It’s a widely held belief in our society, even among groups that disagree with each other on other issues, that meritocracy is a public good. Broadly, meritocracy speaks to our innate sense of fairness—people should get what they deserve. But what exactly constitutes merit? In the dictionary, it’s defined as a “praiseworthy quality” or being deserving of praise. That’s about as subjective as it gets, since it depends entirely on the values of the person doing the praising. In academia, the definitions get more refined, but there is still disagreement. Is merit about “intrinsic excellence,” as defined by philosopher Thomas Mulligan? Or is it context-specific, as argued by Harvard economist Amartya Sen, who says it depends more on an individual’s ability to help achieve a society’s chosen values and goals? Our guest, economist and University of Chicago Professor Steven Durlauf, is the founding director of the school’s Stone Center for Research on Wealth Inequality and Mobility, and has been researching meritocracy and its influence on inequality for decades, particularly in the area of university admissions. Durlauf says that done poorly, so-called meritocratic systems can be gamed by social groups in ways that help them to hoard their advantage and privileges. But done well, meritocracy can create greater equality of opportunity and help individuals from a broad swath of society to fulfill their potential.
Listen to more of Steven Durlauf on The Inequality Podcast
About our Guest:
Steven Neil Durlauf is the Frank P. Hixon Distinguished Service Professor and the Director of the Stone Center for Research on Wealth Inequality and Mobility at the University of Chicago Harris School of Public Policy. Prior to this appointment, he was William F. Vilas Research Professor and Kenneth J. Arrow Professor of Economics at the University of Wisconsin-Madison. Durlauf's research spans many topics in economics. His most important substantive contributions involve the areas of poverty, inequality and economic growth. Much of his research has attempted to integrate sociological ideas into economic analysis. His major methodological contributions include both economic theory and econometrics. He helped pioneer the application of statistical mechanics techniques to the modelling of socioeconomic behavior and has also developed identification analyses for the empirical analogs of these models. Other research has focused on techniques for policy evaluation and the econometrics of cross-country income differences. Durlauf is also known as a critic of the use of the concept of social capital by social scientists and has also challenged the ways that agent-based modelling and complexity theory have been employed by social and natural scientists to study socioeconomic phenomena.
Durlauf received a BA in economics from Harvard in 1980, where he was elected to Phi Beta Kappa, and a Ph.D. in economics from Yale in 1986. He is a Fellow of the Econometric Society, a Fellow of the Society for the Advancement of Economic Theory, a Fellow of the International Association of Applied Econometrics, and a Research Associate of the National Bureau of Economic Research. He was elected to the American Academy of Arts and Sciences in 2011. From 2010 to 2022 he was co-director of the Human Capital and Economic Opportunity Working Group, an international research network linking scholars across disciplines in the study of inequality and the sources of human flourishing and destitution.
(music)
Cold Open:
Dani Rodrik: Economics has changed and, in particular, has become much more empirical. It's become less wedded to theoretical preconceptions about, such as, markets will always take care of problems or that the governments cannot ever solve things…
Stefanie Stantcheca: What we do find in our survey work is that fairness matters a lot to people…
Suresh Naidu: …inclusive prosperity meant thinking about democracy and its relationship to economic policy and democratic politics as something that we were committed to…
Atif Mian: …those deeper questions of how we want to organize ourselves as a society, so we deliver not just economic growth, but we also deliver the values that we need to aspire to.
Intro (Ralph Ranalli): Hi, thanks for joining us for another episode of the Economics for Inclusive Prosperity podcast. I’m Ralph Ranalli, and it’s my honor to be your host. It’s a widely held belief in our society, even among groups that disagree with each other on other issues, that meritocracy is a public good. Broadly, meritocracy speaks to our innate sense of fairness—people should get what they deserve. But what exactly constitutes merit? In the dictionary, it’s defined as a “praiseworthy quality” or being deserving of praise. That’s about as subjective as it gets, since it depends entirely on the values of the person doing the praising. In academia, the definitions get more refined, but there is still disagreement. Is merit about “intrinsic excellence,” as defined by philosopher Thomas Mulligan? Or is it context-specific, as argued by Harvard economist Amartya Sen, who says it depends more on an individual’s ability to help achieve a society’s chosen values and goals? Our guest today, economist and University of Chicago Professor Steven Durlauf, is the co-founding director of the school’s Stone Center for Research on Wealth Inequality and Mobility, and has been researching meritocracy and its influence on inequality for decades, particularly in the area of university admissions. Durlauf says that done poorly, so-called meritocratic systems can be gamed by social groups in ways that help them to hoard their advantage and privileges. But done well, meritocracy can create greater equality of opportunity and help individuals from a broad swath of society to fulfill their potential. I’ve really been looking forward to this conversation, so let’s get to it.
