 In this final lecture I want to answer Lenin's question, what is to be done? And what I mean of course is that it makes some comments about public policy. I first want to start off by addressing a pernicious myth. And that is that efforts to reduce inequality, in particular efforts to reduce poverty, have been a failure in the United States. And so the myth in some sense is to say the war on poverty was lost. I say it's pernicious for the simple reason it's untrue. One might answer, obviously poverty still exists. The question is whether war on poverty ameliorated levels of disadvantage in the United States. And so Christopher Jenks did a very interesting calculation. And basically this is the thought experiment. You start off with a level of poverty in 1964. You have a level in 2013. You might say, well, that's still a very substantial level. And nobody would disagree with that. And what he demonstrated was the following. That if you consider the effects of policy on poverty, in other words, non-cash benefits, Medicaid and the like, you think about the refundable tax credits that exist for the disadvantaged, and you correctly measure the inflation rate, you find that the adjusted percentage of poverty was 4.8%. And so from that vantage point, to move from 19% to less than 5%, that's an extraordinary success of the war on poverty. Now, I don't want to give the war on poverty credit for the way that we measure inflation, but the upshot of that is we have to think differently about the effects of policy. One question is sickly, did it make people better off? In other words, they're well-being. Did it address disadvantage by providing resources? A second question is, how does it affect intergenerational, intertemporal levels of poverty? And what I want to argue is simply that we have a number of policies that matter for understanding mobility. The first is the Earned Income Tax Credit, which is one of the major anti-poverty, anti-disadvantaged policies in the United States. So this very nice paper demonstrates that if you look at the effect on children of the EITC, there are significant influences in terms of their well-being. Now, is that become a mobility policy? Absolutely, because if one accepts the vision that I've tried to articulate of these trajectories of childhood experiences and adolescence experiences determining adult outcomes, improving the well-being of children has dynamic, has the intergenerational and the intergenerational consequences that one looks for in improving mobility. Interestingly, one also finds that the intergenerational mobility gap between blacks and whites is reduced by the EITC. And that's another statement that says by improving the socioeconomic status of disadvantaged families, one that's not only addressed the issues of income inequity, inequalities between rich and poor, it also addresses the, really, the fundamentally salient type of inequality in the United States, which is racial. The second thing I want to say is that our decisions on how to fund public education have important consequences for mobility. In that picture I gave you comparing New York, California, and Texas, it was just an example of the substantial inequalities that exist in per pupil expenditure in the United States. I should say that there's been a long and contentious literature in social science asking about the effects of school spending on student achievement. I believe that that controversial literature has now come to a consensus that school expenditure matters. Data became available, types of interventions occurred, which allowed one to tease out so-called causal influences. Courts ordered reforms, for example, and you actually could see what happened pre and post these interventions. And so work by the economists Rucker Johnson and Kiribati Jackson created very important influences on years of education associated with school finance reforms. You saw improvements in education after the reforms. And so that's just a nice demonstration that spending more actually translates into the things you want. You could ask, how could I raise the high school graduation rate of low-income students? And the answer is a 10% increase in per pupil expenditure can have different influences. And so this, in some sense, is the membership theory of inequality with a vengeance. The extent to which economic segregation means different quality of schools that is translating into different rates of graduation earlier conversations also is translating to different college quality, equalize, spend more on the disadvantage. You have improvements in educational outcomes with life cycle, life course consequences. Another policy I want to say something about is early childhood investment. And this is one that has experienced an explosion of research and an explosion of information. As we think about children developing, there's a period of sensitivity to influences. That period of sensitivity has very important consequences if investments in the children are high versus low. And so having early childhood education can have very important influences. If you look at different interventions in early childhood, there's some heterogeneity in their effects. It's important to recognize that if somebody says early childhood investment is a virtue, you have to think about the intensity of the quality of the program. If one observes heterogeneity in these programs, you shouldn't say, well, the successful ones are outliers or there's something lucky about them. I think the better analogy is to say that in thinking about early childhood, we're looking for the correct vaccine. Different vaccines may or may not be efficacious. Different types of programs are constructed and the intense ones, the ones that involve the parents in them seem to have substantially powerful, substantial returns for their children. I think that the most careful studies by John St. Jackson have found persistent influences to Head Start. And just as important, they have demonstrated an interaction between Head Start and good quality school subsequent to that. There is a complementarity and interaction between them. The best available evidence is that good early childhood investments have persisting consequences for children and that is an important nexus for improving mobility. The third comment I want to make about public policy is going to sound a little bit different and it has to do with criminal justice. How does one reduce persistent inequalities? How does one avoid lock-in within an individual's life course of socioeconomic success? How does one avoid persistence across generations? We have to come to grips with the fact that the American carceral state has very powerful inequality effects. The first fact is for African Americans there are very substantial probabilities of incarceration. The second thing is that incarceration is a very, very important factor in understanding the life courses of less advantaged people. We think about the exposure of children to having parents that have been incarcerated. The development of lower rates of imprisonment and the recognition that incarceration is damaging to individuals and to their families is part of the story of how to reduce intergenerational disadvantage. In thinking about mobility one of the factors is matching. The neighborhood I grew up in, the school that I attend, the network of friends that I interact with, the organization in which I work. And I think that there's extraordinarily interesting evidence that interventions along aspects of these groups could have interesting inequality consequences. And so I want to talk about work that is due to a set of