 When we talk about mobility, we tend to, in some sense, think of it as the correlation of parents and kids. Well, that was a terrible lack of correlation. In other words, there was something about the nature of discrimination in America at that point in time. And I think the data around the late 50s was taken from. Parents could not lock their kids in. And so I think that what we need is, and of course, this is not an original thought to say that the world has non-linearities, I think we need ways to distinguish the characterization of inequalities and persistence intergenerationally between individuals and the morally salient and historically salient social groupings we face. I'm Stephen Durloff. I'm a professor at the Harris School of Public Policy at the University of Chicago. And I'm also the director of the Stone Center for Research on Wealth, Inequality, and Mobility. The name of the Great Catsby Curve in some ways reflects the fact that the idea was politically salient and what I mean by that is that a number of economists, and I think Miles Korak deserves particular credit, identified that if you look across countries, those countries that had relatively tight distributions of income cross-sectionally tended to be more intergenerationally mobile than others. And so in other words, it looked like that those countries that were both unequal at one point in time were more unequal in terms of the extent to which parents could parental income control the destiny of children. When this came out, the then chairman of the Council of Economic Advisers, Alan Kruger, dubbed it the Great Catsby Curve. And I think the reason that it had political salience is that it was one, it really did challenge certain American myth. And so a thing that was said when I was an undergraduate, for example, was that America had made a different choice from Europe in terms of the social welfare programs. The idea would be that Americans care about equal opportunity. And so in a society where you want to maximize opportunities, you, by corollary, allow for lots of inequality in the outcomes. Well, in some sense, what the curve was saying was there isn't a trade-off. In other words, societies could both have very high levels of mobility and not particularly disparate levels of comparatively lower levels of cross-sectionally inequality of outcomes. And so in that sense, I think it resonated not just because it challenged a myth, but because of all conventional measures. Inequality has been going up the last 40 years. Now, the reasons for it, the levels of the changes, all of that is under debate. But the fact of increasing inequality, I don't think is, per se, disputed. And so that intersection did it. Now, then you could ask a second question, which is, what does this mean normatively? And I think the idea of the Great Gatsby curve has salience, because it tells us that there's a consequence to cross-sectional inequality that is particularly morally disturbing. And what I mean by that is that in theories of distributive justice, what are fair distributions, what makes inequality fair or unfair, one of the primary ideas is that inequality is unfair if it is for reasons that a person should not be held responsible for. And that idea should be associated with the economist and philosopher, John Romer. And so upon, you know, reflects upon that, society in which the inequalities we observe cross-sectionally are to a large extent driven by differences in parental characteristics, it's extremely hard to defend that. Because as they say, the lottery is one with reference to parents and kids, you don't get to choose them. And so that becomes a fundamental argument that there's ethical argument in terms of the justifications for policies to reduce inequality. Is what sort of mechanisms generate a relationship between persistence across generations and cross-sectional inequality? Now, one has to be a little bit careful about that in the sense that there may just be mathematical relationships between them. This may sound boring, but the underlying vision of the data in an economy is there's this process that's evolving and it has many features. Some features are cross-sectional, some are dynamic or intergenerational. And so the first question you ask is, this is the nerdy stuff you have to ask is to what extent is it a mechanism versus a function of the fact that there's a process that's evolving. So let me put that there. And the reason I say it that way is that one could ask the question, if I see a positive correlation between persistence and inequality, you know the banality, correlation's not causation, how do we think about the relationships between them? Most of the theories have embedded in them the implication that inequality begets persistence. And it turns out that the canonical mathematical models which can generate it mechanically, its persistence begets inequality so that you could edit all that out as being boring, but that's kind of the background to it. I think the theories of the Gatsby curve come in in three broad categories. One of them are ones that are based on family influences. And so one idea would be that family incomes matter directly for children. There's investments by the parents. A more complex version of such a model would say that family characteristics are both the family's income and the educational levels of the parents. And so what that would do is it would say that there was a direct income effect from parent to child and an indirect effect because education of course affects the parents' income. From that vantage point, what happens intuitively is that as income's spread out, you have greater disparities in the investments in children and from their parents and the influences of parents interact with that because that'd be an implicit idea that highly educated parents are probably pretty good at spending money to educate their kids and get them to succeed. And so more inequality amongst the parents' education and income levels generates greater persistence when you look at the offspring of the rich and the poor families. So that would be one very broad way to think about it. A second category I'm gonna call that social and this is the area that I personally worked on as an economic theorist. And there the idea is to say that many of the mechanisms that matter for the success of children are socially determined. And so to make that concrete, one thinks about neighborhoods and schools. Now parental income matters for neighborhoods and schools but it's a different way to think about it. What parental income does that determines what sort of neighborhoods and what sort of schools a child grows up in. From that vantage point, as parental income becomes more unequal, the quality of schools, the characteristics of neighborhoods become more dispersed and that is what allows the inequalities amongst the parents to be perpetuated across time. Now in that vantage point, there's really kind of, there's several reasons why this would matter. One obvious one is the way the United States finances education. There's many cases, very large differences in per capita expenditures depending on the influence of the neighborhoods. So the political economy does that. The