Seeing the Taj Mahal is a striking experience in its own right because the monument is so majestic. But what struck me was the incredible contrast between the monument itself and what was outside. It was one of the most impoverished areas in India, in Agra. Just seeing that incredible contrast between the tourists and the people who were begging for food, basically outside from the tourists, left a mark on me. I think back to that experience, wondering why things were so different on the inside versus the outside.

COWEN: And that was a social justice feeling? A problem-solving, puzzle-solving feeling? Both? Or what did you think at the time was the reason?

CHETTY: I think a bit of both. I don’t think I fully understood the determinance of why people ended up in such different circumstances. I certainly felt it didn’t seem very fair to me. From a justice perspective and also from the perspective of trying to think about ultimately how might you change something like that. At that age, I think I was too young to put those pieces together. But experiences like that shaped the interests that I have today.

COWEN: And in India, how good was your kindergarten teacher?

CHETTY: [laughs] I went to a Catholic school, actually, in India my first few years, and it was a very strict school. What I remember is you had to learn all your —

COWEN: What school was this?

CHETTY: This was called St. Columba’s in New Delhi.

COWEN: I’ve visited, I think, St. Thomas and spoken there, but yeah.

CHETTY: And so, it’s the type of school where a lot of discipline . . . You really have to do all of your multiplication tables and memorization and so forth. Insofar as those kinds of non-cognitive skills — something we see in our research — discipline, focus, things like that — seem to be highly predictive of later outcomes, perhaps that had a payoff.

COWEN: So now let’s fast-forward to 1997. You’re at University School of Milwaukee, right?

CHETTY: That’s right.

COWEN: And Milwaukee, in your later work, you show to be a city where there’s not a lot of upward mobility. It’s a highly segregated city by income, by race. It’s doing poorly on some social indicators. What did you learn from those years you spent in Milwaukee? You were quite young, but if you were thinking about the Taj Mahal, what was your thought about Milwaukee back then?

CHETTY: I went to an outstanding college prep school, private school, in Milwaukee called the University School of Milwaukee. At the time, I recognized the fact that the resources and opportunities we had at that school, which was on something like a 150-acre campus on the north side of Milwaukee, were extremely different from other schools that I visited, which I happened to visit in the context of the chess team and tennis team that I was on.

So you’d see other schools and you’d see the vast differences in terms of how this private school looked relative to the public schools and what the neighborhoods looked like and so forth. It struck me — we lived on one side of the city, and I’d take the bus through the entire city, essentially, to get to the school. So you could feel the degree of segregation, which was one of the other things. In our current work, we see Milwaukee as one of the most segregated cities. Those types of places tend to have the lowest levels of upward mobility. So I happened to be lucky and be on the other end of the spectrum. For kids of affluent families, Milwaukee has as good outcomes as any city in the US, and my experience was very consistent with that.

I think, once again, seeing those disparities made me wonder how much of where I will end up in life happens to be because of having the opportunity to go to this school as opposed to going to some other school within that city.

On growing up in academia

COWEN: Now your father, he’s a well-known economist, and he studied econometrics with Arnold Zellner at University of Wisconsin. At what age did he start talking to you about Bayesian econometrics?

CHETTY: [laughs]

COWEN: Which is one of his fields, right?

CHETTY: That’s right, my dad did a lot of early work in Bayesian econometrics with Arnold Zellner, and the academic environment was something I grew up with since I was a kid. I’m the last person in my family to publish a paper. My sisters are also in academia on the medical and bio side. Whether it’s statistics or thinking about scientific questions or thinking about how to change things in the world, that’s the environment in which I grew up from the youngest of ages.

I’m the last person in my family to publish a paper. My sisters are also in academia on the medical and bio side. Whether it’s statistics or thinking about scientific questions or thinking about how to change things in the world, that’s the environment in which I grew up from the youngest of ages.

My dad, I remember, brought home a computer from Belgium when I was three or four, and I remember starting to learn how to do some simple basic programming on it.

COWEN: At age three or four — that’s amazing.

CHETTY: You know, just trying to make simple input-output things and a small game and so forth, so that’s the kind of environment I grew up in.

COWEN: Even in high school, you wrote a paper on Time on the Cross, the economics of slavery, right? Do you remember what you argued in that paper?

CHETTY: I think I remember. I don’t remember the details, but I remember questioning the evidence, whether you could draw precise causal inferences, but whether you could say, given that data, that the central thesis of Time on the Cross, did that really seem to be valid or not? I can’t quite remember the details of my argument or how rigorous it was. I am sure I would poke holes in it myself, given my current interest in data.

COWEN: And in high school you also published a paper, I believe, on electron microscopy staining techniques. How did that come about?

CHETTY: [laughs] When I was in high school, I thought I would become a biomedical scientist — that’s what I was interested in.

COWEN: Like your mother, in a sense.

CHETTY: Right, like my mom and like my sisters, I was really interested in things like biomedical engineering. So, to get more experience with that, I worked in a lab for a couple of years at the medical college of Wisconsin when I was in high school. The lab I happened to work in was an electron microscopy lab, and they were developing staining techniques to identify different types of cells and so on. That’s the project I was involved in.

I learned through that experience that while science was fascinating and extremely important, I was more excited about the statistical aspects of that work rather than the staining itself or the pipetting or the microscopy. That experience showed me that I was more interested in the numbers side, thinking about social problems — coming back to the earlier experiences we talked about — and that led me down a different path.

COWEN: And then from your father you always knew economics was out there, right?

CHETTY: Yes.

COWEN: And now, graduate school. You go to Harvard, and you’re working with Martin Feldstein. I’ve long been of the opinion, he’s one of the very best and most underrated mentors in addition to being a great economist himself. But other than economics, what is it he taught you that was of central importance, strategically or tactically or thinking about research?

CHETTY: Well, Marty was such an inspiring figure for me in terms of the impact he’s had on the profession, the questions he was working on. What struck me immediately was that the types of questions Marty asked and worked on are some of the most fundamental questions in economic policy. He didn’t shy away from asking the big-picture questions, even though it might be difficult to answer them perfectly.

