Ep 96: Harvard Economist Roland Fryer on the Truth Behind Police Shootings & Using Data to Supercharge Meritocracy

0:00We watched cities in protest over these videos and I wanted to do something about it. Everyone had to be thinking, is this really the norm in America? And so I collected millions and millions of data points. For officer-involved shootings, we found no racial differences and no bias. And that's the part that people got really upset about. It was not only the media, but my own colleagues. The backlash was quite tremendous, but I'd do it again tomorrow.

0:35Roland's one of the most interesting public intellectuals of our time. He's also an economist at Harvard. We don't want to hold that against him. Roland's someone who's curious about the world. He goes where the data takes him, and he's found some amazing results in education, in labor, and other areas like police use of force. He's a really courageous thinker, and he's building some important things. There's a big debate over DEI and meritocracy. Roland believes we could actually use data to solve a lot of these problems rather than just sit here and yell and talk about it. Let's actually figure out what the data says and let's apply that and make the world work better for everyone.

1:07Let's chat with Roland about it. I'm Joe Lonsdale. Welcome to American Optimist. Really excited to have Roland Fryer with us here today. Roland is a professor of economics at Harvard. Despite that, he's also one of the most creative and interesting minds around these days. Senior fellow at the Manhattan Institute and also an entrepreneur these days. Roland, thanks for joining us. Thanks for having me, Joe. It's going to be fun. I'm excited to talk about your new company Sigma Squared in a little bit, but I want to start with your background for our audience, you know, a little bit. You grew up, I think, in a really challenging situation in a broken family.

1:36You know, how'd you make it out? I don't know how I made it out. I had a couple of things, maybe, Joe. I had a lot of people who took an interest in me. My grandmother was a schoolteacher and was phenomenal, even though she wasn't my parent. She liked to say she brought me home from the hospital and she did. And I had teachers along the way, but also, and I hope this doesn't sound too self-absorbed, but I was a big dreamer, man. You know, I had things I wanted to accomplish in life. And there were people in the neighborhood I grew up in who were like, you know, to show how tough you are, you should go to prison.

2:14And I was like, that doesn't make any sense to me. Like, why would I want to, I got customer service needs. Why would I want to do that? And so I really thought I was going to be good at something and I was willing to put in the work to do it. And it started out as a little toy athlete, you know, playing football at age five years old. But that gave me the confidence to know if you really put in the effort, you can be great at something. And so throughout life, when things came up, you know, my father went to prison, et cetera. People told me I wasn't going to amount to much. I just knew that probably because I was a dreamer that big things were possible if I put in enough effort.

2:56So Roland, you went on to graduate Manicum Laud from UT Arlington while you're working at McDonald's drive-thru. And you earned a PhD in economics from Penn State. By age 30, you'd become the youngest black-changer professor in Harvard's history. How's that possible? What led to that? You know, one tiny thing. When I was in college, I was working as an intern at Verizon then, but I did work at McDonald's drive-thru, Pizza Hut 2, all of them. I had the worst jobs America has to offer. The worst one was I was a waiter at Golden Corral, which is horrible, a waiter at a buffet, right? You make half a minimum wage and no tips.

3:32It was terrible. But I did knock out employee of the month. In any event, when I walked into my first economics course, literally economics 101, it was taught by Ryan Amacher. He was the former president of the University of Texas at Arlington. I tried sociology, I tried other things, nothing sang to me. They were all so complicated. My first sociology class, they said, oh, the world is interlocking chains of oppression. And I was like, that's really complicated and depressing. I don't know how to think about that. Economics was people are self-interested and they maximize their own revenue. I said, now that makes a lot of sense to me.

4:13They maximize their own happiness. In any event, this professor had office hours at 7 a.m. The class was at 8 a.m. Now that I'm a professor, I know that was a joke. But as a student, I showed up every Monday and Wednesday and Friday at 7 a.m. to debate with this professor. We talked about government's role in the lives of poor people in America. We talked about optimal tax structures. We talked about optimal inequality because according to him, it wasn't zero. And, you know, having grown up where I was, that was unheard of. And I fell in love with economics. And so the way that I was lucky enough to accomplish what you described was because as soon as I went to that class, I was in a big hurry to catch up.