other is simply that segregation more broadly generates intergenerational disparities. There's, if you look at things such as exposure to lead, exposure to violence, the list goes on. Fluid parents can isolate their children from these phenomena in ways that poor families, lots of fluent families cannot and the evidence of the effects of those two, for example, is incontrovertible on future success. And so I would say that the difference between the social and the family ones by definition is that the family ones say, you sort of change resources in a certain way for the family or you augment the family resources through early childhood investment might be an example. Whereas the social ones really focus on segregation. In other words, even reducing socioeconomic segregation or reducing its consequences. So if I already equalize expenditures across neighborhoods, I'm sounding like a real social planner, that would make the consequence of living in neighborhoods with different incomes less. The other, of course, is to make affirmative actions to increase integration of communities. Now I put those there as a good economist. I of course mentioned the two, the body of economics work. There's work by economists and political scientists that we were in political economy, which would ask about how inequality begets persistence. And so the sort of idea there would be that as you have more unequal distribution of income, the policies that emerge from that system become less generous towards the disadvantage. And so you have a cycle in which inequality leads to a less generous, so either social welfare state or efforts to improve the quality of poor schools, et cetera. And that would generate the dynamic. Most of what I've had to say is thinking about a gatsby curve within a country. You take the United States and you ask intertemporally what inequality is greater in this generation where you have more persistence of status. A different question is comparing the United States to Scandinavia, for example, which is what the original gatsby curve did. I think that what I would emphasize in thinking about the socioeconomic mechanisms within the gatsby curve is this is a canonical example of the importance of multidisciplinary perspectives. If you think about the explanations that I gave you, I started with a very neoclassical economics idea. Parents invest in their children. The budget constraints determined by the parents' investment and so that's what generated persistence. Then I switched a little bit. I said, well, that's 21st century. You say it's not just the parents' income, it's characteristics of the parents. That would be the skills idea. I then said something about social. That's really a sociological idea. In other words, if I say to you that there are phenomena in terms of the relationship between the milieu in which people grow up and what happens to them, that's where economics and sociology naturally meld. When I was making arguments about the relationship between the distribution of income and associated political outcomes, and of course I could have decomposed that to say there's differences in political power if you have certain disparities in income so on and so forth to explain why there is, that's where political science integrates with economics. So I think the most important thing I'd put on the table is that if I take what were the canonical explanations, those represent a real need for the integration of political science, economics, and sociology. Now, in saying that, I didn't mean to give short shrift of psychology, because, and in fact, in my old synthesis of the literature, I've emphasized, and with my co-authors, Kimmington and Andrews Cortellos, that there's also ideas of identity that have to do with the psychology of individuals, societies that are segregated. There's another set of consequences, how do people think about themselves? And so the example I would give is that, we also, we're talking about college is the gateway to success, is an empirical matter, I'm happy to agree with that. If one grows up in a community in which nobody has gone to college or parents haven't, it's not just a matter of saying they don't have role models, it's saying that you're requiring an act of imagination that isn't necessary for somebody, and let's say, you know, the joke I always tell my students is, did my children have a chance to go to college? The answer is, well, I guess it was a choice, but their parent was a college professor, all their parents' friends were college professors, all of their friends were the kids. In other words, that's not just a matter of, as I said, as a role model, it's a matter of something else which is the acts of imagination, the capacity to see oneself in a different world than the one one starts in. There's been some very interesting work, for example, on so-called low-cost interventions in improving college education, and one of them is in cases of students of color, and students in which, in freshman orientation, you take steps to positively facilitate how they're going to process the college experience, and so what do I mean by that? There is an irreducible fact at every college, and that is people went to high schools of different quality. If you went to a lousy high school, people went, it's likely freshman year's gonna be pretty tough. If you're 17 or 18 years old, and you show up in college, and you get a poor grade the first time, how do you process that information? You say, I'm gonna need to bear down, I'm gonna have to be patient with the fact that people came from different backgrounds, or do you engage in self-questioning? To me, those are issues of identity, those are the issues of the psychology of feeling that this college, this institution, later on in the labor force, this organization, that's a place where I fit. So I see very important areas of psychology which speak to the Gatsby curve, but I wanna be honest, there's only a little bit work's been done on that, and it's not as far as I know by economists. I think that in terms of general research and intergenerational mobility, there's a great need to think in a richer fashion about the process that transmits things about parents to children. I mean, the mobility is definitionally about parents and kids. So here's one, I'll tell you things I'm working on, and so in that sense, it's self-serving, but I hope it's important, and so one question would be, should we think about parents as having this single measure of income when a child's growing up? Economists call it permanent income, it's operational, you take the average of the incomes from, let's say, eight, zero to 19. A different thing to say is that incomes at different ages matter. So maybe the influence of income when you're born is different than the income your parent has when you're 16 years old. And so a number of people have studied that, and so the particular work that I'm doing, and I wanna mention my co-authors at Yusung Chang and June Park at University of Indiana Bloomington and Sung Hee Lee, who is not the Korean Development Institute. We actually asked that question, if you sort of look at the trajectories of incomes between zero and 19, how do they predict future