What had a big impact on me and working with Marty is him giving me quite a bit of independence, even at a very early age. When I was a sophomore, technically working for him as a research assistant, he said, “You seem to have a number of ideas. Let’s talk about them. Rather than working on a specific project as an RA, go work on the things you’re interested in. Let’s meet from time to time to talk about your ideas.” Hearing that when you’re 17 or 18 years old from somebody like Marty Feldstein builds a lot of confidence and was very inspiring.

COWEN: So you were working with Marty when you were 17?

CHETTY: Yes, that’s right when I got to Harvard.

On deconstructing the Raj Chetty production function

COWEN: Now to fast forward up to the present, we’ll go back to some of your earlier papers, but Adam Ozimek — who writes for Forbes — he said a few months ago, he was making predictions and he wrote on Twitter, quote, “It has been months since Raj Chetty unveiled some new really big results, feels like we’re due for a big one any day now.”

And three days later you came out with a paper about the chance of rising up to be earning more money than your parents were. And here is my question: We’re at Silicon Valley, where every company asks, what is their comparative advantage or what is the moat surrounding them that makes their position stronger, invulnerable.

If we look at your papers, they’re about topics people have already thought about. The data work is completely state of the art, but I don’t think it would be said you’re doing something other people can’t do, and yet several times a year, you come out with papers of great import that make a big splash, and the results seem to hold up. So what in fact is your competitive advantage? [laughs]

CHETTY: That’s a tough question. Part of what we try to do is exactly as you said: take old questions. I think some of the most important questions in economics and social science have not yet been fully answered, and the recent availability of big data of various types allows us, for the first time, to tackle those classic questions. What our research group tries to do is bring those two things together.

COWEN: But those both sound replicable, right? What’s the non-replicable asset?

CHETTY: What hopefully our contribution and scale is, is showing how you can take those large datasets and not get lost in them, and bring out the key lessons that are relevant for thinking about these classic questions.

It’s very easy — students often have this reaction, that all I need to do is get access to this big dataset, and then I’m going to be all set for my thesis. And what you end up finding is that that is often not the case. It’s very easy to write a paper that is not that good, even with cutting-edge data and modern techniques. So one of the things that I try to do — and the easiest way to see this is if you internally, within our research group, see the iterations of the papers we’ve been working on — where we start out is often very far away from the papers that people see as the finished product. We work hard to try to write a paper that ex post seems extremely simple: “Oh, it’s obvious that that’s the set of calculations you should have done.”

[S]tudents often have this reaction, that all I need to do is get access to this big dataset, and then I’m going to be all set for my thesis. And what you end up finding is that that is often not the case. It’s very easy to write a paper that is not that good, even with cutting-edge data and modern techniques.

Let me give you an example, a concrete example, in the context of the paper you just mentioned — the one that came out after that Twitter comment you mentioned, where we look at the fading American dream and look at how children’s chances of doing better than their parents have changed over time. Well, it turns out, when you dig into the details of that analysis, in order to know if a child is doing better than their parents, you would think you need to have a dataset where you see both the parents’ income and the kid’s income. It turns out you don’t have that data in the US going back to 1940, which is when we wanted to start our analysis. In fact, you only have that data starting really for kids born in the 1980s.

So the key conceptual idea of that paper was to figure out a way, even though you could not link parents and kids, a statistical method of determining the fraction of kids doing better than their parents despite that data limitation.

Often our conceptual advances are in the context of taking a big question, using the best data that we can possibly get, but then closing that gap and trying to find creative methods and creative answers to those old questions, and executing in the best way possible.

One thing I take pride in is the conclusions we reach and the statements we make, we’re quite confident in them. We take pride in the fact that subsequent studies do not end up showing, “Oh, there was some assumption made that ends up being questionable,” or “The data is not quite up to par,” etc. We’re very careful on all those steps.

COWEN: If I’m trying to model the Raj Chetty production function and I described it as such, it’s a multiplicative model, so there is getting the data, but that’s not the key point.

CHETTY: That’s not the key point.

COWEN: There’s then some conceptual advance that allows you to see the data can test something that other people hadn’t seen, and then there are numerous stages of execution, and then there’s also recruiting and managing the team. There’s a whole bunch of different steps, and you’re trying to do well at each of them and very few other people can do well at each and every one, and maybe that’s the way to think about your moat. Is that fair to say?

CHETTY: That is our strategy. I think the other thing that’s extremely important is, we spend a lot of time on trying to achieve clarity. There are ways to write papers in economics that are more accessible to the public and thereby have greater impact, and there are ways to write papers that are more technically oriented and narrow the set of readers.

COWEN: How do you learn the focus on clarity?

CHETTY: That’s just something, I think, coming from a science background from childhood, I really appreciated the clarity, for instance, of scientific experiments where you can say with precision that this is what we think happened when you changed something.

Having that interest from a young age, I want to be able to explain my findings to my parents or my wife, who’s a neuroscientist, and so forth. So that type of focus is something that distinguishes our research group, and trying to think in innovative ways about problems that people have worked on in the past.

For instance, there’s been a lot of work on neighborhood effects for 30 years. Our group by no means is the first to think about neighborhood effects. Where we’ve tried to push forward a little bit is thinking about the mechanisms through which neighborhoods affect kids and showing that, in particular, every year of exposure during childhood seems really critical, something that struck us as being an intuitive hypothesis, but then figuring out how to document that clearly in the data, and that has led the field to be able to now look at neighborhoods in a very different way than people were looking at in the past.

On the peculiarities of tax salience

COWEN: We’ll get back to that paper, but let’s briefly talk about two of your earliest hit papers that are still, I think, your two most heavily cited papers.

The first one of those is on salience of taxes and prices. One of the results in this paper — there’s more to it than this — but consumers are less discouraged by the price when the price takes the form of a sales tax added on rather than when the price is embedded in a single net total price up front, correct?

CHETTY: That’s right.

COWEN: And you still like that paper, yes?

CHETTY: I still like it, and there have been plenty of follow-up studies verifying the results. I’m happy to see that.

COWEN: Now let me ask you, people from Europe — often when they come to the United States, they say, “Why is it that you charge prices this way?” They feel irritated. Consumers here don’t feel irritated. They’re in an equilibrium where you just pay the price you see, and we’re in an equilibrium where the sales tax is almost always added on. What accounts for that difference in terms of salience or other factors?