4:55I really, I, you know, I, that was my first economics class and I then sped through undergrad and graduated in two and a half years and I did my PhD in three. So five and a half years from, you know, kind of hanging out. Were you in a rush? Cause it was expensive also, or what was, what was the rush? I was in a rush because I was enthralled by the actual subject matter and I, I could see the power in it. Um, and, uh, you know, I was a nerd, so they were paying full freight for me. I wasn't in a rush. Back in those days, Joe, I made as a graduate student at Penn State, I made$15 ,000 a year. And I remember the first time I looked into a bank account and I had $3 ,000 in it.

5:35And I thought, what could happen to me? Like I have$3 ,000, you know, there's nothing that can happen to me today that would cost this amount. I, this is pure profit. So no, I wasn't in a big hurry because I thought I was rich,$15 ,000 a year, but I was in a big hurry to use the tools of economics to actually increase social mobility. Because what I noticed was these tools were beautiful, but used for silly things. And the things that were most important for American optimism, like how do people get jobs that are equated with their skill level? How do you actually make true equal opportunity in America.

6:17Those things were led to politicians and other pundits to debate, and we weren't using the brilliant tools of economics to study those. And that's what I wanted to change. I was in a big hurry to change that. You really did kind of ride all the cages when you got to Harvard, trying new things, exploring and studying that. One of my favorite studies you conducted, I think with one of our mutual friends is sponsoring it as you paid kids for better grades, right? You tested what works get kids better grades. I think you spent like$10 million in the inner cities, but what were the results you found?

6:44Yeah. You know, uh, there were a few, few sets of results. You may be the only person who liked that. My grandmother didn't even talk to me when I was doing that. Um, so you know, look, I didn't meet many adults that liked it except for you, but I never met a kid that didn't. Uh, and so, but the actual statistical results were very positive when we paid kids to, so we did two types of experiments. One, we paid them for inputs, right? So we paid them to do, to do specific behaviors or accomplish very specific things that we wanted them to do. Do your homework, come to school, don't beat people up, stuff like that.

7:18When we did that, it was wildly successful. So I was having a conversation yesterday with my old friend, Michelle Rhee. We worked together in DC doing those experiments and she was remarking, yes, they were wildly successful. On the other hand, when we designed experiments exactly as economic theory thought we should, which is you pay for the actual output, not the input, we got nothing. And so what we realized was that incentives can be extremely powerful. And we even measured what the kids spent their money on. And so it was interesting. The kids actually saved and it was really an interesting experiment.

7:53They'd actually save the money. It's a, it's a, it's a extremely high return investment in kids is to pay them to do specific behaviors, like reading books and doing their homework. One of the reasons I really liked it, other than the fact that it works to me, you were showing something that people all need to see is these kids in the inner city who were failing and suddenly succeeding. Clearly, they weren't failing because they were dumb or because they were incapable. It was clear they weren't interested or didn't care or didn't think it mattered to get good craze. I thought that was a really interesting insight, right?

8:21The incentives weren't there for them, right? And it was interesting. An 11-year-old really schooled me one day in New York City, and I think it was in Crown Heights. And I asked him, why do you like the experiment? And he says, oh, I said, is it the money? He says, no, it's not about the money, man. He said, although it's nice. He said, in every other aspect of life, I'm basically need to be an adult. But when I come to school, I'm not treated like one. And so for him, the incentives, this really was his job. Now he could focus on it. And I never had, I mean, you know, sometimes you do the right thing for the wrong, for lucky reasons.

8:55And I never thought of that as a mechanism, but these kids really liked the idea of making this what they were truly focused on in their occupation. We do that in middle class and upper middle class households all the time, but in the inner cities, not so much. And so I was, I think that was one of the reasons it worked well. I love it. I grew up with my dad paying me for good grades. So to me, it's all intuitive, you know, that's weird. I grew up with my grandmother whipping me for bad ones. We can't do that in schools. We used to do that. I want to ask on the top of the education, because you've done all these studies and found lots of cool things.