Ralph Ranalli: Steven, welcome. It's great to have you.
Steven Durlauf: Thank you so much.
Ralph Ranalli: When I was reading some of the research and preparing for this podcast, it really took me down kind of a rabbit hole, but I also feel like maybe I didn't go far enough because this is sort of an endlessly fascinating and multifaceted subject. But starting off, I was hoping you'd kind of make the case for us about the why. Why studying meritocracy is important, why you've devoted so much time to it, and kind of what's at stake if we don't sort of collectively get a better grasp on what meritocracy actually involves and what its consequences are.
Steven Durlauf: Well, I think one reason to study it, and one reason it motivated me is how salient it is in public policy debates. So if you think about the way that DEI has been criticized, it's often portrayed as an attack on meritocracy. Similarly, debates about affirmative action have often opposed affirmative action to meritocratic admissions. So that of course leads to the question, what is meant by merit? What is meritocracy? And so as I said, one reason I'll call it pragmatic, which is this is where so much controversy is. But second, there's a principled reason, which is that I think that meritocracy is a constituent of just decision making and good decision making in a society. Now saying it's a constituent does not mean it's the constituent. And so what I've tried to do in my research is clarify the ways one might think about meritocracy and provide the ways in which I think one can make reason defenses of it as something to consider in public policy evaluation.
Ralph Ranalli: If you look up the word merit in the dictionary, which I did, Websters defines it as a praiseworthy quality or virtue. And the OED defines it as the quality of being good and deserving of praise. And I was kind of surprised because I thought there was sort of a fundamentally intrinsic element to it, but both definitions mention praise, which is about as subjective as you can get depending on who is doing the praising and what they consider praiseworthy. So I'm starting to think that maybe meritocracy is perhaps the term most deserving of being used with quotation fingers. And I was listening to your podcast with Lauren Rivera of Northwestern who wrote this wonderful book about who gets hired for elite jobs at banks and investment firms. And she said the term was first used in 1958 by a sociologist as a joke, as satire. And yet we seem to have a rough popular consensus that meritocracy in general, although those variously defined by various groups, is a social good. And both sides in the argument, as you said, over affirmative action, for example, make versions of meritocratic arguments. How did we get here where meritocracy became such a thing?
Steven Durlauf: Well, let me start by giving a type of meritocracy where there's a consensus now. And that is the use of civil service examinations versus other ways to appoint officials to the government. So when the United Kingdom developed civil service exams or the United States did, that was a meritocratic set of policies. And I would say they're universally praiseworthy, but what was underlying them was the idea that you hire somebody that's best for the job as opposed to who knew the Lord or the political party boss, et cetera.
Ralph Ranalli: Right. It's a way to combat cronyism.
Steven Durlauf: But it also had an efficiency idea there. In other words, there was something good about matching people to positions because they would be most productive at them. So there's a long background of that. In fact, the imperial examination system in China is often touted, I think properly so as a historically revolutionary way to have something that's meritocratic. And of course that comes from the tank dynasty. So the Europeans and Americans are only 1100 years late. But the serious point is this notion of trying to match people to positions because they will do something.That's much of the source, I think, of why the public reacts positively to that. Now, Michael Young's satire was equating merit essentially with effort and ability. And his satire was if you had a society that had this mechanical mapping of the positions of authority, positions of economic success to them, and parents, of course, were transmitting many of these things to their children because you can hoard opportunities to become more able through elite schools, et cetera.
You would end up the equivalent of the British aristocracy, but it wouldn't be because of bloodlines per se. It would be because of this worship of merit as the only factor that's determining outcomes. But you brought up a very important point, which is that merit, as we use it in common language, is not what social scientists are referring to as merit. And an example of that would be that we often is what you said, it's praise. So we would say somebody has merit because they've been brave, for example. That's not the same thing as saying they would be good at a particular job. That all said, I think that within the discussions of meritocracy, there are some different ways to think about merit that have been the focus of both philosophers and social scientists. And so one idea of merit, and Thomas Mulligan, for example, philosopher, would emphasize the notion of intrinsic excellence.
So if I'm deciding what people to assign to jobs as physicists, he would have to the notion that there's some intrinsic excellence to being a physicist, and that would sort of determine who gets what type of position. The notions I take of merit, the way I think about it is different, which is I, and this way, I'll surround myself with deep thinkers and say that I follow Marty Sen's writings on meritocracy, where he actually argues merit's underdefined and meritocracy is underdefined. Unless you tell me exactly what the context is.
Ralph Ranalli: Right.
Steven Durlauf: So the way that I proceed is to say that merit is functionally defined. In other words, I kind of work backwards. I would say you have a decision you have to make, who to hire, or you have to decide what students to admit to the University of Chicago or any other institution. And merit is determined by who best meets the objectives of the organization. So the point is that you start with asking what it is you wish to accomplish, then merit is implicitly ranked by the people that best meet it. Now,
Ralph Ranalli: Is that the same thing as retrospective meritocracy versus prospective meritocracy?