CHETTY: What accounts for the difference in how people are reacting?

COWEN: In Europe versus United States, but Europe does it where there’s no add-on.

CHETTY: So are you asking what accounts for the difference and why does Europe do it differently from the US? Or why do Europeans react differently than Americans do?

COWEN: Well, both.

CHETTY: Yeah, you would think that . . .

COWEN: The suppliers would favor the add-on.

CHETTY: Yeah, yeah. Well, the suppliers definitely favor the add-on. I think the nature of the value-added tax in Europe, which is levied at various parts of the production process, makes it a little bit more difficult or cumbersome to levy the taxes and add-on at the end. It’s not that you can’t do it, but just when you think about the mechanics of it, it’s more natural to build it in along the way. In the US, we have the sales tax as opposed to a value-added tax, so you can see why it might be natural to just add it at the end.

Companies recognize that, when the tax is added at the end, it’s in their favor because consumers don’t cut back on demand as much as if the tax was included in the price, and that would force companies to basically cut prices and bear some of the costs of the tax. This paper — coming back to your earlier question — the way I started thinking about it, is just walking around in the real world. If you think about how people react to taxes, economists like to make the assumption that everybody pays attention to every tax in the economy.

COWEN: Right.

CHETTY: And my sense, from just having conversations with friends, is that, in fact, people have no idea on a day-to-day basis about what the tax rates they’re actually paying are. So I thought, how can we take that disconnect and ask in the simplest possible context to sales tax, what could be simpler than that? It’s much simpler than an income tax, for example. Are people even reacting rationally, so to speak, to the sales tax? And you end up finding that they’re not.

And my sense, from just having conversations with friends, is that, in fact, people have no idea on a day-to-day basis about what the tax rates they’re actually paying are.

COWEN: Would it make sense for suppliers to shove the sales tax into the price and then have their own add-on, which would vary? In equilibrium, can the suppliers do better by tricking customers more or by having additional shrouding, but not wanting it to be determined by the sales tax? Why are we in this funny equilibrium, where it’s always by the percent of the sales tax?s

CHETTY: Yes, I don’t think sales taxes are the only attribute that are shrouded. There are plenty of other fees you pay. For example, think about airline fees or cellphone fees. Often there will be a set of surcharges that you pay after the initial quoted price, or when you’re renting a car, or when you’re looking at hotel room prices. So sales taxes are one of many examples like that.

And companies not wanting to include the sales taxes is just an example of how they don’t want to include other prices as well. Another classic example is printers and the toner cartridges. You make the printer relatively cheap, but the toner cartridge is quite expensive.

COWEN: Do you believe that macro/micro literature seeming to show that retail prices are quite sticky? There can be demand shocks, and maybe for 18 months or longer that price won’t adjust in nominal terms. Do you think that’s actually what’s happening?

CHETTY: Yeah, I think there’s pretty good evidence that companies don’t change their nominal prices that often. Whether that’s literally because of menu costs or some other sort of inertia, I don’t know, but I think that evidence —

COWEN: Is there a salience story? Can your work help explain this? It seems it can’t be menu costs anymore.

CHETTY: Yeah.

COWEN: Not here in Silicon Valley, at least.

CHETTY: Yeah. But it could be menu cost more broadly written in the sense of it takes time to think about optimal pricing strategies and re-optimizing that problem every single day. Devoting labor to that might not be the efficient thing to do. Could it be about salience? That’s an interesting question. I think perhaps, although I would think the salience model would push you towards wanting to change prices more in some ways, because if consumers aren’t paying attention, in essence you’re facing an inelastic demand curve in the short run, and so you might want to catch them by surprise and occasionally have higher prices or lower prices.

I actually think sales have a lot to do with salience, if you think about all those tags you see at the store. It’s not so much the dollar price reduction. It’s the fact that the thing says it’s on sale, I think, that attracts people’s attention and makes them want to buy.

COWEN: I always tell my wife to ignore this if something’s always on sale.

CHETTY: Yes. [laughs]

COWEN: But for her it’s salient.

Now, another of your famous early papers was on corporations paying dividends, and yours is one of the first pieces — I think it’s from 2003 — where you show there’s more rationality to the dividend process than we used to think. If I think of what I learned in graduate school about dividends — this was before some of the later tax reforms — it seemed at the time there was no good reason to pay dividends at all, or at least there was no good reason to pay dividends and have the firm raise new funds through the capital market at the same time. That was the more general paradox.

Now, your paper shows that changes in dividends can be explained quite well, but do you think, especially in those earlier years before tax reform, that the existence of dividends at all is a puzzle, either statically or in the time series, or do you feel that also makes sense?

CHETTY: I think the existence of dividends itself is a bit of a puzzle, and our paper, which shows that changes in tax rates affect the amount of dividends firms pay out, doesn’t fully address that puzzle. So in particular, in the point you’re alluding to, is that you would think it’s more effective to repurchase shares and effectively return money to shareholders.

COWEN: Or do nothing.

CHETTY: Or do nothing, thereby avoiding the dividend tax and just paying the capital gains tax. In practice, companies pay dividends. There have been explanations put forth, which I think are plausible. For instance, you need to constrain managers — the amount of capital managers have on hand — because otherwise there’s a tendency to empire-build and invest in projects that aren’t as efficient from the shareholder’s perspective as you might otherwise like.

And dividends, for some reason, are viewed as a commitment. If I’m going to start paying a dividend, I’m going to keep paying that dividend. If I choose to cut it back, it’s usually a negative, very negative, signal that a firm is sending. And so that forces managers to pay out money, which could be efficient from a shareholder’s point of view, if you think managers don’t have the right incentives.

COWEN: But that’s maybe why the real puzzle is not dividends per se, but the mix of dividends and managers who raise funds at the same time. It seems you’re pulling in with one hand, taking out with the other, and paying in that price in the net cash position. It would seem to be some kind of arbitrage loss because you’re paying more tax than you need to.

