Capitalisn’t: Surprising New Insights on How Children Succeed

Raj Chetty: There is still an enormous divide in terms of where Black kids and white kids end up—even conditional, critically, on being of the same class. And so, I wouldn’t want our most recent study to be interpreted as saying it’s now only class that matters, and race is much less important. Directionally, that is true. If you project forward, that’s going to be just a key dividing line in the United States.

Bethany: I’m Bethany McLean.

Phil Donahue: Did you ever have a moment of doubt about capitalism and whether greed’s a good idea?

Luigi: And I’m Luigi Zingales.

Bernie Sanders: We have socialism for the very rich, rugged individualism for the poor.

Bethany: And this is Capitalisn’t, a podcast about what is working in capitalism.

Milton Friedman: First of all, tell me, is there some society you know that doesn’t run on greed?

Luigi: And, most importantly, what isn’t.

Warren Buffett: We ought to do better by the people that get left behind. I don’t think we should kill the capitalist system in the process.

Bethany: Even to me—a noneconomist, horrors!—Raj Chetty is almost a household name. He’s an economics professor at Harvard and the director of Opportunity Insights, which uses big data to study economic mobility.

How and under what circumstances do children move up the income ladder and make it to a higher rung than their parents? And, if they don’t, what can we do differently?

Luigi: For people who are not aware, he has access to the IRS data. Our tax records are anonymized, so he doesn’t know how much you and I make, Bethany, but he can use this data to actually see how children of people with a certain income are behaving 30 or 40 years later.

Bethany: Essentially, outcomes improve for children who grow up in communities with increasing parental employment rates. This may sound obvious in some ways, but no one had shown it systematically before.

Chetty has just furthered his work with a brand-new paper in July showing how quickly this can change. He and his team used IRS data from 57 million children to show that for children in the US born between 1978 and 1992, earnings increased for children from high-income families but decreased for children from low-income families.

Earnings increased for Black children at all parental income levels, actually reducing the gap between low-income Blacks and low-income whites. The overall gap between what Black people make and what white people make, on average, remains large.

For children born into high-income families, household income increased for all races between birth cohorts. Among those from low-income families, earnings rose for Black children and fell for white children.

Essentially, the researchers found that Black millennials born to low-income parents had an easier time rising than the previous Black generation did. At the same time, white millennials born to poor parents had a harder time than their white Gen X counterparts. Chetty has said this: “Class is becoming more important in America.”

Luigi: Chetty and his coauthors do emphasize that this is not a zero-sum game, where mobility for poor Black people is improving only because mobility for poor white people is falling. Indeed, places where opportunities for Black children improved the most are also where white children did the best.

But you can see how there can be an argument for everyone, especially as we head into a presidential election. Conservatives have long argued that white working-class Americans are falling behind, while progressives point out that Black and brown people remain far behind their white counterparts and, therefore, need more help from social programs.

The study provides an argument for both sides. But this study also shows that, far from being fixed, opportunities within a place can change significantly and rapidly. Neither history nor place is destiny, or, at least, not an absolute destiny.

On a less-positive note, Chetty’s work also continues to show that the top of the income distribution remains ossified, particularly for white kids. Rich white kids are overwhelmingly likely to remain rich, and poor Black children continue to have only a 3 percent chance of rising from the bottom to the top quintile. For poor white children, chances have fallen from 14 percent to 12 percent. Whereas nearly all American children born in the 1940s could still expect to do better than their parents, only two in five could do that by 1984.

Bethany: Let’s bring Chetty himself on to address our questions.

I wanted to start with a pretty basic question. What has motivated your work into social mobility? Why do you think this became a passion of yours? Was it accidental, or was it something that you set out to do?

Raj Chetty: I think it’s partly my personal background. I was born in India, lived there until I was nine years old, and then came to the United States with my parents. Like many immigrants, especially those who come from less-developed countries, you obviously observe the tremendous contrast between the different places.

I was struck by the opportunities, and also, more broadly, in my extended family, seeing how things like the fact that my parents got a chance to get a higher education, which was very unusual in the low-income villages where they grew up in south India, has had a cascading effect over the generations in our family. I think that kind of experience has motivated my interest in understanding what helps people thrive. And then, after doing a PhD in economics, I realized I could study those questions more systematically.

Bethany: We always think—at least, I did—mobility, good. But why does mobility matter?

Raj Chetty: I think mobility matters for two main reasons. First, the degree of mobility fundamentally controls the degree of disparities, which people clearly care a lot about.

