 Thank you, Carson, for that introduction. I appreciate that. I will introduce Rasmus, and indeed he is a co-author and co-presenter. So I was asked to participate in this interesting conference, and I found it challenging. If you look at the title, you can see why, because I'm supposed to comment on Denmark, and I'm not Danish, I don't read Danish, but I do have a Danish co-author. And I would say a high-quality co-author, and we've been working on this question for the last three or four years, and have even published a paper, and hopefully will produce other papers. So this talk today really reflects a partnership, and it's an ongoing partnership, really trying to understand a comparison between the U.S., where I come from, of course, and Denmark, and the Danish welfare state. And so as an American, I have heard lots of discussions about Denmark. In the 2016 presidential election, Bernie Sanders, who was running for the Democratic primary, for the candidacy to be the presidential candidate of the Democratic Party, spoke lovingly and frequently about Denmark and its ideal welfare state. And even Hillary Clinton, who was less vocal about the issue of Denmark, was also talking about central features of the Danish welfare state. And so in the U.S. political context recently, but of course, often even before, there's been much interest in what the welfare state can do and what Americans should take from lessons about Denmark, and its great progress in advancing the cause of equality and social mobility. So Rasmus and I approached this problem a couple of years ago, now three years ago I guess it was, really. Rasmus had visited the University of Chicago, and it was a great visit, and we've continued the work. But one thing that came out, and this is really the theme of the talk today, is that, of course, as Carson just said, well, that the whole question of inequality, and I would add social mobility on top of it, which is a separate but important issue, I think this is a central theme of the Danish welfare state. And we know, if you look at the data, and many people have commented on the data, that after looking at the tax and transfer system, income inequality is low, and income mobility, and by that I mean mobility across generations, how important the parent's income is as a determinant of the child's income, that that is very high, a lot of mobility, much less dependence on the accident of birth, at least in terms of income. However, and this was something that staggered me when we first found it, and that is that educational mobility is remarkably similar between the United States and Denmark. And to me, I say remarkable, because of course there are many features of the American welfare, the American welfare state, and we have one, that are really very much inferior to what's happening here in Denmark. And so to me the question is, this is very odd, we're having greater income inequality reduction, we're getting much more social mobility in terms of income, yet in terms of a basic skill, central to success in the modern economy, anywhere in the world, education, that there really isn't so much mobility. And so, begin to look more closely, and we're still continuing to look closely, at what are substantial skill and education gaps between children with different family backgrounds. And that's true here in Denmark, it's certainly true in the United States. And of course the skills that we're going to talk about, we'll briefly mention that list of skills, it's not just IQ, it's not just PISA scores, it's not one item, it's a whole bundle of items, but in fact these skills are very important. And the effects of some of this inequality and the lack of or success of and promoting skills lead to multi-generational impacts. So let me just show you what is now a central figure in world discussions of mobility and inequality. This is called the Gatsby Cur, it's after a book, Alan Krueger, now a Princeton but previously chief economist under the Obama administration coined the term the Gatsby Cur. And this is a curve that displays two relationships. One is a relationship at the bottom, if you look at the bottom of the graph, it gives you inequality after taxes and transfers. This is inequality and family income. And if you look for example at the Scandinavian states together, Sweden, Denmark, Norway, Finland, and you'll see that the Gini coefficient, which is a measure of inequality, family inequality, very low in Denmark and relatively high about the same level as that of Singapore and China in the United States. So there's real difference if you just look at Gini coefficient differences. And if also you look at income mobility, and here I'm showing you an equation that relates the income of the child averaged over the lifetime or a substantial stretch of the lifetime related to the income of the parent. And if the relationship is one of high mobility, this coefficient beta should be close to zero. And so the higher beta, the more the origins of the family affect the origin, the final destination of the child. And you can see then in the United States, the relationship is very high. In Latin America, it's even higher. You can see Peru, Brazil, some of these countries are very unequal and they're also very low social mobility. But of course in Denmark, Norway, Finland, Sweden, you find very high measure of social mobility, a low value of beta. This relationship, however, is also very suggested. It's an empirical relationship. Those countries that have high inequality also tend to have high immobility. And that is what this Gatsby curve shows. Gatsby comes from this book, The Great Gatsby by F. Scott Fitzgerald, documenting the high life of the affluent in the 1920s in the U.S. So it was Kruger's whimsical title. But it captures the notion that there really seems to be a relationship empirically. There is between