9:29Is there anything else that strikes you as like a really cool insight that you could share? It's one of your favorites, you know, from all the education stuff you've done. Oh man, I think my favorite work in education is the work we did in Houston. You know, I think it, and there's two basic insights there. One, for the first half of the work I did in education, Joe, I was really focused on IRR, which is really important. And, you know, did this work or not? I wasn't really focused on, is this something that can fundamentally close the gaps in American education. Because incentives are great, but let's say the gap between poor and rich is one standard deviation.

10:09Incentives will give you a quarter of that, but you've got three quarters left over. So what we did that I'm really proud of is we went around to charter schools, some successful and some not, and we used their flexibility and their natural variation and their policies and what they do with kids and their pedagogy, et cetera. To extract from that five things that explain 50 % of the variance in what makes some charter schools good and other charter schools not so good. And we didn't stop there. We took those five things and we put them in a randomized experiment in Houston in the worst 20 schools there.

10:44And what we showed was that in three years in the secondary schools, you could close the racial achievement gap in math. And in five to seven years, you could close it in reading. And so this is something where we distilled down the best practices from the innovation that's happened in charter schools and then put them in regular old public schools without a lottery, with no parent consent, and demonstrated that you can actually turn these schools around and fundamentally change the lives of kids. So it just showed me that what's possible in American in education if we have the courage to do it.

11:19That's amazing. Where can people read about that if they want to go read about it? Yeah. Hopefully we'll link to some papers on my Harvard website and it's there. It's called Injecting Charter School Best Practices in the Traditional Public Schools. That's amazing. Well, we'll definitely link to that. I want to go over one more area study of yours that was even maybe more controversial, if that's possible after the education work you did, which I love. The most noteworthy probably was your study on police use of force, right? And this was in 2016, shortly after the shooting of Michael Brown. I apologize, you probably told the story a lot of times, but can you reprise it briefly for us?

11:52What did you discover about the police use of force and why did it piss people off? Yeah. Well, you know, look, I was upset because I was watching just like you and everybody else, these videos come out and we were watching this and everyone had to be thinking, is this really the norm in America? You know, and we watched and we watched cities in protest over these videos and I wanted to do something about it. Now, as you know, I'm an economist, economist, and I believe in comparative advantage. So protesting is not my comparative advantage, but I'm a data nerd. So I decided, let me see, let me go collect some data and I will demonstrate that police are biased and then, you know, that'll help, that'll help this movement along.

12:41And so I collected millions and millions of data points, uh, both on the lower level uses of force, putting your hands on a suspect, throwing them up against a car, maybe hitting them with a baton and other physical types of force that are not lethal. And then we also collected a lot of data on actual lethal uses of force, uh, officer involved shootings, for example. And what we found was that on the lower level uses of force, there are racial differences that we couldn't explain. And we put them through formal bias test and there is bias that goes on in the lower level uses of force. What we didn't find was any that that bias actually was still there as the force ratcheted up.

13:25So for officer involved shootings, we found no racial differences and no bias. And that's the part that people got really upset about. I didn't expect them to get that upset about it. But it was not only the media, but my own colleagues. And, you know, but we put the paper out and the backlash was quite tremendous. But I will say I'd do it again tomorrow. Well, it's definitely courageous. I know you had people even inside of Harvard who went after you, caused you a lot of trouble, disinvited you from some things all around this. And I think one of the ones who went after you, she ended up being the president and then got thrown out.

14:04So that must have been interesting to watch after all of that. That happened? I didn't even notice. Didn't even notice. Let's go to the DEI debate. We'll skip some of that. That's good. As a transition to Sigma Squared, I want to ask you about a 1995 study on the difference between black and white wages in the workplace. Can you explain the study and how it impacted you and your work on these issues? You know, I grew up in a lot with my grandmother and my grandmother was, I think this is important. I'm sorry, it's a roundabout way, but my grandmother was born in 1925 and she integrated Florida public schools in 1969.