To take a concrete example, very relevant in the US context are racial disparities. Because Black kids are less likely to move up and also less likely to remain at the top if they are born to a high-income family, they’re in a perpetual treadmill, where even if you try to close racial gaps at any one point in time, they’re going to end up re-emerging down the road. You can make that same argument on many different dimensions: geographically, by gender, by ethnicity, and so on.

A completely separate reason to care about mobility is that it matters for productivity and growth. Here, I like to give the example of a paper we’ve written on who becomes an inventor in America and this notion of lost Einsteins. The idea that there are lots of kids out there who could be doing great things, not just for themselves, but also for society—maybe inventing a new drug, starting an important new company—but because of a lack of opportunity to rise up, don’t end up coming through the pipeline and doing that. I think that’s of interest to all of us.

Luigi: Let me play the devil’s advocate a little bit because I agree with you, mobility is important, but the question is, at what cost?

My favorite example is that Philip of Macedonia knew that his son, Alexander, would be king. And so, he got the best education for him, he got Aristotle to train him, and that’s the reason why he became Alexander the Great.

On the other hand, in mobile America, we have a bright kid named McNamara that climbs the ladder and becomes secretary of defense, and he doesn’t know anything about Vietnam, and he doesn’t know about the fact that the Vietnamese hate the Chinese more than they hate the Americans. To his own admission, had he known that, he would have been waging the war in a completely different way.

We have an enormous inefficiency because we don’t allocate resources properly. After all, if I’m not mistaken, your father is an academic economist as well, right? So, the reason why we have Raj Chetty is because you were, basically, serendipitously raised in such a great family. Had we forced equalization, we would have lost Raj Chetty. So, who are the Einsteins we would lose by enforcing equality?

Raj Chetty: Yeah, I think you raise a good point. I guess implicit in that argument, Luigi, is the idea that there’s heterogeneity in talent, maybe ability to make a significant contribution for various reasons.

I think the question is, to what extent is that talent distributed equally across the parental income distribution, across the racial distribution, by gender? Should it be the case that kids from high-income families have much better chances of getting into places like Harvard or the University of Chicago? Are they truly more talented and more likely to benefit from those resources? Or is it that we want to go more on the basis of things like SAT scores or other notions of merit and make sure that we’re admitting the most talented kids to these places, where they could really be launched into doing great things?

My perspective is not one of enforcing equality, as you put it, in the sense of, let’s have random admissions to colleges and have everyone have equal access to different institutions, because surely there are gains, as you’re noting, from having talented people matched to certain resources.

But rather, my sense is that at present in our society, we have gone maybe too far in certain dimensions, in not just putting a focus on talent, but also other forms of resources, wealth, other factors that play into the process that take us away from that meritocratic ideal that I think you were positing in your examples.

Luigi: But what is the optimal level of mobility you have in mind? Is the benchmark we should think about that from wherever you start, you should have equal chances of ending up wherever? This would mean that if you start at the bottom, you have an equal chance of getting to the top. If you start at the top, you have an equal chance to get to the bottom. And what are the costs of deviating from that?

Raj Chetty: That’s a tough and complex question. If you take one extreme of complete immobility, where your parents determine your fate, I think most of us would have the view that that doesn’t seem like a very fair society. It’s probably not a very efficient society in terms of productivity and growth. It doesn’t harness talent effectively.

To take the other extreme, where your odds of reaching the top or your outcomes are completely independent of where you started out, that also seems unrealistic and maybe in some cases undesirable.

To give you an example, take a case where you could have perfect mobility, where incomes are just determined by a flip of a coin. There’s nothing meritocratic about it. We just randomly determine how much people earn. That would be a society with perfect mobility. I think we would also think it’s probably not a desirable society in terms of other values we care about, like meritocracy or allocation of talent and so on.

There are many examples you can point to where we have higher levels of mobility, kids have better chances of rising up, than you do in the United States at present, where, arguably, you had better outcomes in terms of allocation of talent, productivity, and so on, in addition to less persistence of inequality.

My instinct based on that is more mobility than we have at present in the US is likely desirable. Whether we want to go all the way to perfect mobility, I think, is much less clear.

Luigi: Let me get into some of your results that are really fascinating. When I look at the Asian families, the Asian families seem to have a remarkable equality of opportunities, especially in the latest group. If I’m not mistaken, the chance of an Asian born in the first quintile to end up in the top quintile is almost 20 percent, which would be an equal distribution.

Number one, what is special about Asian . . . I understand I’m touching home here, but feel free to brag. What is special about Asian families, I guess, that they seem to be not affected by local shocks as much as anybody else? What can we learn from them?

Raj Chetty: When we dig further in the data, we find that that Asian advantage, so to speak, in economic mobility is really an immigrant advantage. It turns out that if you look at second-generation kids, so kids whose parents were born in the US of Asian descent, their rates of economic mobility actually look very similar to white Americans.