inequality at a point in time, in a society, and what happens in the next generation. So this is an important relationship that's very much documented around the world. Many different variants of this, Miles Korak has probably done the most work, but so forth. Now one thing that's extremely interesting about the Danish welfare state compared to the U.S. is, of course, that if we look at the expenditures on educational institutions, and it was this that attracted so much attention in the 2016 election, and has still attracted attention around the world, state of Vermont, many other countries, states within United States pointing to this, is that expenditures not just in public schools starting from kindergarten on, but in the pre-primary schools are much more favorable towards children, young children in Denmark than in the United States. And if you look at the whole bundle of educational expenditure, Denmark is much more generous. And so in this sense, the question is, look, Denmark is far more generous, and it shows up, as you'll see, in the sense of reduced inequality in test scores taken, measured by children in school. So using a lot of the current measures that we see and use around the world, we see that Denmark is far more generous. And this is part of the welfare state, part of the expenditure that Carson was talking about. Yet, and this is what staggered us, and I think this is what's important today, is that there is steep gradients of children's education in parental education, income and wealth in both the U.S. and in Denmark. And so if you look at what's called the IGE, this is the analog, not any more of looking at the relationship between the income in one generation than the income in the next. But here, a relationship between the education of the parent and the education of the child, you can see that Denmark, which has a much lower genie coefficient in terms of after income tax and transfers, Denmark's IGE, the relationship between parent schooling, child schooling, actually, if anything, is more strongly related, but of the same order of magnitude as in the United States. And that, to me, is very surprising. And here, I'll let Rasmus show some data that he discovered recently, which I think is extremely important, which will form the core of our discussion. So we're going to move back and forth. I hope you don't mind this orchestrated presentation, or maybe not so well orchestrated but you get Rasmus. Well, thank you, Jim. Yes, as Jim said, there's a strong dependency between children's education and parents' education, even in Denmark. But actually, this dependency starts at birth. So what this figure shows is birth characteristics, so birth weight of a child, the fraction of children who can basically go home from the hospital after birth, who are not admitted to a neonatal ward, by their mother's education. So blue is if mother only has compulsory schooling. So red is if mother have high school, or academic or vocational, or green is if mother have a college degree or higher. And what we can see is that they're even at birth, by birth weight, are significant differences between different educational levels of mothers in Denmark. And if we look at the fraction that are admitted to a neonatal ward, I mean, these are not trivial differences, 40% more are admitted to a neonatal ward if mothers have a compulsory schooling only, relative to if mothers have a college degree. So these are huge differences. And these continue. So if we go to, say, daycare, so Bernhauer-Ellen, we see basically the same differences. Again, sorted by mothers' education. So blue is mothers have compulsory schooling, red mothers have academic or vocational high school, and green is mothers have college or higher. So now we see rated self-regulation scores from age three to five. And again, we see the same differences, and age three. So these are differences that have emerged earlier, actually, than children start daycare. And they basically persist all the way up until children start school. And we can look at what happens in school. So now we see test scores in Danish, in school. So in second grade, if your mother had college, on average, you're doing almost 50% better than if your mother only had compulsory schooling. And this is constant, from grade two to grade four to grade six to grade eight. So of course, we can see that the pattern we saw basically at birth is the same pattern we can trace all the way up until age 14 to 15. So very stable. Then some might think that, well, life is more than test scores. So if this was just test scores, then it's this important. Well, we could continue in life. So now we can look basically at who stays clear of crime. So a fraction of children that do not commit crime, again, by mothers' education. It's basically the same pattern. So twice as many commit crime, if your mother, on average, if your mother has compulsory schooling relative to if mothers have a college degree. Or we can look at education. So final years of education, that goes back to the education idea. So there's a two and a half year difference on average between the green and the blue dot, so by mothers' education level. Or even income at age 40. So almost a 50% difference in wage income at age 40 by your background. So a huge dependency basically from birth all the way up until adulthood. But it continues. So we can look at who has a health such that they haven't contacted a hospital between their age 40 and 50. It's basically the same differences again. Who are in the labor force at age 54, the same differences again. It's even who's alive at age 60. So perhaps a bit dramatic, literally from the cradle to the grave. And not only that, it's basically all the core areas in the Danish welfare state, so we have prenatal care. We have daycare, childcare. We have public schools. We have crime prevention, the judicial