14:40And, you know, her first day of integration, Joe, she was spit on. And I only say that because she had very different views of race relations than I did. you know, thank God that the country was in a different space when I was a kid so that I had a different purview. And so that's one of the things we would argue about a lot. But I will admit I was inculcated into the view or I was just, that's how I was raised, that discrimination was omnipresent. And it was, you could look someone in the face and understand if they were biased or not. Although we didn't use words like bias back there. And so when I got to graduate school and started thinking about what to work on, I was really interested in testing some of my grandmother's theories, those things that I sat on the couch and sipped sweet tea with her for years.

15:30I was interested in that. And one of them was the prevalence of discrimination in the labor market. And I stumbled on a paper in 1995 written by William Johnson at the University of Virginia and Derek Neal at the University of Chicago. And what they wrote was that, what they found was that once you accounted for pre-market skills, that the racial gaps went from something that looked like 33%. So before this paper, what people did was they ran a bunch of regressions without great controls, meaning they weren't accounting for context in the right way. And the result was a 33 % difference in wages between blacks and whites.

16:15That's a huge amount. So for every dollar a white person makes, a black person makes 67%, even when they're equally qualified. But the issue is it's hard to understand if someone's equally qualified. And so what Neil and Johnson, they had a brilliant idea. The idea was what if we accounted for and only compared people who went into companies with the same skill set. Right. And so that skill set was obtained before they went to the company. So they're every. So it's really apples and apples at the time of hiring. Yeah, that's oversimplifying, but that's basically it. And what they found out was that that 33 percent went down to seven percent.

16:52Wow. OK. And in other words, they would say if you're really interested in racial inequality in America, you've got two options. You've got two margins. Economists would say to think about the developmental margin, which is, should we spend resources on K-12 education, making sure people obtain skills? Or on the employment margin, should we spend our energies on regulation and making sure that employment practices are unbiased? Okay. Well, their results would say, look, if 80 % of the effect is happening before people ever hit the market, that means we need to spend more time in K-12 education.

17:30And it's because of that paper that I spent so much time in education reform, trying to ensure that people get to the labor market with the same skills. And less of my time, although I spend considerable of it, figuring out innovative ways to ensure that we don't have bias in companies. So the paper is really, really important because it says two things. And typically in America, people say one or the other. The two things this paper says, just to repeat, is yes, there is bias in the world, but the vast majority of disparities is accounted for because of K through 12 skills. This is more department of education where it run correctly in the department of labor, basically.

18:07Like we need more, more training and less, less going and harassing people. Right. And both of them are statistically important, but one is a lot more important than the other. That makes a lot of sense to me. I want to, I want to ask you one of the things that really has been bothering me lately, Roland, we're trying to study how to make government better. You probably see there's a huge election going on in the background is debates about all these things. And, you know, my friend Elon has gotten really passionate about making government, you know, less broken and wasteful amongst others. And one of the things I studied, you know, there used to be tests in the government starting in 1883 for almost a hundred years.

18:39We had these tests that are merit-based. And the idea was you can only run certain departments if you can get certain scores, right? And so there's, I think there's a hundred point test called the PACE exam in the 1970s. And it turns out you need about a 90 score at least to run something really senior. And as of the late 70s, there's about a single digit percent of people who got that really high score. But unfortunately, it was only 0.6 % of the black applicants that got that score. And so the courts, they got rid of these tests because they said that's disparate impact. I'm curious how to think about that and disparate impact in general, because it seems like this disparate impact thing has stopped us from having a lot of these merit-based tests in general.

19:16To a lot of us, garbage just got stupider over the last 40 years, because you can't even quiz people anymore. And yet people seem to be still obsessed with disparate impact. Are they correct? Like, why is that? Yeah, it's not how I would think about it. I would say, look, it's not whether or not there's a disparity in the test scores that should cause you to throw out the test. Although they've been doing this, as you say, forever. They continue to do it. I remember years ago that on the SAT, they threw out the following test because it was biased. Seven is an integer. Could you pick out the other integers?