Steven Durlauf: Yes. Yes. And so in my own writings, what I've argued is that the notion that somebody has an intrinsic excellence is retrospective because it means that you don't look at the future. You basically say, "This is who you are. " And there's some characteristics that are defining the excellence. The other type of retrospective meritocracy is a reward system. And so when I teach this, I ask the class, if I did everything I could to train to run as fast as I could in the hundred meters, is there any circumstance where I'm going to ... You would say, "I should receive the Olympic gold medal and I can assure you it's not going to matter." So in other words, we do have context in which we think of the merit is defined because it's a contest and we have a reward system. So those are kind of the retrospective examples I would focus on.
But then think about college admissions. Right. That's entirely prospective. In other words, you admit people ... Well, it doesn't have to be, but I would argue it should be. The reason you admit students to an institution is because the institution has some objectives. And I put an S on that very deliberately because I think it's important to think about the multiplicity of the objectives. So let's take the University of Chicago.
One objective you might have as an institution, and so think about very schools like Caltech, MIT, I'll put in Chicago given loyalty, is you might want to admit students because you wish to produce as many future Nobel laureates as possible. Right. If that's the definition of your objective, then that tells you implicitly what the merit is in terms of the admissions. You would look for people who are most likely to be outliers as scientists. By the way, I think that's a perfectly legitimate objective for societies. One thing you wish to do is push the scientific frontier and you therefore want to cultivate the possibility that individuals are going to do that.
Ralph Ranalli: Yep. But you also said there are multiple objectives, right? And those involve balancing, don't they?
Steven Durlauf: Well, so partly I want to say that you can simultaneously do them. So I would say that, I'm going to make up a number, but I think it's illustrative of a fact. And that is, if my goal is to produce Nobel Laureates and I'm going to admit 1,800 students, I've told you what to do with 30 of the slots. What do I do with the other 170? And so the first point is I think that if you give an objective, there's no guarantee that that is exhaustive in any interesting sense, but let's suppose that it could be. That's where you get balance. And so you can say one objective, again, I would argue one objective is that you would like the university to try to promote extraordinary scientific excellence. The second thing you'd like a university to do is to produce outstanding leaders in organizations.
And so that tells you a different type of individual you would look for. Yet I could say that what I think is important in the university is also the production of artistic excellence, musical excellence, literary excellence. And so of course that becomes scope for other people. And so once you go down the types of individuals you wish to produce, that gives you a much more heterogeneous class. So that would be kind of the first thing that makes it complicated is that we would have a plethora of excellences in the future, not retrospective, but the ones that you're going to produce as a consequence of admissions. But then the second thing is to emphasize that you don't admit people in isolation. You admit a class. So then a question has to be not just, do you look at everybody's test scores in isolation, light them up like a sausage and cut them off here, and you can add grades to make it a two-dimensional sausage, but the principle's the same.
You have to think about the interactions between students. And so that leads to something that has been controversial, which is whether or not diversity, for example, is intrinsically conducive to the educational process. And I think that, to be honest, that some of the arguments for that have been overblown and the history of the Supreme Court decisions is a little bit of background is the original arguments for affirmative action were twofold. Number one, that it represented a proper response to historical discrimination. It was a way of trying to dynamic, speed up the process of the diminution of terrible and historical injustices. And second, that it was meritocratic in the traditional sense, which is it was identifying people of great promise who, in terms of standardized test scores, et cetera, aren't going to be identified. Now, I have great sympathy for those two, but over time, the Supreme Court moved towards requiring a explicit argument that diversity was intrinsically valuable to education.
So I give that as background. And I think that some of the arguments were not well crafted and the evidence was not commensurate with the strength of the effects. That said, I think there are important reasons why diversity mattered. So if somebody said, I think the diversity in a mathematics class had some intrinsic effect on the learning of topology, that's not a good argument. If you said the diversity of a class matters for the interactions of the students and friendship networks, et cetera, which may be relevant to how study groups formed, there's probably some truth in that. I'm not sure it's first order. On the other hand, if you said to me the diversity of the class matters for the way that people understand American history or American literature, I would say absolutely that is surely the case. And I think that there's evidence that attitudes about race, et cetera, are affected in college context by who you interact with.
And I think we have very good reasons to believe from other examples that the diversity is productive in that sense. So if I can digress for a second.
Ralph Ranalli: Sure.
