women are more likely to choose heel jobs. Maybe not a term that everyone is familiar with, but Richard Reeves introduced the term a while back. He stands for H is for health, E, education, the A for administration, and L for literacy. So we have these different forces push and pull that lead men and women to go into different directions, and they're also compensated differently.

22:29So that's one way to kind of start thinking about, okay, where should I focus if I wanted to kind of narrow that gap? And so career advancement still is just a biggie because that's where a lot of inequities are hidden. Is there any other data that you think our listeners are going to find surprising that, you know, on any of these dimensions, right? I wrote a book about 10 years ago on what works actually. And already in that book, I speculate about a feature that many performance appraisal processes have in common. Namely, many performance appraisal processes ask employees to self-evaluate and then share their self-evaluations with their managers before managers make up their minds.

23:21So I don't know whether that's the type of process you used or not. But I was concerned about that because we just have so much evidence that not everyone is equally as able or willing to shine the light on themselves. And so that could be due to self-stereotyping, but that could also be due to fears of backlash. right kind of maybe not culturally appropriate in my context to do that etc so then i found a few years ago i am working with a financial services company that is headquartered in the us that had this feature but they told me that they had a glitch in the system and couldn't share self-evaluations one year and so for a researcher that's like you know nirvana i'm like oh wow own.

24:05It's not quite an A-B test, but it's almost an A-B test. So, okay, so we evaluated the data, they gave us all the data, and we did find that women, and particularly women of color, gave themselves lower self-evaluations than anyone else. And so that was kind of my target group. So I was interested whether they benefited from their self-evaluations not being shared. So in the glitch year, initially, we didn't actually find a lot of movement. So we totally saw in the data that the managers could not have built their assessments on the self-evaluations. They didn't have them, and the data told us they didn't do that.

24:43But my smart managers, of course, went back to last year's self-evaluations. And so we could show, again, in the data that their evaluations this year were much more correlated with last year's evaluations. So then I tried to outsmart them and we focused only on newly hired people. So, you know, sometimes research is a bit of a detective. So then we focused on the newly hired people in that particular year that, of course, decreased the sample size dramatically, but still. So we did find what we expected in that group of newly hired people. So the gender to race gap, so for women of color, was eliminated in the glitch year for newly hired people.

25:28So then they were evaluated on par with white women and white men. I liked the study because, I mean, obviously I did it. So that's a very unwomanly thing to say, right? God forbid we praise ourselves. Yeah, that's right. Good job, Iris. Yeah, good job. Thank you. Thank you. No, no, no. So I like it because it highlights, you know, how maybe something well-intentioned. Because some of these design choices, including this one, came about in the 70s thinking that, oh, we have to give employees voice. And so it was kind of well-intentioned, but, you know, not really thought through. I think that's the first takeaway.

26:11And the second takeaway is that, you know, data just speak louder than words. It does help to have data tell the tale. I want to make sure that your listeners don't leave the podcast thinking, you know, only people in power can make change happen. It's very tempting when you talk about systemic change and different ways of structure your processes, et cetera, that somehow this is only top down. So top-down plays an important role and certainly helpful if the CEO or leader is supportive and, you know, role modeling the types of behaviors that we want to see. Obviously, very important. However, I want to end with an example of something that was more bottom-up and also introduce a term that you might like.

26:59And that's the story of Ross Atkins. Ross Atkins is a talk show host on the BBC. And he has a team of four people. and he kind of quite literally woke up one morning asking himself who they feature on his show. Meaning he was like, I have no idea about the gender composition, about race composition, about whether we have, you know, he already had, of course, a feeling that probably they weren't representative of the people they serve. So anyway, they started to count and they literally counted on sticky notes. So this was a daily talk show, every evening, little sticky note, tabulating, took like a couple of minutes.

27:39And they learned that, indeed, that was true, that his suspicion was true, that they were featuring fewer women, they were featuring fewer people of color. So it started with gender equity initially, and then it expanded to ethnicity, and then it expanded to disability. And his show did this for a bit and improved. And then he started talking to his colleagues and said, look, here's what we're doing. So it wasn't top down at all. So here's what we're doing. And here's how we think about it, why it's helpful, how it actually improves journalistic excellence, how it, I'm going to go back to the business case, how it might actually affect our viewers, how we might attract new viewers, et cetera, et cetera.