It’s really that first generation where you’re seeing exceptional rates of upward mobility for Asian kids. Now, why is that? One is a measurement issue, where you hear the anecdote of the engineer who comes to the US and can’t get a job commensurate to those classifications and ends up taking a job driving a cab or in the service sector. That person is going to look like they have very low income and low human capital, in some sense, in our data. Their kids may end up doing extremely well, but they’re growing up in a household where there’s a lot of education, a lot of knowledge, a lot of resources in a different sense. Maybe that’s why you see high levels of economic mobility.

But I think more broadly—partly speaking from personal experience, partly from the data, partly the work we’ve done, but also the work of other excellent scholars like Ran Abramitzky and Leah Boustan, who focus specifically on immigrants and economic mobility—we find that immigrants tend to gravitate towards communities that offer the best chances of rising up and often make choices that we see in the data translate to better outcomes for kids.

I’ll give you a couple of concrete examples. Often you will find immigrant families—Asian immigrant families, in particular—choosing the smallest house in the best school district so that you can get your kids the best chance of doing well.

I think that simple decision of where to live is extremely important, but it reflects a much broader set of decisions, where there’s an emphasis . . . The goal of coming to America to begin with was to rise up, and we’re just going to put front and center that goal. We’re going to do everything possible to make that happen.

I think that also maybe affects the social capital and the norms in those communities in the initial immigrant generation. Now, that doesn’t seem to persist, as I was saying, in subsequent generations, but I think there’s something to learn there about how mobility works for everyone.

Luigi: You seem to suggest that some ethnic groups or cultural groups, whatever you want to call them, prioritize their kids more than others. They’re willing to sacrifice today for the benefit of tomorrow. So, if we are trying to intervene from a social point of view to equalize mobility, we end up overpenalizing the people who have a long-term orientation and are willing to sacrifice their well-being for the well-being of the kids.

I am prepared to live in a crappy place because I want my kids to go to the best school. But then, because my kids are overachievers, they cannot get into Harvard because they’re Asian, and I need to give an equal opportunity to the ones who did not sacrifice. And this is perfectly consistent. There is nothing racial about this. It’s cultural, because once you get incorporated into American society, you lose your long-term orientation. We know there are societies that value this very much, and clearly, a lot of Asian societies are like this. American society, let me say, is not particularly oriented to the long term.

Raj Chetty: I think those cultural factors can be very important. One distinction that I feel is quite important in the argument you just made, and I want to emphasize, is between equality in terms of people’s starting points.

In the example you gave, you can have kids who start out in low-income families that have different orientations or different social communities and so on. I think you were suggesting that it might be unfair or not meritocratic to ex post penalize the kid whose family has invested a great deal in order to help them get ahead.

I can certainly see that argument. I think a lot of our focus has been on, how do you give the kids who are from those low-income families who are very interested in rising up, how do you give them access to the opportunities that will allow them to do so?

To give you an example, in the context of the K-12 school system, if we’ve got a society where school funding is based on local property taxes, and there’s tremendous stratification in terms of resources, so you basically can’t afford to go to a good school, or you have to go to a private school, in order to have access to those opportunities, that’s going to create a lot of stratification.

That’s not on the basis of ex post merit. It’s on the basis of ex ante income. I think that’s the kind of issue that it seems unambiguous is both unfair and inefficient.

Before we get to the more complex question of equality in the context of people with different levels of merit when they’re applying to college or entering the labor market, my view is that there’s a lot of low-hanging fruit in terms of inequality in opportunities much earlier in the pipeline that we could address.

Bethany: I was trying to fit some of your various pieces of research together, and I might be misconstruing it or trying to look for illusory coherence, but I wanted to push on one thing. A piece of your research pointed out that Black and white boys who grew up on the same city block and attend the same school still had very different outcomes in adulthood.

If community is what’s most important, why does this result hold true? And that was especially true for men. And so, if community is what’s most important, why does gender matter? Or do you see differences between boys and girls and how they react to community?

Raj Chetty: Yeah, it’s a great question. First, the way we think about community, both from that work and other subsequent work, is really who you’re interacting with—not just the place in which you live, which, of course, has a big influence on who you interact with because you tend be friends with the people around you, especially in childhood—but also, we find very clearly, in various data sources, that there’s a lot of stratification in society by race and by class.

A Black and a white kid growing up on exactly the same block, as in the example you gave, if you look at their social networks, they tend to be extremely different. Black kids tend to have more Black friends; white kids tend to have more white friends. And the same goes by income and so on, and by gender as well.