19:46And they threw that out because they had disparate impact against non-minorities. so it's just it's just a culling of test scores to make us all look like to blunt the instrument here's what i would do i would say look um is the test equally predictive for the groups yeah yeah that's what matters and if not then maybe you want to use different instruments for the different groups right i know that's controversial but the idea is what you're trying to do in your mercy in the most earnest way is figure out what predicts success right That's what these instruments are for. And so if you told me that, look, there were really, really skilled black people who scored very low on this test in aggregate at scale, not two people, but on on average that that was true, that it was essentially zero correlation between the test and a objective measure of productivity for black people.

20:40But for white people, it was different. Then I would say, well, this test is not a good test. Let's find another instrument that's correlated for both groups. But that's really important. I get really nervous, Joe, when people say when the first thing, when a thermometer tells you you're cold and your first action is to throw out the thermometer. What we need to understand is whether or not it's actually accurate. Yeah, that's a great way of putting it. And the worst benefits of the time, some of these tests did seem to be predictive, but that's a great point. You need to really confirm are they predictive for each of the different racial categories as well, because it might be different depending on the culture you came from.

21:08That kind of takes us into sigma squared a little bit, because that's something that's kind of tied to what you're trying to do now. You're trying to predict how people will do in their job performance. So we have a lot of people, especially big corporations, talking about doing DEI, talking about doing diversity. I guess what they really care about with hiring, with promotions, is how people are actually going to do on the job, how they're going to perform, how value they're going to create, right? So how do you approach this? In the simplest but pretty data-driven way, what we're trying to do is supercharge meritocracy, right?

21:35Because at a high level, and I'll get into the details if you want, but at a high level, if companies have bias, then meritocracy is completely consistent with a DEI strategy. Because if you're getting to your talent frontier and you have bias, you're removing that bias and you're saying, look, I'm just going to hire the best person for the job. then you're getting higher productivity and not or diversity. That's ironic. So if you assume the company actually is like super racist, then the DEI would be working better. That's the point, right? Then you don't have to call a DEI. You can just say, hey, I'm trying to ensure meritocracy in my hiring.

22:19And so that's what we're working with large corporations to do is to supercharge that meritocracy. And the way we're doing it is understanding for each company their specific success phenotype, right? And, you know, you know this better than anybody, but I don't care what industry you're in, whether you're at Harley Davidson or you are a an AI company or some other type of analytics company. You are a in the people business. Right. And so what we're doing is we are helping companies predict who will be successful before they ever start in a really data driven way. right and so we are understanding the types of folks who have been uh um successful in the past by role by region all sorts of things like that and then when people apply we're able to give them a screener score immediately instantly to say hey this guy joe applied and here's what we think his expected tenure is in the company and his expected performance how are you getting this data your data to find success traits right like where is it where are these coming from what's predictive Well, they're coming a lot from the HR.

23:28They're hidden in plain sight in the HRS systems that are already in use every day in these companies. And they're just not making use of them in this kind of data-driven way. I think that's the first step. The first step is we help companies optimally use the data they already have. But the second step is even more exciting, which is we're working with companies who are really trying to get innovative and even doing a lot of A-B testing on finding the types of questions that are predictive of the folks that they want. So some people are looking at softer skills like grit and resilience and locus of control.

24:00Do you internalize things when they go wrong or do you blame other people? Folks are looking at all sorts of different characteristics and traits that they can assess at the time of hiring. And then we are helping them A-B test those things so we can get an application and a set of interview questions that are the most predictive of high performance. Let's go back to your previous example and say, look, Roland, we had this test, disparate all that. This process would help you figure out the types of questions and instruments that are most predictive for you specifically as a company. And they won't, they'll be different.

24:34They'll be different for Starbucks than they are for Pizza Hut. But the key thing is you can build these. We help companies build these models internally. which turns out to be their own specific talent IP. And then they can operationalize that to make sure they get the best people and the best jobs doing the things that they're capable of doing. And that is when you get people to their talent frontiers. No, it makes sense. Let's talk a little bit about hiring and conversation as well, because that's tied into this. You know, one interesting data point you discovered, I thought was kind of crazy, is that high-performing black employees are often even more likely to be promoted than similarly performing whites.