Steven Durlauf: There's been very interesting studies of the effect of racial composition on juries. And so there's been, it's experimental evidence. We have a kind of a mock trial and you have a jury assess evidence. And what's been found is that the composition ethnically of a jury, whether you have African Americans and whites together, just whites, they can come to different decisions. And there's been a study, of course, as a trial, you actually can listen, you know what people were doing, what they were saying, and you find the interactions are different. And I think that it's actually a beautiful example because let's think about the two of us. If we were on a jury together and we were an African American defendant and was a white motorist, we were going to process all of that through the template of our lived experiences as the term is often used.
An African American's going to process it differently. It's not a matter of saying one person is right, it's saying one person's wrong. And it's saying that in this complicated process of truth creation or truth seeking, all of those perspectives are fundamentally important. And so I put all that on the tables. I think that is something a university has to think about.
Ralph Ranalli: Yeah. Ultimately, I think a lot of this comes down to values, like a lot of the very base kind of determinations of what people see as meritocratic. Now, this podcast is called Economics for Inclusive Prosperity. We're open-minded, we're fundamentally empirical, but we're not necessarily values neutral here. Inclusive prosperity is in our name. And we start with the proposition that for many reasons, inclusive prosperity is a fundamental imperative and a public good. Given that orientation, like our orientation, what do you think are the most important questions we should be asking and the most pressing issues that we should be addressing about meritocracy as it's currently practiced?
Steven Durlauf: So I think the most important thing I would say is that I think the publics or most of the debates about meritocracy don't think about meritocracy in a comprehensive fashion. And as a result, there is a false opposition in much discussion between meritocracy and efficiency. And much of my work has been specifically trying to argue against that. And so one way to think about that is the role of equality of opportunity in all of these conversations. And there are debates amongst philosophers. Is equality of opportunity part of the definition of meritocracy or is it not? My work, I'm not a philosopher of course, is to say something different, and that is equality of opportunity is extremely important in things being meritocratic and therefore it can be efficiency enhancing. So let's kind of go down that route a little bit. When we think about meritocracy, often it's often the discussion is a point in time.
So you have a bunch of college students, these are their test scores, these are their grades. Let's stipulate they are indicative of where students are in terms of their skill levels at a point in time. And you're supposed to max students to universities in such a way that basically the better your test scores and grades, that's monotonically determining likelihood of admission. That's an odd thing to think about because that, again, that's a reward system. The first thing you want to ask for universities is what the value added is. In other words, you admit people, and going back to what should universities do, think about a public university. Maybe the objective there is you want the most favorable distribution of education levels, skill levels, human capital to use the economist standard phrase in your workforce to have a productive economy. When you say that, you're not asking about the levels where you come in, you ask what the gains are going to be.
So one type of efficiency is to focus on value added. So if efficiency defines who ultimately has the merit for the admissions, then you have this idea that you admit people in order to be efficient in terms of the productivity of the workforce. So that would be the kind of thing I would put on the table, I think is essential. A second thing to say is that if I'm worried about the disparities in students at age 17, a fully meritocratic system, the unconditional one is that I have to think about the development from birth all the way to age 17. So you design a system in such a way that when you get to people at age 17, many of the disparities you see haven't been produced for reasons that are inconsistent with the quality of opportunity. And that's kind of a convoluted way to say the following.
And that is, now I'm going to give you my value judgment, that the society we should aspire to is one that everybody fulfills their potential. And that is in the crudest sense, that's going to be the efficient society. I tell students that one of the things I teach is the economies that win the 21st century will be the ones that allow the potential of all the entire population to be fulfilled. All the issues of gender inequality, racial inequality, all of those are their efficiency denying. And so that's kind of the background. And then you ask the question, what sort of equalities of opportunity and points in time do you generate so people fulfill their potential? Now, all of that is consistent for me with meritocracy. That means that the outstanding students with scientific potential, they're the ones that you make sure have the opportunities to go to Caltech and MIT, et cetera.
But it means many more things. It also means you focus on opportunities at every point in time to make sure that you're constantly leveling up as it were and producing the dynamically efficient economy.
Ralph Ranalli: So you've said it's actually really difficult researching those trade-offs between equality and efficiency. It's not an easy thing to do. Can you talk a little bit about that and what you had to sort of overcome to study those trade-offs?
Steven Durlauf: Well, I think to be clear, my work has been largely theoretical, so I don't want to exaggerate there's been empirical work that I could say these are the findings. The challenges I would say are the following, and that is that to be dynamically efficient and what I'm calling truly meritocratic, you need to know about lots of things such as the ways that individuals interact in an organization, be it a school or a firm. And you need to know the dynamic consequences of creating opportunities for people. So to give one example of that, the parents, the college students of today are the role models of tomorrow. And so in thinking about the efficiently, who do you want to admit to college, et cetera, you have to keep in mind people after college, they're going to do something. So the first thing you would think about is kind of where they're going to be in the labor force, and the second's going to be what they do when they become parents.