CHETTY: I do think you can explain it though by temporary needs for cash, which make you want to raise equity, but then wanting to have a steady commitment to pay that money back, signaling to shareholders you’re not just trying to collect money in order to invest in some pet project or buy a private jet. You have a commitment to actually delivering a return and paying back the dividends.

What we found in that paper is that the sense in which there’s some rationality here is that managers who own a larger fraction of shares and thereby have interests that are more aligned with shareholders’, they tend to pay more dividends — exactly as you might think.

On the value-add of kindergarten teachers

COWEN: One of your best-known papers, series of papers, is on the value of kindergarten teachers. Let me ask you a few questions about this. The result seems to have held up really quite well, even though some people were skeptical to begin with, and I’m interested in what the mechanism is.

As you know, much better than I do, there’s a lot of measures of teacher quality that don’t seem to affect future earnings of the children in the kindergarten class. And the one measure that does matter is if the teacher has been teaching for more than 10 years. If that’s the case, again oversimplifying, it’s possible that’s worth as much as a quarter million dollars a year in value for this typical kindergarten class if they have a teacher with more than 10 years of experience.

But the other measures of that teacher quality don’t matter, and the results you get measuring those children in grades one, two, and three, they don’t seem to be doing better because they had the kindergarten teacher with more experience. In terms of what’s really going on in those numbers, what do you think is the best story for the Chetty result?

CHETTY: Well, if I can add one bit of nuance there —

COWEN: Sure.

CHETTY: So, experience seems to matter, but then the other thing, more generally, I would say is a strong predictor of teacher quality is so-called value-added measures, which are based on test scores — so teachers’ impacts on students’ test scores. If they raise their students’ test scores a lot, we’d call them a high value-added teacher. If they reduce their students’ test scores on average, we’d call them a low value-added teacher. And these value-added measures, along with experience, seem to be pretty strong predictors of students’ later success.

Now, as you noted, what is quite puzzling and surprising about this result is that prior studies in our own work show that if you look at intermediate outcomes, like test scores and later grades, you don’t find much of an impact, but the impact reemerges when kids become adults.

So why is it that you see this kind of puzzling pattern of effects that disappear and then reappear in adulthood? Our best hypothesis is that, what’s going on is that the teacher is not teaching you skills that end up showing up on later tests. It’s not so much that they’re teaching you how to do math better, or teaching you trigonometry or something in later grades. Rather, it’s that they are teaching you things like discipline and social skills and other things that are valued in the labor market directly.

Our best hypothesis is that, what’s going on is that the teacher is not teaching you skills that end up showing up on later tests. It’s not so much that they’re teaching you how to do math better, or teaching you trigonometry or something in later grades. Rather, it’s that they are teaching you things like discipline and social skills and other things that are valued in the labor market directly.

COWEN: But it doesn’t help you on second-grade tests? That seems odd.

CHETTY: Yeah, in something like discipline, that seems a little bit odd, but take social skills. For example, suppose the kindergarten teacher helps you get a higher test score in kindergarten, but in the process of doing so, she’s also a teacher who is able to manage the classroom effectively and help kids get along with each other. All right? Now, getting along with somebody else is probably quite valuable in the labor market. However, it’s not necessarily going to have a direct payoff in terms of helping you do better on your third-grade test.

Now, this is a bit of conjecture — I don’t know whether that’s exactly the mechanism — but what I can tell you is what does seem to persist are these gains in terms of student’s non-cognitive skills. There are subsequent surveys, which in the context of the study we did of 12,000 kids in Tennessee, they assessed these children’s ability to get along with other kids, their ability to sit still, various non-cognitive measures. And you see that, if you were in a better kindergarten classroom, you’re doing better on all of those non-cognitive dimensions even though your academic performance is no longer better. That is suggestive evidence that that’s the type of mechanism that matters here.

COWEN: The teacher mattering after only 10 years of experience, that to me also seems counterintuitive. I would think it normal if teachers needed a few years to learn how to teach. But I almost would think their skills would peak years three through seven, and after that, there’s a bit of burnout, like “Oh, my goodness. These kindergarten kids, I can’t take it anymore,” that somehow, years seven through nine, they’re like the immature teachers.

CHETTY: Sorry, sorry, let me clarify there. That is, in fact, in a subsequent study, that’s actually the pattern you see in the data, so it’s the first couple of years of experience that matter the most, so novice teachers don’t do as well as slightly more experienced teachers, but then you see essentially a flat profile.

The above-below 10 that you’re referring to in this earlier kindergarten STAR paper was just a simple fifty-fifty split of the data, which, when you have a smaller dataset, where when you’re running an experiment, that’s the most natural way to maximize statistical power and split the data.

In a subsequent study where we study all children in the New York City School District and we have two-and-a-half-million observations instead of 12,000, you start to see that it is indeed the first two years that matter the most.

COWEN: Do you think this means it makes sense to pay teachers on some kind of seniority basis?

CHETTY: No, actually. I think paying teachers who are more experienced a bit more can make sense — the novice teachers versus teachers who have two or three years of experience. However, the current system of pay essentially is a function entirely of experience, so the more experience you have, if you have 5 years, or 7 years, or 10 years, or 15 years, your pay continuously goes up. Your pay does not depend in any way on any measure of performance.

As I was saying earlier, while experience is one predictor of teachers’ impacts, in a way a much more important predictor are these value-added or other types of measures of performance. And I think tying pay or bonuses or efforts to retain teachers to performance measures — whether they’re test-score-based, value-added measures, or principal ratings or something else — is quite sensible. That is not something that’s currently done in most US school systems.

COWEN: It’s a common view heard anecdotally, though I’ve never seen it tested, that kindergarten teaching in this country became much worse once smart women had better labor market opportunities. Do you think that’s true?

CHETTY: Yes, and that also might reflect the earlier finding, where we find the teachers who had 10 or more years of experience seem to be better teachers. They are also coming from an earlier cohort relative to the teachers who have less experience who are becoming teachers more recently. So, one of the things you see from other research is some evidence that the quality of teachers in earlier cohorts, where women were more likely to go into teaching — high-skilled women — was much higher. So I think that’s plausible.