We have evidence from a series of studies showing that who you’re connected to and who you’re interacting with, especially in childhood, really seems to influence your trajectories. If you take the example of Black boys in particular, Black children tend to grow up in environments, unfortunately, where there are very few Black fathers who are present in the local community.

There tend to be very high rates of incarceration—historically, at least—and high rates of contact with the criminal-justice system. These issues, you might think, have a particular influence on Black men because they are the ones who are most directly in contact with those groups in a way that’s separated from white men or Black girls as well. There’s growing evidence that boys are especially influenced by male role models, for example, more so than girls. I think those very particular, community-level effects could explain the kind of divergence that we’re seeing.

Luigi: It’s interesting because you mention social capital very often, which, of course, I think is very important, but when it comes to policy implications, I don’t see you push that very much, even in the analysis. I would be very curious to know if, for example, the presence of civic associations or church attendance or any form of association that tends to be cross-class, or cross-racial groups, should help in that dimension. I’m trying to see why you don’t push this line. Is it because there’s no evidence or because it’s more difficult?

Raj Chetty: I agree with you that in our context on this issue of economic mobility, we’re finding that, specifically, the extent to which low-income people are connected to high-income people—what we call economic connectedness—that really seems to be what’s critical.

We find that that is determined both by the extent of segregation in a society, but it’s also, importantly, mediated, as you’re touching upon, by the nature of the institutions we have that connect people to each other.

In particular, we show using data from Facebook that people are much more likely to make friends that cut across class lines in certain institutions—in religious institutions, for example, or in recreational groups—than they are in their own neighborhoods, in their own workplaces, in their own colleges, and so on.

Then the question is, what exactly does that mean in terms of interventions? We’re not going to directly change, necessarily, how religious people are or how much they’re participating in sports, but maybe there are lessons for how we can reshape our institutions to create more of these bonds, along the lines of the example you just gave in Siena.

To take a more prosaic example, we find that the size of the groups in which people participate is also a strong determinant of the extent of cross-class interaction. Concretely, if you are in a very small school, you’re much more likely to connect with people from different backgrounds. If you go to a very big school, like these big, urban public schools that are often very diverse on the surface, they’re incredibly segregated underneath the hood because kids break into their own subgroups.

I think the intuition is very simple. If you imagine going to a party where there are 10 people, you’ll probably talk to everybody by the end of the evening. If there are 500 people, you’ll probably gravitate towards people like you or people you knew initially.

And so, I think that simple insight maybe suggests a more actionable course, which is, we could think about creating smaller cohorts, think about smaller schools. I think more deliberate efforts like that can be quite valuable.

Bethany: Is it an overstatement of your work to say parenting doesn’t matter? In other words, if it’s all about community, is parenting just incidental? I know the dangers of an anecdote of one, but I am an anecdote of one. I grew up in a very downwardly mobile place, a small mining town, where the economic bottom fell out in the 1970s, and I left, purely and simply, because of my parents. Don’t some of your results fly in the face of other studies that suggest that parenting can matter more than peers and community?

Raj Chetty: I don’t think anything in our data says that parenting doesn’t matter or even that it matters less than peers or community and so on. I think parenting likely matters a great deal.

But how are we supposed to change that through policy? There have been some efforts to educate parents, partnerships with parents, and so on. I think there’s some mixed evidence on their success.

But what I think is very clear, especially in terms of determining systematic group-level differences between Black and white Americans, between different immigrant groups, and so on, is that there are these community-level factors that are quite important, that I think can potentially be reshaped, at least to some extent, through policy, which is why we have focused on those.

I’ll also note that, exactly as in your own personal example, perhaps one of the most important decisions parents can make is where to live and where to raise their kids.

Bethany: A different wrinkle on the solution idea. You could, on the surface, I think, look at your research and say, OK, any parent who cares about their children should just pick up and move to one of these communities where lots of people have jobs, and lots of people are doing well. But given all these wrinkles that happen, as you say, all the stratification even within those places, is that the right answer? Or is that a deeply flawed answer?

Raj Chetty: I think the limitation there is that may not work for everyone. We’ve seen in some work we’ve done in Seattle, where we tried to make the housing-voucher program more efficient by helping families use those resources to move to higher-opportunity areas, that there were certain communities, for example, a big Ethiopian community, where they said: “That may well be good for the average kid, but we feel very comfortable in our current community, even though it may not look like as high an upward-mobility area on average. And so, we’d rather not move.”

I actually think that could be a very reasonable conclusion that, for that particular subgroup, maybe those moves aren’t going to be as beneficial.

And so, I think what that motivates, then, is thinking about not just moving—or desegregation is the broader way I think about it—as a pathway to creating more opportunity, but how do you bring opportunity to people where they currently live rather than bringing people to opportunity?