And so in other words, the inequalities of opportunity at the family level, they're dynamically affected by what happens today. And so what makes these things hard is that's a lot of information you need. If I just say to you, admit the person with the highest test scores, that's easy. You have the test score. If I ask who's going to have the highest value added, that requires a lot more information. That all said it doesn't mean we don't have examples where you can make compelling cases. So let me just take examples from the literature. It's been very interesting work that has found that exposure to a single African American teacher has very large effects on the outcomes of African American children. That observation is very important. And if you said, "Do we know the reasons why?" No, it is not obvious. Does it have to do with some cultural sensitivity in the way teaching is done?
Is it having higher aspirations for the students, so on and so forth? There's many different interpretations, but let's take that as a fact that I think has been well documented. You have to know that fact in order to think about what's efficient. And I claim if it's efficient, it is meritocratic because my efficiency criterion is make sure that we have the best educational outcomes for African American students.That's a thing you have to study and that requires pretty hard statistics in order to uncover that. Or think about the example I gave of the juries, you got to be clever. And the authors that did that, the researchers, they had to actually sit down and construct an environment. You can do what's called an experiment in the social science sense in order to extract that information. So those would be kind of the examples of the challenges that you have.
Ralph Ranalli: Yeah. So you've said that there's a lot of missing information when policymakers go to make policy about sort of meritocratic related and areas where meritocracy is kind of at issue. But you've also said that there can be robust policies despite the limits on policymaker knowledge. That seems a little counterintuitive. Can you sort of explain how that works?
Steven Durlauf: So what I was trying to say is that in thinking about policy decision making, one has to be, this is going to sound like an obvious thing to say. One has to be very upfront about where we don't know things. Right. The question is, well, we have to do something today. In other words, I have to make a decision today, and therefore one needs a theory of decision making under uncertainty. Now you're talking the economist language, and that is that sort of the part and parcel of much economic theory is decision theory under uncertainty. How do we handle uncertainty? Well, one idea would be that I suppose you and I were going to make a bet on flipping a coin and we got it from the bank, so we know it's a fair coin. We don't know the outcome, but we know the probabilities. The social science problems are often harder because we don't even know the probabilities.
So then the question is, if I don't know the probabilities, can I come up with procedures where I make choices where they're pretty robust? In other words, no matter what happens, the worst case scenarios aren't that bad. And that's really what I was alluding to, is thinking through policies where we have good reasons to know that the downside risk is not high.
But the upshot is that in thinking about policies, the first thing one asked to do is distinguish between the cases where we know probabilities and where we don't, and it's often the case we don't know the probabilities. The second is, well, if you don't know the probabilities, you got to come up with some rule for that. Now, in other writing, I have kind of a view of this, and that is that ... And let me give it context to make it specific, and that's racial profiling for traffic stops. And so racial triling for traffic stops, the justification is that it's supposed to reduce crime. The evidence on that is extremely weak at best. And there's various reasons to be skeptical of the empirical literature, but I don't want to say it's zero for the sake of argument. My view is the way to think about profiling for stopping is we know something with certainty, and that is it's unjust.
And the reason for that is the profiling policies that have been argued to be efficient. They automatically make it more likely that an innocent African American motorist will be stopped than an innocent white motorist, and that's an easy violation of any fairness consideration. What I'm trying to say is the following. Here's where we are. We have a certain unjust injustice and we have an extremely uncertain efficiency claim there.
My policy rule is, if you're going to do something unjust, you have a very good reason for it. In other words, it's a very threshold. So
Ralph Ranalli: The hippocratic oath of policymaking,
Steven Durlauf: First
Ralph Ranalli: Do no harm, right?
Steven Durlauf: Yes. And so from that perspective, I know it may have been a little convoluted way they said it was to say that the upshot is I think we may have clear egalitarian fairness considerations. And if we're going to overturn them, there has to be a strong argument for it. So that's at least one way I would talk about how to resolve it. The other is simply to ask, are there policies where if we don't know the probabilities, the worst case scenarios are not so bad, and that is kind of a way one might proceed.
Ralph Ranalli: Yeah. So your work on this topic has been very multidisciplinary. I was reading through some of the papers and you seem to cite philosophers and sociologists as often as you do other economists. So my question is, what do philosophers, sociologists, political scientists and researchers from other disciplines bring to this study of meritocracy? And do you think that's a good example of how maybe they should be more widely brought in as collaborators by economists on other topics besides meritocracy?