COWEN: If we’re trying to separate out the influence of environment on children, young children, what’s your opinion of twin studies? There’s plenty of studies of genetically identical twins , but due to adoption, they’re raised in different families, and then you can look at them down the line and see where they’re similar, see where they’re different. If I understand those papers correctly, they seem to suggest the environment doesn’t matter all that much. Are you convinced by those, or do you feel that’s somehow a select sample of people we’re looking at and it’s not an indicative result more generally?

CHETTY: Well, there are a number of twins studies which show that there are significant differences in kids’ outcomes depending upon the conditions in which they were raised.

COWEN: You can wreck them easily, but if they’re both in decent families, they seem to do more or less the same.

CHETTY: Yeah. To me, the most convincing piece of evidence along those lines is studies of adopted kids actually, which directly try to assess the effects of environments. I’m less a fan of the twin studies because I feel like you don’t have a clear sense of why one twin ends up in a certain condition versus another twin. You don’t have an experiment there; you don’t know what’s driving that variation.

In contrast, there’s some very nice work by Bruce Sacerdote, an economist at Dartmouth, looking at Korean adoptee kids who were effectively randomly assigned to families because of quirks and ordering of families’ applications and so forth. And he shows very convincing evidence that, if you happen to be an adopted kid who ends up in a family that’s more educated and more affluent and so forth, you do substantially better in adulthood than if you end up randomly assigned to a less educated, less affluent family living in a worse neighborhood. So I think that’s really clear evidence that, conditional on genetics, environment really matters.

On ways to improve primary school education

COWEN: It’s a common view, derived from William Baumol and Bowen, that education is subject to a kind of cost disease, that it’s harder and harder to augment productivity, wages rise in other sectors of the economy, education takes a rising share of GDP but doesn’t really get much better. Do you accept that story, or, if not, how would you modify it? Are we doomed to low productivity growth in K–12 education?

CHETTY: I don’t think so because, while in some limited case that might end up being true, at the moment I see so many opportunities within the US K–12 education system to potentially have significantly higher productivity without dramatically higher cost. Let me give you an example. Coming back to the case of teachers, my sense is, if we were to try to keep the most effective teachers in the classroom and either retrain or dismiss the teachers who are less effective, we could substantially increase productivity without significantly increasing cost.

COWEN: But say we do that. What do we do next?

CHETTY: I think eventually it’s conceivable that you move up the quality ladder, and you’ve got everybody getting a very good primary school education. Then you need to work on secondary education and so forth. But there again, I would say there are lots of bargains to be found.

In our most recent work looking at colleges and upward mobility, we see that there are a number of colleges where kids seem to be doing extremely well that are not all that expensive. Also, I think, here a macroeconomic perspective is useful. If you look at countries that have some of the best educational outcomes, like Scandinavian countries, they’re not actually spending dramatically more than the United States.

At some abstract level, I think that logic has to be right, that eventually, in order to raise the level of education beyond some point, we’re going to have to spend more and more on that, but I don’t think we’re close enough empirically to such a point that that is really a critical consideration at the moment.

COWEN: Some people I know, who studied the Finnish system and who also know the South Korean and Singaporean systems, claim that simply giving teachers a higher status, especially in smaller economies that may not have quite the opportunities that smart people would have in the United States, has a big value-added to K–12 education. What’s your opinion of that?

CHETTY: Yeah, I think status seems incredibly important. My sense of the K–12 education system in the US is, unfortunately for many kids graduating from top colleges, teaching is not near the top of the list of professions that they’d consider. It’s partly because, in a sense, they can’t afford to be teachers because it entails such a pay cut. But also because they feel that it’s not the most prestigious career to pursue.

One example of that, that I think illustrates that point, is take the Teach for America program. Teach for America is incredibly successful in getting very talented kids to go into teaching, at least for a short time. I think the issue with Teach for America is, those kids tend not to stay in teaching after a couple of years. Now, what I think is striking about Teach for America is, the reason it’s so effective in attracting top talent is precisely because it’s very difficult to get into Teach for America. I think kids at top institutions like to do things that are hard to do, basically.

“Now, what I think is striking about Teach for America is, the reason it’s so effective in attracting top talent is precisely because it’s very difficult to get into Teach for America. I think kids at top institutions like to do things that are hard to do, basically.”

COWEN: It’s like a Giffen good. It’s more of a signal then.

Some European systems such as Germany, they have much more extreme forms of tracking than we do. So you get put on an on-college track. You are evaluated early. What’s your opinion of those systems?

CHETTY: My feeling is too much tracking at an early age is detrimental because children bounce around. Test score measures are imperfect. There is no way you can perfectly gauge a person’s trajectory at a very young age. However, I think there’s some exceptions to that. At the very upper end of the distribution, gifted and talented programs, I think there is good evidence suggesting that they can be extremely effective in promoting those kids. Learning and getting them on a path to be kind of the superstars in the economy who are starting the new companies, discovering the new drugs, and so forth.

So, while I would not advocate tracking at the broadest level across 95 percent of the distribution, I think there can be a role at the very top to give attention to gifted and talented kids, especially, I would say, from disadvantaged backgrounds, where often you don’t have the stimulation and the resources, if you’re a very smart kid at an early age.

COWEN: Is there a tension here with your own work? Because part of what you are doing is trying to explain, or you could almost say predict, the mobility children will experience. And that’s fairly well predicted. We might think it’s wrong that it can be predicted as well as it is, but people actually don’t move to better neighborhoods as often as we would like.

Isn’t, in a sense, that work potentially a rationale for tracking? Not a normative — this isn’t the way the whole world should look — but in the sense of we can identify these children, run them through the Chetty equations, and then track them. Yes, no?

CHETTY: To some extent, but that’s why I was emphasizing the upper tail. I’m not convinced we can actually identify these children in terms of their potential outcomes at a very early age with a great deal of accuracy. I think there’s some predictive power. You can explain maybe a quarter of the variation that you’d see in adulthood with early childhood test scores and things like that. There’s certainly strong predictors, but by no means are they . . . They don’t tell you your fate.

There’s lots of variation even conditional on how well you’re doing early in childhood. So it’s important to recognize that and not condemn people to different tracks purely based on how they are doing at the earlier stages. Again, with the possible exception of kids in the upper tail, where I think it requires a lot of investment to get children to the point where they’re going to be highly productive members of society — having positive spillovers on others — and so there is a rationale to potentially invest there.