psychologists. It goes to high schools and it presents a program to teach what they call a lay theory of achievement. And the idea essentially is to talk to individuals about what it means to socially belong, to culturally fit, to think through the process by which one achieves in college. Even though this looks like a modest intervention, it had persistent consequences for grades and for enrollment. Teaching that people belong, teaching them that the adversity they're going to experience in college is not something that reflects on them in terms of something intrinsic about them but reflects perhaps on high school background, etc. That turns out to be very important. There's other interventions that have been done prior to the freshman orientation that again talk to students about what it means to be in college, what it means to face adversity in terms of classes being harder than one expected, having peers that have backgrounds that are different from you, etc. And again, what happens in these programs is that the teaching of mindset has measurable consequences for children. I put this on the table because it has an important message for thinking about the interplay of intergenerational and intergenerational mobility. I'll be honest, the first time I heard about these studies in which you had these specific interventions before freshman year of college, I was skeptical. Not because I didn't believe the others had written in good faith. I just couldn't understand how it could happen. And then I realized it's because frankly I had the wrong mathematical model in my head. The model I was working with was why you go through college, there's all these things that happen to you, some are positive, some are negative, they average out, so any individual thing can't be that important. That turned out to be the wrong way to think about it. Suppose one goes to college and a freshman year gets a C in the first exam. How does one process that? Do I say that I'm not smart? Or do I say I didn't have the background necessary, so I have to double down? So if I don't do well in a class, I take classes that are not as hard so I learn less. If I have a disciplinary problem and the response of the school is one to take me out, push me behind, engage in what's called retributive versus restorative justice, we're going to have very different consequences for children. And so what I want to put on the table, that might sound like a very modest microintervention, but these types of studies tell us they hint to something very broad, which is the development of individuals, and of course development doesn't stop at age 18. If you think about it through childhood, through adolescence, through college, where can we intervene to make sure that negative experiences are not persisting and be getting future negative experiences? And that of course interacts with the whole incarceration issue. Putting teenagers in jail has persistent consequences because of the fact that it attenuates their education, their connections to society, etc. What I would like to put on the table is a different vision of what policy ought to be doing. And that is shaping the trajectories of children, adolescents and young adults in ways that adverse experiences, adverse events don't become self-reinforcing. The final thing I want to put on the table is something that I have called associational redistribution. This is something that I personally focus on as a researcher. So if you accept my position that a fundamental determinant of mobility, both intergenerationally and intergenerationally is via the memberships theory, in other words the location that people have in the socioeconomic structure, that leads to the natural question as to how policies can influence the locations. Affirmative action is a statement that colleges, for example, have one way that they engage in matching and it asks the question, relative to that benchmark, should other evaluative criteria be used? There is very good evidence that affirmative action substantially benefits those who are admitted under diversity criteria and so I want to highlight recent research by Zachary Bleeber, who has looked at California and asked about the changes and outcomes of underrepresented minorities associated with bans of affirmative action and what he found was if you look at their wages, you look at performance in school, so on and so forth there were substantial harms that were done to those who were not assigned to the more elite UC schools as a result of the ban. Many policies if you think about them have to do with the matching process. If I ask the question what should be the evaluative criteria for a university, should it mechanically look at SATs and grades and admit people based on being greater than or equal to two numbers, that's a rule. Another rule is to say that you evaluate people more holistically, you recognize the backgrounds they come from affect those types of numbers. Another is to put an intrinsic value on diversity, all of those are alternative values and what I put on the table for you is that associational redistribution is an important dimension of mobility enhancing policies. I of course may be subject to a question and that is what happened to meritocracy. In other words there's a philosophy which goes back to something I talked about in lecture one, this notion that some things we deserve them, people deserve to have this and so the idea in other words would be to say that high test scores I deserve to go to this university. What I want to argue is that if we think about meritocracy correctly it is perfectly compatible with associational redistribution. It's perfectly compatible with what is conventionally called affirmative action. When people say that I deserve to go to a college because of my test scores in high school that is a backwards way to think. It says based on past performance I deserve a reward. That isn't how firms hire people, firms hire people based on whether they'll be productive in the new position. What is missing in affirmative action discussions is attention to the prospective consequences of the maxes and so if you ask the question what will be most productive in terms of who is admitted, in terms of what the value added of the school could be how that individual because of the social structure will influence others you get very different answers than the backwards looking one. There really is no logical conflict between affirmative action and quote-unquote meritocracy. Once we conceptualize merit as a prospective and social object in which we're asking what is best for society. So that's what I really wanted to put on the table. I want to say that some of the conventional policies are highly efficacious for mobility things like the earned income tax credit. I want to argue early childhood investment that is I think the evidence that high quality programs work is extremely powerful. In thinking about this process of maxing we should recognize that people are on trajectories and so that self-reinforcement of adverse events that's the source of persistent disadvantage. And so asking how can we mitigate those they may be in some cases low cost interventions making sure that incoming freshmen understand the difficulties of college understand what it means to be in a place where people can have very different backgrounds of them. I further want to argue about the carceral state. In other words that incarceration has adverse inequality consequences and finally I want to argue that associational redistribution is an important frontier for future egalitarian future pro-mobility policies. And so with that I'd like to thank you.