Steven Durlauf: I very much do. And most generally, I would say that in wanting to focus on inequality, that's a subject matter as opposed to a discipline. And so it would be almost surprising if it wasn't enriched by the different perspectives. So that's a cheap compliment. So starting with meritocracy, the contribution of philosophers has been essential because they are focused on the normative dimensions of what could provide a justification for meritocracy. And so going down that route, one thing that comes into play is the relationship between thinking about the idea that we want societies to be fair versus thinking about what the idea that people deserve things. So what does that mean? Certain types of fairness, there's simply nothing of ... There's no debate about them that's interesting. In other words, if I said, "Is it fair or just for a firm to racially discriminate or to pay women less than men because they think men are supposed to be the breadwinners?" When the workers are doing exactly the same job and equally productive, then if somebody wants to debate it, I'll be honest, I would say I'm not interested in it because it's been resolved.
So that's a cheap one. Here's a harder one. I have two workers, they work equally hard, just one of them is twice as good at it. Should the person who produces twice as much be paid more? I still remember I set the thought experiment. They didn't work harder. They just were born that way. That's not fair. In other words, the genetic lottery, as John Rawls called it, that isn't a basis for a normative claim about inequality. On the other hand, I think most people who listen to that would say, "That doesn't sound like it's the same." And the answer there you would say is, "Well, because we also have some notion of whether people deserve things." And once you go down the dessert route, that's most philosophically controversial. So Rawls would reject dessert, other philosophers wouldn't. But my point is, once you think about it that way, now you see what becomes part of the ethical justifications for meritocracy is in addition to saying, and this is an idea that associated with John Romers of philosophers or an economist who's done major work in philosophy and others, is inequalities are unjust if you're not responsible for them.
The question is whether or not inequalities are just because you don't deserve them, and those aren't the same sentence. And once you go down that route, you begin to think about one of the normative justifications for meritocracy. I have pushed the purely efficiency ones. In other words, that is many contexts in which even if you have no intrinsic interest in inequality, it's simply efficient. And so those other interests you have are also met. But philosophers clarify all of that. As for sociologists, psychologists and political scientists, that's what it means to have good social science about an inequality question. So I think that I'm going to make the following generalization, but I think it's true. And that is if you ask sociologists what your view is of meritocracy, it's very negative. You ask economists, it's positive. Now, why might that be? I think a lot of it is because economists equate meritocracy with efficiency.
Sociologists equate meritocracy or they're focusing on whether or not the system is in fact matching people for the reasons that they're considered to be merit versus simply family background. But once you understand those differences, then you begin to think about the complications of actually thinking about merited Each point in time. So that would be kind of the immediate thing. The other factor is that sociologists obviously have different axes and different natural kinds in thinking about the determinants of outcomes. And so an example, sociology has been absolutely integral to economic studies of inequality is neighborhood effects.
Those are all social factors that were long studied in sociology. And then economists have benefited from that and imported those ideas. Economists have our own comparative advantages. My own career choice says what I think is a worthy thing to do, but all of that's going to come into play. Again, psychologists are going to bring in, I call them sort of different national kinds, different axes along which you organize things that are going to be integral to any consideration of inequality policies. So let's kind of focus on meritocracy. Sociologists and psychology would be much more attuned, I think, to the harms of meritocracy that have been discussed and critiques of meritocracy, such as I think the most compelling one by Daniel Markovitz, in which you think about the effects on individuals from losing these competitions. And so that can be an extremely high intellectual cost. Intellectual is the wrong word, excuse me, mental cost.
Ralph Ranalli: Right.
Steven Durlauf: And so again, that's an idea that would come from them and they would be very sensitive to that. In terms of political science, again, I don't want to sort of put everybody in their own box. In the design of policies, political efficacy, feasibility, that's going to be their bailiwick. And so I think in actually advocating policy, they're going to have insights that don't actually come from economics. One thing I think that comes when at least you bring the sociological perspectives in with the economic perspectives is the following. And that is Lauren Rivera has some very interesting work on looking at the admissions policies for elite preschools in New York City.
Ralph Ranalli: I was listening to that. Yeah, that was amazing. Those sort of developmental, the hoops that the applicant families had to go through to show sort of developmental stages and milestones.
Steven Durlauf: And so she's a fantastic researcher. Let's kind of take the two things. So she worked on the three year olds and she looked at elite organizations, firms and hiring people. And what did she find? Number for the three year olds that there were some criteria that reputably meritocratic that were being used. And in the elite firms, she found that I would say cultural similarity was helping to determine the placement. And so merit was being conflated with homophily, in other words, similarity between individuals. So those are important insights into the actual social mechanisms that are producing these matches that are supposed to be meritocratic.