COWEN: What’s the most memorable thing a K–12 teacher did for you?

CHETTY: Well, I remember in high school being influenced by a number of teachers. One anecdote that comes to my mind off the top of my head, in the context of this conversation, is a history teacher I had who taught an economics class in my high school. He said, “You know, Raj, one day you are going to be on the Federal Reserve Board.” And I remember thinking to myself, “There is no way because I am going to do biomedical science. This teacher clearly doesn’t know my interest.” And I look back upon that and think, “Well, he must have thought that I would be good at economics or interested in economics or something.”

On things under- and overrated

COWEN: There’s a segment in all of these conversations, in the middle; it’s called overrated or underrated. And I’ll toss out a thing, a name, a something, and you feel free to tell us if you think it’s overrated or underrated, and you are free to pass on any of these, of course.

First would be the Mozart of Madras, A. R. Rahman, the Indian musician. Overrated or underrated?

CHETTY: I’d say underrated.

COWEN: Why?

CHETTY: I think he’s had a huge influence on Indian music and bringing it to the Western world in particular. And I like his music very much.

COWEN: What about it speaks to you?

CHETTY: What about it speaks to me? I think the combination of Eastern and Western influences on music. Where there is more of a beat basically in Western music than in traditional Tamil and Carnatic music, and A. R. Rahman really was the person who saw how to bring those together.

COWEN: Many westerners know Bollywood, but Tamil cinema actually produces quite a few films. Overrated or underrated?

CHETTY: Overrated.

COWEN: Why?

CHETTY: Because there tends to be very little variety in what those movies are about, at least traditionally. It’s typically about marriage. That’s what a lot of these movies revolve around, and arranged marriages, and kids not wanting to have a particular arranged marriage, and ending up in a different situation. There is kind of a recipe to a lot of these movies.

COWEN: Given your work on mobility and all the data you have looked at, what do you feel is the most underrated state in the United States?

CHETTY: Utah.

COWEN: Utah. Why?

CHETTY: Salt Lake City, Utah, is maybe not a place you’d think of as having the most vibrant economy and so forth, but it’s a terrific place for upward mobility. It always shows up in the top of our data in terms of rates of upward mobility, health outcomes — lots of different dimensions. Maybe it makes sense ex post when you think about things like the community and the Mormon Church and so forth. But certainly ex ante going into it, if you had asked me, where do you think the place with the highest upward mobility is going to be, I would not have said Utah.

COWEN: I’ve often argued more Mormonism would be good for this country.

CHETTY: [laughs] Right.

COWEN: And good for the middle class.

Atul Gawande, and indeed your mother, I think, actually endorsed this idea in an interview: the notion that you get things done by having a checklist and checking them off as a way of being more productive. Overrated or underrated?

CHETTY: Underrated.

COWEN: Why?

CHETTY: Checklists I think are incredibly useful to stay organized and solve complex problems. The way to make progress on complex research problems, for example, is to divide them into smaller, manageable problems, and then go down a list and solve that checklist.

COWEN: In music, the group the Piano Guys, speaking of Mormons. Overrated or underrated?

CHETTY: Underrated. I love the Piano Guys.

COWEN: Why?

CHETTY: I think the Piano Guys are great in terms of doing renditions of popular songs.

COWEN: Not too triumphalist? Do you mean the major chords?

CHETTY: Maybe in some cases, but I like them.

COWEN: Bhindi or okra. Overrated or underrated?

CHETTY: [laughs] Underrated.

COWEN: [laughs] I’m also a fan. What’s the best way to cook it?

CHETTY: There are many Tamil preparations of okra. It’s a popular dish and supposedly, it’s supposed to make you smarter. So I ate a lot of okra when I was a kid, trying to get that.

COWEN: So let’s say a friend of yours, non-Indian, is going to make a trip to India; maybe they’ve even been there before. And they ask you, “Where should I go, other than the obvious sites, Taj Mahal and Mumbai?” Plenty of obvious things they might do. But they want a tip from you about how to better appreciate, understand, see India. What place would you recommend to them? And what insight would you package along with that recommendation?

CHETTY: I would say the South. I think most westerners’ perception of India is based on North India: Taj Mahal, Delhi, Agra, so forth.

But going to places in Tamil Nadu, going to places in Karnataka, increasingly . . . If you think about Silicon Valley, the corresponding place in India is in Hyderabad and the tech sector there. If I were to pick a specific place, the area where I’m from, there’s a city called Madurai in Tamil Nadu, which is a typical Tamilian city, and you get a very different feel for what India is relative to what you’d see in Delhi.

What is striking is that India is really a combination of many different countries, but completely distinct languages, completely different food, culture, and customs, and so forth. So what I would suggest is visiting at least two or three different states so you can see the diversity in the country.

COWEN: Chennai is actually my favorite Indian city, it may please you to know. So I agree with that.

As you probably know, in the history of South India, there’s this long-standing tradition of mathematical prodigies. Your work isn’t exactly mathematical, but broadly speaking, of course, it is. And why is it you think there’s this connection between South India and this history of mathematical prodigies? Do you have any thoughts on that? Is it something about the abstract nature of the philosophical thinking embedded into South Indian Hinduism?

CHETTY: Yes, it could be. The name that comes to mind most prominently is Ramanujan, The Man Who Knew Infinity . And I think that type of thinking is encouraged once you have a tradition of people in that vein. I remember hearing a lot about Ramanujan since I was a little kid. And so you look up to that kind of thing, and maybe that influences subsequent generations of children as well. I think that kind of mechanism is actually extremely important in terms of what children choose to do.

COWEN: There are a number of writers, as you know, who’ve tackled the highly diverse Indian American experience. One of them is Jhumpa Lahiri, who’s actually been a guest on this podcast series. But of that whole literature, who or what in it speaks to you the most, given that you have your own Indian American history?

CHETTY: There are a bunch of books that broadly focus on integration of Indians in Western society and what that entails and how different America is from India. What speaks to me is that there are obviously quite significant differences between the two cultures.