permanent income of the kids? And we found something pretty surprising, and that is the incomes in late childhood and adolescence have much more predictive power, they're more sensitive than the early ones when children are young. So in other words, if I wanted to sort of save you mechanically, there's a dollar higher here that have a bigger bang on my prediction than down here, it was almost monotonic. Now that's surprising because there's all this very compelling evidence, I believe, about the efficacy of early childhood investment. But then this is the point I wanna get to, and that is, I think where the research has to go is to not have this black box for parental incomes somehow produce some mysterious thing called education that then the kid or the child takes to the labor market. If one thinks about the results, and I think, okay, this is speculation, but I think this is what's going on, that whatever criticism is on the hands of the American Welfare State, they're worse for adolescence than it is for children. And furthermore, what income does is different. Adolescent incomes later, those have to do with schools and neighborhoods. There's a different thing that's being purchased. And so it's not a matter that, one wants to say early childhood or something else, it's rather, there's complicated complementarities, interactions between them. And in my judgment, we have to step back. An important thing to do is to step back from this scale or this blob of income produces a blob, according to this very scientific terms I know of education, and that produces income. But rather to say the parents have this vector of things that they do for a child. And I think that that therefore will speak to that. It turns out that when you sort of take this perspective, you can generate a Gatsby curve based on the distribution of ages in a population. And the reason for that is, as children get older, if I look at 17-year-olds, the variance of their parents' incomes is bigger than what we have one-year-olds. And so actually it generates a different type of Gatsby curve. That is, the cohort, the distribution of cohorts actually could create a relationship between inequality and persistence. So that's kind of the first thing I would say, I think is an important measurement issue. The second is that I think that this literature, and to be clear, many people are working on it, needs even greater focus on issues of group inequality. This is an anecdote which you're not gonna believe, but it's true. I was typing in that debate team in high school and I was looking some book up. And you're looking for stuff on the guaranteed annual income was the topic. And I saw a paper that had something which I never forgot in it. It was an occupational mobility table, not income, but occupation. And it was for African-Americans and whites. And so it was around 1962, so it was during, ending of Jim Crow, in which you could sort of say you had upper white color, lower white color, blue color, I don't remember if it was upper or lower, and then agriculture and manual. And if you look at the white category, it's kind of what you would expect. Because most are likely that you have a white color show, that the parents white color. For African-Americans, it didn't matter what the parents were. They always were likely to have kids that were manual workers. So why do I bring this up? Because when we talk about mobility, we tend to, in some sense, think of it as the correlation of parents and kids. Well, that was a terrible lack of correlation. In other words, there was something about the nature of discrimination in America at that point in time. I think the data around the late 50s was taken from, that parents could not lock their kids in. And so I think that what we need is, and of course, this is not an original thought to say the world has nonlinearities. I think we need ways to distinguish the characterization of inequalities and persistence intergenerationally between individuals and the morally salient and historically salient social groupings we face. And in saying that, that's kind of, again, these come down to some of the devils in the details. It's kind of, you know, figuring out the right ways to measure these things is hard. But I think that that's just another direction we have to go. I mean, to be very provocative or not, the reparations arguments in my judgment, the strongest ones have to do with persistence. In other words, independent of one's views of contemporary discriminations. And I'm not taking a stance that they're not existent. I'm saying, even if you thought that, the persistence of what happened is so vast that the idea of affirmative, you know, societal level efforts to break those, that persistence to me is a very, very morally powerful argument. And I think evidence on nonlinearities on sort of, the term would be bottlenecks in which there's certain configurations of family, ethnicity, geography, they just make it really hard for kids to get above that. That type of focus should be explicit in the measurement of mobility. So I think that many of the inequalities we face in society are because of these disparities in social interactions that are faced. So from my perspective, you know, I think that, you know, you ask issues about, you know, what are the kind of standard things that people worry about? What are the rezoning restrictions and how can relaxation promote socioeconomic integration? Now, relaxing some of these restrictions by itself, that's not a, you know, that's not, it's not self-evident that you actually have the construction going on. That's kind of a thing you would want. That's kind of the first step. Obviously, government subsidies for middle-income housing, that's a thing that can be done. So at very micro levels, you can think about economic integration. If we're gonna focus on race, I think there are extremely compelling arguments that diversity both intrinsically matters and it matters intergenerationally. So for me, the preservation of, you know, the way in which we think about the construction of college classes and all that, that's all very important. And I think there'll be a, you know, be terrible move backwards if the Supreme Court bans those types of policies. And so I realize it's getting into politics and ethical issues, so I'm telling you what I think is exactly appropriate, those things have to be defended. I think another thing that I would probably emphasize in all of this is that I think is the need for thinking about how to simply reduce the sources of disparities that are race-based in terms of everything from labor market connections to what I recall to the act of imagination. And so those are just the issues of information flows. If you said, what's the policy consequence? I don't, that I'm not qualified to say, but I think that it's extraordinarily important that there be actions that, you know, when you go to the University of Chicago and you see in the spring, the kids get off buses and walk around the campus, all that actually is meaningful. And so what I don't know, and this is where I can tell you who would, that's those types of things I would scale to dramatically break the psychological isolations that I think exist.