Ralph Ranalli: Right. And those definitions have changed. I mean, I think I remember that podcast she was saying that at one point it was sort of civil service exams and standardized tests were seen as bringing in immigrants and other sort of groups that had been excluded by the traditional upper class. And then when the upperclassmen didn't like that, and they sort of shifted the criteria to being more like, oh, their qualities were less tangible. It was more sort of like that matched what the preferences of that elite
Steven Durlauf: Class. Yes. Yes. And so I think partly why I claim that the economist contributes to that is to clarify that
You want to distinguish between what is defined as merit versus whether or not the system is matching with respect to what you've defined as merit. The critiques you make when Harvard suddenly introduced all these character issues to reduce the ... I don't think that I'm saying anything controversial in the 1920s to reduce the number of Jews that were being admitted. What they did is they defined merit in a new way. And by their offensive lights, they had implemented their own version of meritocracy. But I think this distinction between what the objectives are and kind of the way you match the objective, that's a thing that one has to be very careful about. What I want to take from the Rivera examples and say, how does that kind of lead into things that the economists approach? And again, this is something I think is important in general in thinking about society, is that we have all these contexts where we think society wishes to say meritotocracy here is a virtue.
In college admissions, we want them to be meritocratic and admissions to elite high schools in New York City. We want them to be meritocratic. Now to be clear, I think that they ... I support the elite high schools in New York. There's not a critic. I'm saying something different.
At each point in time, there's some notion of meritocracy. And it can meet ... Again, it wasn't morally objectionable the way I gave the Harvard example. It was saying you want to do something. And in isolation, it's a worthy objective. The problem's the following, and that is, are two problems. Number one, they cannot be thought about in isolation. In other words, if my goal is to produce the best scientists, that doesn't in isolation tell me what to do with admissions to a graduate school, to undergraduate, to high schools, et cetera. I have to think about all of them collectively and ask what sequence of rules, what sequence of a criteria at the end produce something that is our objective. And it is not the case that what looks like a meritocracy that's static at point, point, point, point, is dynamically efficient. And the way to think about that is suppose that I wanted to maximize the average academic performance of high school seniors, and I know exactly how the system works.
Everything's perfectly measurable, and it's efficient to sort of take the top 30 students and put them in a classroom because they'll be really productive at the end, so on and so forth. That doesn't tell me what to do with five year olds.
You could say, well, maybe it's best to take the highest schooled five year olds and put them in the classroom and kind of keep segregating them across time. You can show that's actually dynamically efficient, even if that's the policy that maximize the performance of six year olds. So we don't care about that. We care about the 17 year olds. And I don't know if the example is clear, but there's a few points in time where we have essential reasons why we would think about the sorting and all of that as providing social. But all the other ones don't have that, and they should only be thought of as part of the process that's producing the attributes where the meritocracy admissions or the matching in the admissions really matters.
Ralph Ranalli: Really matters.
Steven Durlauf: Yeah. So what I'm saying is you kind of do what I call the unconditional meritocracy. There's a few points in time where it's essential that the dynamic solution coincides with the static ones. And so my big view of society is we have much, too much segregation, too much sorting because we're doing it in cases where we don't care about the intrinsic consequence of the sorting. We care about it as a means to moving to the place where the sorting is particularly important.
Ralph Ranalli: Right. And the more sorting we do, the more entry points there are for people who have the ability to game that sorting system are able to do so, right?
Steven Durlauf: Yeah, absolutely. And that's part of what is a limit in everything I said as I said when I was talking about a theory. I assume that you actually can implement exactly what you want. And I want to say under the best case scenario, and that's a little bit of how I've done my writing on this, is to say, under the best case scenario, do you really want to segregate students by test scores year after year after year? It might be that when you're talking about who gets the PhD programs at the end, that's when you do it, because that's when you really can identify stuff. But for the three year olds, I say the madness is not only you can't measure anything, you don't care what the performance is of four year olds. You care about the performance of 25 year olds, and you have to basically shape the distribution of achievements over time.
So when you get to the 25 year olds, you have the most favorable one from which to do the other thing.
Ralph Ranalli: Yep. So before this podcast, I was a policy podcast. I had a podcast called Policycast at the Kennedy School. So I've been a policy podcast for almost a decade now, and I really know how hard it is communicating complex and counterintuitive ideas, of which there are definitely a number of them here in this area. How do you hope to give the general public a broader understanding of what meritocracy is and what it entails? And in, as you said, best case scenario, what do you think that could achieve or what would you hope it could achieve?
Steven Durlauf: Well, I think what I would like is to get away from this opposition between meritocratic, the idea that meritocratic admissions are inconsistent with admissions to colleges, for example, which promote diversity. Now, I understand the Supreme Court's made a decision, so I'm putting that question aside. On the way to think about admissions to colleges, to recognize the value of diversity for everybody. A second thing I would like people to understand is that the processes which we educate people, and I'm hardly the first person to say this, is determining the quality of the workforce we have. And so the imposition of what people think are meritocratic admissions for high schools or colleges may produce a workforce, which is collectively less productive, and that adversely affects everybody. And again, that argument's made by Roland Benabund others, so I don't want to ... In the past, so I don't want to claim deep originality there or anything.