The honest truth is, from a scientific point of view, we don’t really know yet exactly how you do that effectively. But there are changes in terms of where we target job-training programs or where we site local businesses or how we design schools. Certain types of charter-school models have been shown to be quite effective. I think there are targeted things that we can do in communities that are struggling that can change their trajectory. And then, through these social-capital forces, you have more people who are doing well, who are employed, and so on, which will have knock-on effects in subsequent generations.

Luigi: The nugget I found reading your paper is that you use, at some point, county-level parental-income drop in unemployment to see the shock that you have at the county level. I couldn’t believe my eyes when I saw it. First of all, the drop among white kids can be as large as 15 percentage points, which is pretty dramatic.

The second, which I don’t think is mechanical—and correct me if I’m wrong—when you present the data, and you do it very well, you actually see that the white and Black kids almost don’t overlap, that the drop for the white kids is all to the left of the drop of the Black kids. What’s going on?

Raj Chetty: Yeah. What we’re finding is that opportunity is changing fairly rapidly in the United States over the past 15 years or so. Black kids, especially at the bottom of the income distribution, are doing better, while white kids born to low-income parents are doing worse, and kids born to high-income parents are doing better. So, racial gaps in economic mobility are shrinking, and class gaps are growing in the United States.

Why is that? As we’ve been discussing, kids’ outcomes are heavily shaped by the communities in which they’re growing up, and the communities in which Black kids have been growing up have been improving, as measured, for instance, by their parents’ employment rates. While the communities in which white kids, especially low-income white kids, are growing up have been deteriorating, as measured by their own parents’ employment rates. And those changes are quite substantial in magnitude.

I think that’s because of factors that other economists have written about over the years related to things like globalization, technological change. But importantly— and this I find quite interesting—among many white Americans, especially those with lower levels of education, when those shocks occurred, rather than switching to different sectors, many people dropped out of the labor force entirely and just stopped working.

And so, in those communities, kids are now growing up in an environment where they had thought maybe that they were going to get a job at the local steel plant or auto-manufacturing plant, but now, instead, they end up being discouraged; they end up dropping out of high school. They go down a very different path than they otherwise would have.

For Black children, two things, I think, are different. One, the sectors in which their parents were employed to begin with were different. I think Black folks had less of a connection to union jobs in the steel or auto industry, for example, or more in the service sector, et cetera.

Second—and here I’m drawing on some interesting work by Lisa Kahn, another economist—when a given shock hits both a Black and a white worker, like the local plant closes, while I said earlier that white folks were likely to exit the labor force entirely, you see that Black people are more likely to switch to a different occupation, switch to a different firm. That’s why you don’t see those same very sharp reductions in employment in the face of the same exact shocks. That may be because they were less attached to those jobs to begin with, their identity was less tied up in those jobs, and so on.

Bethany: Were you surprised by how dramatic and how fast the transmission rate is? I guess if I would have thought about this beforehand, my intuitive belief would have been that, yes, this would happen, but it wouldn’t happen immediately, and it wouldn’t happen to as great a degree as your research shows. Were you surprised by that as well? And, overall, does your research make you more optimistic or more pessimistic?

Raj Chetty: Yes, absolutely, I was surprised by that. Indeed, we’ve been working on economic mobility for about the past 10 years. About 10 years ago, we put out a study called “Land of Opportunity” that released for the first time public statistics on how economic mobility varied across counties and metro areas in America.

A number of researchers used that data following that study to ask, what are the determinants of economic mobility? What a number of studies have shown is that you can identify long-standing historical and institutional factors that seem to predict present-day differences in upward mobility. Rates of slavery back in 1860 or the prevalence of Jim Crow laws or the patterns of the Great Migration—these things that happened 70, 100, 150 years ago—seem to be highly related to where you have higher versus lower levels of economic mobility today.

In light of that kind of evidence, I was increasingly concerned that I devoted much of my life to studying this issue, talked to lots of policymakers who are interested in changing opportunity, but when you look at this kind of data, you wonder, are these things actually going to change in a policy-relevant timeframe and a timeframe where I might actually see some change in my own life?

With this new data, basically because of the gift of time, we’re now able to look at more generations of children with these tax records covering all Americans. What you see is that things have indeed changed quite a bit as a result of these area-level changes in employment rates. I think, in light of the tremendous historical persistence of some of these factors, it’s amazing to see in 15 years that you can have a one-third closing of the Black-white gap and so on.

On net, that makes me more optimistic that change is possible. Yet it also creates a challenge because I can’t say I have the answer for exactly how you create more of that change. That motivates me, and I think many others, to try to study these questions more carefully.