But the experience I’ve had is that America’s been an incredibly welcoming place since I was a kid. When I first came to the United States, it took some time to adjust, as it would for any child coming from a completely different background. But I found, on the whole, Americans to be very interested in Indian culture and to be very welcoming. That’s the experience I’ve had. I know other people have had different experiences, but to me that’s the main thing I think of.

COWEN: In 1969, your father published a pretty well-known piece in the American Economic Review, called “On Measuring the Nearness of Near-Moneys.” And it implies there are a lot of close substitutes for moneys. So, in light of this, given that India has recently tried a demonetization, which has been highly controversial — the Indian demonetization, would you say underrated or overrated?

CHETTY: I’ll have to pass because I don’t have a good sense. [laughs]

On drivers of upward mobility

COWEN: Let’s go now to some of your research on mobility, which is maybe, at this moment, what you’re best known for. You can identify counties or parts of the United States where mobility for generations is going to be especially high. To what extent do you think that’s picking up that simply some of those regions end up with resource booms or other good events that is, in a sense, just random? It doesn’t per se have to do with the region? Or do you think we can adjust for that?

CHETTY: Yes. Some of what drives upward mobility, of course, is just having a very vibrant economy. To give you an example, parts of North Dakota, with the natural resource boom there, we see are having very high upward mobility. Of course if you discover natural resources, that’s going to help more people move up the income distribution.

But by and large, that is the exception rather than the key driver of the differences in upward mobility that we find across places. I say that for a couple of reasons. First, even if you hold fixed the rate of growth, the rate of economic growth, you find that some places have much higher rates of upward mobility than others.

To give you an example, Atlanta is a city that’s booming in terms of jobs and economic growth overall. But Atlanta’s one of the places with the lowest levels of upward mobility for kids growing up in low-income families there.

The second thing you see is these rates of upward mobility, to the extent we have data, they tend to be quite persistent overtime. It’s not like the places that have high upward mobility in one decade, suddenly a very low upward mobility in the next decade. It’s a pretty persistent phenomenon.

A striking example of that is states in the middle of the US, like Iowa for example, which historical data going back to work by Claudia Goldin and Larry Katz has always looked like a place with very good outcomes for kids in lower-income families. And what’s amazing about that is, Iowa suffers from a brain drain phenomenon where the most successful people often end up leaving the state, going to Chicago, going to New York, to get higher-paying jobs. Yet generation after generation, Iowa seems to produce very good outcomes for low-income families. So that again suggests it’s not about natural resources or temporary booms. It’s something more persistent.

COWEN: And getting back to the kindergarten teacher question. The physical mechanism as it operates through geographic space, what does Iowa do to you, so to speak?

CHETTY: Yeah.

COWEN: If you told the story about molecules impinging on your body and impelling you to action, what’s the best story you can come up with for Iowa, say, or Utah?

CHETTY: Yeah, a few different things. Iowa is known for having very good public schools for a long time.

COWEN: But that too is arguably just part of the package.

CHETTY: Yes. Where did that come from? Why does Iowa have good public schools?

COWEN: Right.

CHETTY: One of the strong correlates we find is that places that are more integrated across socioeconomic groups, that have lower segregation, tend to have better outcomes for kids. And that kind of thing in a rural area — you can see why that occurs and why it might lead to better outcomes.

If you live in a big city, it’s very easy to self-segregate in various ways. You live in a gated community, you send your kids to a private school. You essentially don’t interact with people from different socioeconomic classes. If you live in a small town in Iowa, pretty much there’s one place your kids are going to go to school. There’s one set of activities that you can all participate in. And that is likely to lead to more integration.

If you live in a big city, it’s very easy to self-segregate in various ways. You live in a gated community, you send your kids to a private school. You essentially don’t interact with people from different socioeconomic classes. If you live in a small town in Iowa, pretty much there’s one place your kids are going to go to school. There’s one set of activities that you can all participate in. And that is likely to lead to more integration.

COWEN: And you think that’s causal rather than just restating the same fact about the quality of the place?

CHETTY: We don’t have definitive evidence on this, and we’re working on trying to establish clear evidence. But our sense is that integration — actual contact with people from other backgrounds — is a strong predictor and likely a causal determinant of kids’ long-term outcomes. I suspect that that’s one major factor in what’s going on.

COWEN: Now, as I read your 2014 papers on mobility, they seem to suggest that, at least in recent times, intergenerational income mobility in this country actually has not been getting worse, contrary to what many people thought was the case. Your former Harvard colleague, Robert Putnam, in his latest book, he basically wrote he doesn’t believe the numbers. He’s not faulting your work. He just thinks, given the return to higher education is rising, there’s wage stagnation, plenty of evidence of regional decay, that it must somehow be the case that intergenerational income mobility is getting worse; we’re just not measuring it properly. Do you agree with Putnam? Or do you agree with your own numbers?

CHETTY: I think our numbers are right. The fact is that relative intergenerational mobility — by that I mean, your chances of moving up or down in the ranks in terms of percentiles — has not changed significantly over the past 40 years or something like that. That is, if you’re born to parents in the bottom 20 percent, your odds of making it to the middle class or to the top 20 percent look roughly the same today as they did 30, 40 years ago. Now, why is that the case?

I think Bob correctly points out that there are lots of trends going in a direction that might have led you to think that this would be getting worse over time. For instance, he shows in his work that the rich are investing more in their children’s education, spending more time reading to their kids, for example, relative to kids from lower-income families.

But you also have to remember that, at the same time, over the past 50 years, we have been fighting the War on Poverty and investing more in kids from lower-income backgrounds, trying to expand access to higher education. Expanding pre-school programs, Head Start, and so forth. So the fact that you don’t see any change in relative mobility in light of all those major social investments, from that perspective, it’s actually surprising that things haven’t improved. My view is you’ve got these two countervailing sets of trends, and that’s left us at a standstill in terms of relative mobility.

However, in our more recent work, we show that absolute mobility — your chance of doing better than your parents, which arguably is what people really think of as the American dream — that has fallen dramatically over the past 50 years.

COWEN: And that’s your most recent paper from December?

CHETTY: Yes.