I want these to be things people think about, and that's the dynamics. You brought up the issue of communication. So an example of something I would like people to think about is, and I, again, which I have discussed in other context, the University of North Dakota and their medical school, which has policies to try to admit Native Americans. I don't know the state of play with them with the Supreme Court decision. Let me be clear. But the University of North Dakota's medical school had explicit policies to try to create more Native American doctors and they were a significant source of the United States. How do we think about that as a society? If I have this narrow conception of meritocracy as a reward, and I'm not rank ordering people by the MCAT, that sounds like a violation of meritocracy. I want people to ask the question differently, which is why do we admit people to medical school?
We want to produce the best medical outcomes in the United States. I think that's an easy one. Right. Now take the second fact, which is North Dakota has Native American reservations that are among the poorest places in the United States, very, very extraordinarily sad situations. Now add to the fact that you have to ask who's going to be willing to be a doctor in Native American communities who's more likely. Then you have a different vision, don't you? In other words, if I want to make sure that I've produced a group of physicians that are helping the entire country, I need to account for the fact where they're going to help. That to me is absolutely meritocracy. I don't see how it could not be because I have given you a very clear objective function and I then functionally, as I said, define meritocracy as that which meets it.
I don't think that that's something that is going to be offensive to people who think hard about desert and meritocracy. Maybe a part of what I would say is I think the communication has to do with giving very concrete context where one thinks a consensus exists. In other words, there's a reservoir of common ethical ground.
Ralph Ranalli: Now, do you think that given where we are with sort of deepening inequality, a lot of public discontent over where the economy is, over the level of wealth concentration, in particular in the top 1% and the struggles of the middle class working class, do you think that could be kind of a fertile ground for planting the seed of a new vision of meritocracy? Do you think that time is sort of opening a window of opportunity for maybe new understanding?
Steven Durlauf: I do. And I will say that part of what, not taking off the academics hat, but my citizens hat, is part of what I want to do in my writings is reclaim meritocracy for liberals. In other words, the dichotomy now is that somehow it's a conservative idea. I think it should be a liberal idea. It could be both. In other words, I'm not claiming they conservative can't believe in it, of course, but I think that there has been a tendency to think that it somehow sits on one side of the political spectrum. Much of what I would argue is ... So this philosophical question is, does meritoxicity require quality of opportunity?
I don't actually want to take a stance on that because I want to say something different, which is that many increases of improvements in equality of opportunity are meritocratic, even if you don't intrinsically care about equality. In other words, the only thing I care about is the average academic performance of students or the average incomes in the economy. There are good reasons to believe that opening up many opportunities will maximize that. And so that's where the economist comes in, not the philosophers. There's a wedge between them. In other words, I can't say all equality of opportunity, but my lights is meritocratic, but a lot of it is. And so that's part of what I want to say is that if we could focus on policies that are both equality of opportunity enhancing and efficient, and they are meritocratic, so that I think should be claimed.
And then go down that route. If I said to you, we're going to replace the system of civil service appointments with an examination that's trying to find people that are able, I think that a card carrying egalitarian should have said, that's great because it was eliminating something we thought there's a consensus was an unfair reason for the assignments.
Ralph Ranalli: And
Steven Durlauf: So I really do think we've gotten into the liberals or liberalism and liberal left, however you want to define the coalition, should embrace meritocracy as a virtue, not just for pragmatic grounds because of public supports, but in principled grounds, because the idea that people deserve things, that's worthy of respect. And the idea that we want a society to be efficient, that's worthy of respect, and those are things that should be embraced, but I think the fight should be, so to speak, to deny the opposition between it and equality enhancing actions. Think about all the work on early childhood investment that's been done or play a very childhood investment in Head Start, those things are efficient.
Ralph Ranalli: Right. And I think it's not an accident though, that those things, that Head Start has been a place where there's been a fair amount of agreement across the spectrum.
Ralph Ranalli:. Well, Steven, thank you so much. This has been really interesting conversations. As I said, it's a fascinating rabbit hole. I encourage anyone who wants to dive down it even deeper to read your work and listen to your podcast. And thank you so much for being here.
Steven Durlauf: Oh, Ralph, it was a wonderful opportunity to meet you and talk to you. Thank you very much.
Outro (Ralph Ranalli) Thanks for listening. This podcast is a production of Economics for Inclusive Prosperity, a network of leading economists looking beyond neoliberalism and economic nationalism to answer the question: How do we build a more inclusive, more sustainable, and more prosperous economy for everyone? Our show is produced in collaboration with the Reimagining the Economy Project at the Kennedy School of Government at Harvard University. The co-producer of this podcast is Tony Ditta.Please join us again in two weeks for another new episode, and if you like our content, please share us on social media and remember to subscribe on Apple Podcasts or your favorite podcasting app.