Luigi: Is the glass half full or half empty? It’s true that this gap shrunk, but half of that shrinkage is due to the fact that the whites are doing terribly. I might attribute that to the fentanyl crisis. One of the reasons why Black and white communities might respond very differently to being fired is, as you know, doctors are more lenient in prescribing opioids, or were more lenient in prescribing opioids, to white people than to Black people because there was this stereotype that Black people abused them. Am I right in saying that half of this shrinkage is due to that, which is not a matter of being very optimistic?

Raj Chetty: A couple of things there. First, on the point of the opioid epidemic, my view there is that’s more a symptom rather than a cause. In our work, we also look at the subset of places where opioid prevalence was very low, and we find very similar changes in terms of outcomes for white folks and Black folks. So, we don’t think that that is, in and of itself, the key driver.

In fact, if you look at the data, while Black folks were protected, in a sense, from the opioid epidemic initially, in more recent years, you’ve actually started to see an uptick among Black people relative to white people. So, I think these things could be broader than that.

Stepping back a bit to the more general point you were making, you’re absolutely right that the one-third statistic that I quoted about the shrinking of the Black-white gap, about half of it is coming from Black kids doing better. About half of it is coming from white kids doing worse. There’s a little bit of nuance there, where there are certain dimensions where we genuinely are seeing, I think, meaningful progress among Black Americans.

In particular, Luigi, if you focus not just on average incomes among Black Americans, but what fraction escaped poverty or are not incarcerated, what fraction are working—so, basically, avoiding the most dire outcomes—you see quite a bit of progress among Black Americans. Where you’re seeing less progress is breaking into the upper middle class or the top echelons of society. Much more remains to be done there.

But you’re right that it’s a half-and-half story. Now, maybe I’m too much of an optimist, but why do I have an optimistic reading of that? It just comes back to the fact that these things can change. Even if we see change in a negative direction, it still shows me that this is not something that is set in stone over a 150-year period.

Apparently, coming back to Bethany’s point, this stuff can change either positively or negatively on a decadal horizon. And so, I think it raises the stakes in terms of figuring out how you make that change go in a positive direction going forward.

I want to emphasize that one thing we found throughout in this research is that there are rarely low-cost solutions to fixing these problems. I think, on the margin, there can be nudges that have some beneficial effects, but at the end of the day, you seem to get what you pay for in this context. I think changing kids’ trajectories over long periods has incredibly high payoffs but is also quite expensive to do so.

Luigi: I credit Oren Cass for raising this question when he was on the podcast, and he made me think. Imagine we’re doing some cost-benefit analysis of pollution, and we are seeing that this caused a certain number in expectation of deaths, and so we impose some cost that might end up closing some plants. Should we, in this cost-benefit analysis, import your aspect of saying, look, if I close this GM plant that, yes, it pollutes, and in expectation kills a few people, but on the other hand, it gives work to a bunch of people? And if we close this plant, and there is not any ready alternative, we have a generation of kids that would be disadvantaged for years to come. So, should our cost-benefit analysis incorporate your research or not?

Raj Chetty: From an economist perspective, yes. That may seem cold and calculating, but what I would point to there is a lot of the effects we’re finding are not just in terms of how much kids earn, but also in terms of health outcomes.

If you look at life expectancy, mortality rates, various other measures of success that go beyond money, you find that if you’re growing up in a better environment, you do much better on all of those dimensions, as well, in the long run. So, it could well actually be in your example, Luigi, that closing that plant, ironically, while it might have beneficial effects in terms of health and life expectancy in the context of pollution, could end up, actually, having net negative effects by increasing mortality rates for subsequent generations of kids.

I’m not necessarily saying that one should take a protectionist approach, that that’s the best policy. Ideally, you would figure out a way to do the efficient thing while supporting the next generation as well. But I do think it makes sense to consider these downstream impacts, absolutely.

Luigi: One way to summarize your paper—and again, feel free to disagree—is that class is becoming more important than race. We’re now facing a huge excitement because we have the first Black woman to run for president, but at the end of the day, she’s a daughter of a Stanford economics professor and an Indian Brahmin who was a researcher herself at the university. The real shocking thing is to have a JD Vance running for vice president because he was born at the bottom of the distribution. Is that correct?

Raj Chetty: I think it is correct that, directionally, class is becoming more important in America, and fortunately, race is becoming a bit less important.

However, in terms of levels, race is still incredibly important in determining your fortunes in America. Even if you take, in the present day from our most recent study, the most recent generation of kids, there is still an enormous divide in terms of where Black kids and white kids end up—even conditional, critically, on being of the same class.