On geography and gender

COWEN: Yes. Have you thought much about within this country, geographic differences in gender inequality? This is a big issue in India. People in South India will tell you, “It’s better for women here than in North India.” I haven’t seen data, but I find that believable. Where in the United States does gender equality come most naturally and where not? Do you have a sense? Is it an interesting question?

CHETTY: Yeah, that’s a very interesting question. We find sharp differences in outcomes by gender across areas for various reasons. Let me give you a couple of examples. One, we find that areas with more concentrated poverty — take the city of Baltimore, for example — we find very poor outcomes for boys in particular, relative to girls, and we think that that has to do with crime, and getting involved in gangs, and so forth — things that girls are less likely to do.

As a result, growing up in a place like Baltimore turns out to be extremely detrimental for boys. We estimate that you lose something like 30 percent of your earnings relative to if you’ve grown up in an average place in America. Whereas for girls, it’s slightly negative but not nearly as bad. There are a set of urban ghettos, places with concentrated poverty, that tend to have particularly negative outcomes for boys.

There are also other phenomena that are more subtle, related to things like marriage patterns. Relating this back to personal experience, I remember when working on these issues and thinking about our decision to move from Harvard to Stanford. At the time, we actually were expecting our first child, a daughter. And I noticed in our data that, for kids in affluent families in the Bay Area, daughters tend to have very low household earnings. And I found that kind of curious and we spent some time trying to dig into why what was, partly given my personal interest in the issue.

COWEN: So, your own moving decision was influenced by this research.

CHETTY: [laughs] In some ways.

COWEN: Yeah.

CHETTY: What you find is an interesting explanation, which is, if you look at individual earnings rather than household income, girls growing up here in the Bay Area do extremely well. However, when you look at household income, they don’t do so well, and that’s because they’re much, much less likely to be married than if they grew up somewhere else.

COWEN: Yes.

CHETTY: So if you’re in your mid-30s, only something like a quarter or less of girls growing up in the Bay Area are married, and we show in our paper that every extra year you spend growing up in the Bay Area, you’re less likely to get married. I remember telling my wife, “I don’t think we need to worry. Our daughter will be fine in terms of earnings. It’s just that she might not be married if we move to California.”

COWEN: So, you’ve lowered your expectations for grandchildren?

CHETTY: Yes. [laughs]

On future research

COWEN: Last question or questions, sort of a bundle. Now that you live here — you’re ensconced at Stanford, up and running again, you’re at the heart of Silicon Valley — what are your plans for future research and questions? Both right now, but even more speculatively, say, five or ten years hence. But start with the right now. Then we’ll get to the longer term.

CHETTY: One of the things we talked about, Tyler, is digging more into whether segregation and integration are really causal mechanisms that are driving differences in rates of upward mobility. One way we’re trying to get at that, which is very much in line with thinking about Silicon Valley, is using Facebook data.

In the past, people like Bob Putnam and others have talked a lot about social capital and integration as being a key determinant, but we’ve never been able to measure those things as precisely as we’d like to be able to. And Facebook data, or more generally social network data, gives us the most precise lens to look at how people are connected and how that might influence their long-term outcomes.

COWEN: So you have a Chetty conceptual trick — that maybe you won’t tell us — that will enable you to convert Facebook data into a causal proposition, that you will test and then execute on, using the Raj Chetty production function. [laughs]

CHETTY: That’s the hope.

COWEN: That’s the hope.

CHETTY: We’ve got a team of collaborators there, and we’re starting to explore the data and figure out, not by running experiments of course, but just with the existing de-identified observational data, Can we find patterns? One example: In the cities that have very high rates of upward mobility, do we see that there are more links between kids from low-income and high-income families? That would be suggestive.

But then to get closer to causal inference, one way you might be able to do that is to exploit the fact that kids tend to be friends with children in the same grade, but not friends with kids who are one grade ahead or one grade behind. Right? Now, the network of kids you’re exposed to is going to tend to fluctuate across grades just by chance. You might have more low-income kids, more girls, more talented kids, etc. And so that’s essentially going to be a random variation from the perspective of a given child. And you can see how you might be able to exploit that type of variation to understand whether the set of people you’re connected to ends up affecting your life trajectory.

That’s one early-stage idea we’re thinking about developing in that space.

COWEN: And longer-term, more speculatively, what do you dream of working on, even if it might seem highly impractical or something you don’t feel you know enough about just yet?

CHETTY: At a very big-picture level, what in our research group we’d like to figure out is, how we can increase upward mobility. We’d like to come with a recipe for how you can increase upward mobility in specific cities. In the long term, you might think that important recent trends like automation, driverless cars — these things could end up reshaping what cities look like. Maybe you’d be able to change how integrated people are taking advantage of driverless cars, which would allow people to commute in very different ways.

Thinking about how technology could be used to tackle some of these social issues strikes me as an important thing that I’d like to be working on. What I see here in Silicon Valley is that technology has completely transformed the way we live our lives through the private sector. I think it’s, to date, had less of an impact on social issues, and there’s a big arbitrage opportunity there.

Thinking about how technology could be used to tackle some of these social issues strikes me as an important thing that I’d like to be working on. What I see here in Silicon Valley is that technology has completely transformed the way we live our lives through the private sector. I think it’s, to date, had less of an impact on social issues, and there’s a big arbitrage opportunity there.

COWEN: As I’m sure you know, since the 1990s, segregation by income has been rising in this country. And here, Silicon Valley is one of the most extreme cases of that. So seeing that, are you on net a segregation optimist or pessimist? If I may ask.

CHETTY: I think current trends suggests that segregation will continue to grow in the US. Take the case of driverless cars, for example. One way that could go is, if you have access to driverless cars, it makes it all the more easy to go live further away in a secluded place, further reduce interaction, right?

So I think it’s very important to think about social policy in the context of that type of technology. How do you set cities up? How do you do urban planning and architecture in a way such that you don’t actually just facilitate more segregation? Such that you make it attractive to live in a more mixed-income community? That’s a key challenge, I think.

COWEN: Raj Chetty, thank you very much for coming to chat with us. It’s been a pleasure.

CHETTY: Thank you. My pleasure.