And so, I wouldn’t want our most recent study to be interpreted as saying it’s now only class that matters, and race is much less important. Directionally, that is true, but race still remains quite significant. And so, I think it is valuable to celebrate progress on that dimension. I think it also will inspire subsequent generations, as it should. But class is becoming more and more important, and I think, if you project forward, that’s going to be just a key dividing line in the United States.

Luigi: Thank you very much. You are fantastic and a very good sport in taking my tough questions. I’m not that right wing, but I like to provoke a bit to make the program more interesting.

Raj Chetty: Great questions. Thank you all for reading the work so carefully as well.

Bethany: Thank you.

Raj Chetty: This is great.

Luigi: First of all, he’s a very good researcher, and he’s also very careful in presenting the right side of the results and not the wrong side of the results. I think there is a pretty clear underlying agenda, in my view.

Bethany: And what would you say that agenda is? I would say that it’s for more interventions. And he made the point that these interventions are costly, but I think his argument is very clearly in favor of more interventions at the community level.

Luigi: No, intervention is fine, but I think that as I was pushing him, I don’t think he spends so much time in looking at more cultural or social interventions. He looks more at economic intervention, which maybe is a bias because you are an economist, but if you attack a problem as big as the one of social mobility, you cannot just be an economist. You have to look broader.

Some of the hypotheses that he tests are indeed sociological hypotheses. I don’t know why not to consider, for example, should we subsidize more religious gatherings? If a church is a place where you actually mix people of different social strata, it could be a real channel to opportunities. For example, the type of teaching that is done in school.

I thought he was a little bit too fast in dismissing the Asian exception because I think it’s very interesting, and if this disappears over time, maybe there is something wrong in American culture that makes us more complacent, and immigrants have more of a drive that allow them to succeed.

If there is a formula of how to teach your children to be more resilient, I would like to know it and teach it in the schools. When he says, “Oh, we should improve poor schools,” that’s, of course, right. I always thought that one of the most horrendous things in the United States is the way local schools are financed because it is really a recipe to maximize differences in starting points and not equalize starting points.

I don’t want to sound like I criticize his main recommendation because I’m 100 percent in agreement with that. However, we know that first of all, money tends to be scarce, and there is a lot of resistance on that front. My question is, are there some alternatives that might be cheaper, even if we don’t like the sound of it? Maybe he’s promoting more religion or promoting more sense of responsibility, trying to create more connection across different strata, like, for example, in some mandatory activities where everybody’s forced to be together.

Bethany: I couldn’t help starting to laugh at that because the idea of a mandatory activity in which people are forced to be together sends shockwaves of horror through me.

But anyway, I am a little skeptical—and again, it’s just the danger of the anecdote of one, as I said in our conversation—of the idea, and it’s implicit in his work, not explicit, that parents don’t seem to make as much of a difference, even whether the parent themselves works or not.

It just seems to me that parents make a lot of difference. I get that that also is not an entirely popular thing to say because it turns it back from a community issue to a household issue. But it flies in the face of intuition—again, which is dangerous to me—to downplay parents as much as his work seems to.

Luigi: But Bethany, your kids, your daughters, are reaching their teenage years just now, right?

Bethany: Right.

Luigi: You will see how little influence you have moving forward. I think, up to now, you maximize your impact, but from the teenage years onward, it is their classmates. No matter what you say, in fact, at some point—this is personal experience, but I think a lot of older parents might agree with me—what you say counts with a negative sign, not with a positive sign. So, I’m sorry to break it to you, but I think that the social community will play a much bigger role moving forward. We’ll have this discussion in 10 years, and then you can tell me.

Bethany: Oh, dear, speaking of optimism versus pessimism, that’s certainly not very optimistic. But, fine, I will take it. And you’re probably right, my children are just at the age where I’m still under the illusion that I have influence over them.

Luigi: Were you satisfied with the answer to my provocative question on the current vice president and the candidate vice president?

Bethany: I expected him to shy away from your question more than he did, and he did not. Were you satisfied with the answer?

Luigi: Actually, no, because then I went back to the data, and I asked the following question: do you have more chance to succeed if you are a rich Black or a poor white? Do you know what the answer is?

Bethany: I do not know what the answer is. I would suspect, based on his work, that you would have more opportunity to succeed if you were a rich Black, given how, as we used the word “ossified” earlier, given how ossified wealth has become in our society. But perhaps I’m wrong.

Luigi: No, no, you’re absolutely right. I think that going by quintile, if you are a Black person born in the top quintile of the income distribution, you have a 17.2 percent chance of having kids in the top quintile. If you are a white person coming from the bottom 20 percent of the income distribution, you have only an 11 percent chance. That’s a pretty big difference. Now, if you are Asian, and you are at the top of the income distribution, you have a 38 percent chance of having kids in that range.

So, when you compare JD Vance with Kamala Harris, you don’t know whether you want to call Kamala Black or Asian—maybe 50/50 because that’s what she is—but even if she were classified as pure Black, she would be way ahead in her chance of success compared with JD Vance.

If you’re saying, as a Black person, do you have more challenges as a poor person? Probably yes, but as a rich, relatively upper-class person, then you have many more chances.

He was very good at deflecting the question into a very politically correct answer. There is no doubt that it is much harder to be Black in America than to be white, and his evidence is overwhelming in that direction. I think it is more difficult to be a rich Black than probably a fourth quintile of distribution white. But at the two extremes of the distribution, I think the things flip.

Bethany: But I didn’t think he said any differently. You’re right, he didn’t expound on this in perhaps the detail that you just did, so maybe your take that he deflected is true, but I didn’t think he deflected and misled. Maybe he just deflected. There’s a difference.

Luigi: Yeah, I think there is a difference. I agree with you, I think he certainly did not want to mislead. I think he just deflected in a very skillful way.

Can I bring you to a slightly different topic, which I think is interesting? I was a bit provocative with him on the issue of what is the optimal level of mobility. I think mobility is very important. With the increase in income inequality, the issue of mobility becomes much, much more important. If you see everybody lifted and becoming rich, the fact that the relative ranking doesn’t change very much is not as important because in the absolute ranking, everybody is super rich.

I think that the interest in mobility is generated by the lack of generalized growth. Giving everybody an opportunity is extremely important, don’t take me the wrong way, but also, making everybody better off is at least equally important. At some level, I don’t want one to substitute for the other. I don’t think that we should focus only on mobility and ignore the importance of uplifting everybody.

Bethany: I actually found it interesting from the standpoint of social stability. What degree of mobility makes the most stable society? That, to me, is a really fascinating question.

I was thinking, as you were talking, about our episode on Chile because that does show that even if you are lifting up the entire society, if the gap between the top and the bottom is growing wildly, you’re going to have instability.

So, I agree with you, you don’t want to see mobility and then a flattening top line. People can move around within a fixed box. That’s not very satisfying. But, at the same time, if the top line is growing a lot, and there’s more for everybody, but the gulf between the top and the bottom is widening, Chile would seem to show that that also is not a great answer in terms of the stability of the society.

Luigi: I think you’re right, but I think that the biggest cause of instability is the dissatisfaction of the people who are close to the elite, but they’re not the elite. Very rarely do revolts start from the bottom 20 percent of the distribution. They might be contributed to by those, but they are generally provoked and fueled and directed by unsatisfied intellectuals. In that sense, I think Chile is a phenomenal example because there was an enormous selling of this idea that, oh, you go to university, and you will become rich like the elite that is rich in Chile.

That’s simply not true, and for a number of reasons. One is that for the Chilean elite, the wealth is based on accumulated wealth, and so, you’re not going to achieve those levels. But most importantly, it is because the elite is quite closed. A former colleague of mine showed that even if you are super smart in Chile and you enter the top school, the Harvard of Chile, your life is not made better off unless you come from the right high school—unless you’re part of the elite. And so, you had a bunch of people who actually took on a lot of debt to go to university, hoping that this would change their lives, and they realized they were sold a bill of goods.

Bethany: That’s really interesting. And so, while Chetty’s work is undeniably incredibly important and groundbreaking in terms of what we can do for the least well off in our society, maybe if what you were trying to study was what makes for a stable society, you would actually look at mobility between the top 1 percent and the distribution from 90 percent to 99 percent. Or maybe you’d even look at mobility between the 0.1 percent and the distribution of 90 percent to 99.9 percent. And maybe that’s what you should be looking at and studying and analyzing and concerned about if your overall goal is social stability.

Luigi: Yeah, I think it is how well college graduates are doing in all the disciplines, basically. And if you have a big cohort of college graduates who do poorly, then it’s a major reservoir for instability and revolts.

Bethany: Fascinating. Fascinating. I’d love to know . . . There probably isn’t because you couldn’t really get the data on this over the history of the modern world, but it would be fascinating to know how to think about mobility in terms of social stability and what strata of society you need to have mobility in order to keep a society stable. Wouldn’t that be an interesting thing to know?

Luigi: We should ask our listeners to send us papers. I’m sure there are papers, but I cannot think of any off the top of my head. We should ask our listeners to suggest papers and books on that, and maybe we should have one on the podcast.

Bethany: Yes, please weigh in, everybody.