system. We have higher education, income taxation, health care, retirement, old age. But this is just within a lifespan, across a life cycle. But it's also stable, these differences across time, across the entire expansion of the welfare state. Many of these differences actually remains quite stable. Even though there's tremendous progress, both in terms of access to services, but also in terms of the technology, as Carsten told us. So one example is child mortality. So child mortality by age five. So he is separated by mother's education. So the solid line are mothers who have compulsory schooling only, and the dashed lines are mothers who have high school or higher. So what we can see is that child mortality has been decreasing, and it has so for probably hundreds of years now. But from the 70s until today, it's been cut by, say, 60% to 70%. So a very, very strong increase, basically, in one of the basic fundamentals, child mortality. And it's done so for all groups. But if we look at the gaps between the groups, we see that they're quite constant. You can see the gray line in the button. That's actually the gaps in absolute terms. And if we consider the gaps in relative terms, they're actually increasing. Another example could be education. So this shows the fraction that attains college degrees by your father's educational level and by birth cohort. So what we see is, for say, for children born in the start 60s. So if your father had compulsory schooling only, around 10% got a college degree. If fathers have a high school, then it was 15%. And if fathers had a college degree themselves, then it was around 30%. Then we fast forward up to children born in the mid-80s. We see that now, well, 20% whose fathers only have compulsory schooling get a college degree. So that's been progress, for sure. 30% whose father have a high school degree gets a college degree. And 60% of those whose fathers have a college degree get one themselves. So relatively speaking, the 10, 15, 30 and the 20, 30, 60 are the same across 30 years, across the entire expansion of the educational system. And actually in absolute terms, it's been expanding the differences. And when we look at educational mobility, overall, as for instance with the education IG, that's also gonna affect how we look at trends. So of course, these differences have large impacts because these gaps in educational attainment, they shape our lives. For instance, we can look at crime. So who commits crime depending on their own education? So here we see crime rates again by year of birth and by educational attainment of oneself. So not the mother, but now the individual itself. We see that, well, for those who only have compulsory schooling, 25 to 30% commit crime. And it continues to be so, both for those born in the 60s and for those born in the mid to late 80s. If you have high school, so much less commit crime, so only around 10%, 10 to 15%, but still a substantial gap. And if you have college, it's only around 5%. So these gaps are basically stable over a 25th period, 25 year period. If anything, they're actually increasing, even though that we have this vast expansion of the Danish welfare state. And perhaps the most dramatic consequence is if we look at lifetime. So this is life expectancy by at age 30 by your own education. So again, the same story we've seen progress. So life expectancy has increased and it continues to do so. Here we see it by education, so compulsory schooling only, high school or college again. And if you look at the differences, even though that all the groups on average see progress, the differences are basically the same. They're the same for those born in the early 60s as they are for those born in the early 80s. And that is the question. That is the central question, right? Why do these gaps arise, even in Denmark? And why do they not close? Why do they appear to be so stable? And that also raises a question about how much do we actually know about the root of inequality? And that's really important, because if we don't know what the roots are, if we don't know what causes these differences, then it's hard to address from a policy perspective. So how do we interpret these differences? So one might think that it's genes, so that it's basically all genetics and we can't do anything. So your life expectancy is just given by who you were born from, by your parents genetically and not by, say, your childhood, et cetera. But of course that's what Jim's work, among others, basically speaks against. That there's a large part of evidence that demonstrates the importance of early investments. Okay, I'll take over again. I would point out though, correctly, right? That in fact those gaps are there even when you remove immigrants from the population, right? This is not just an immigrant phenomena. So this is, I think, very important that this is something, if you want to call it, say, the many generation tradition of living and being born in Denmark and being raised in Denmark, you still find these gaps. Yeah, all these figures are from native Danes. Yeah, exactly. So that's something, and even be worse in some sense if you included the immigrant population. So here's where we're trading off again. What I wanna talk about is some work about understanding these gaps. Now the reason why this is so staggering for an American to see these kinds of gaps here in Denmark is that of course we understand in the United States why there might be these huge gaps. We have enormous differences in schooling expenditure even by, certainly across states and within states by certain neighborhoods where local expenditures really shape the nature of the school investment decision. So there are many ways Americans are accustomed to this kind of inequality, but what's interesting and what is the research challenge that we face and we are facing right now is, look, in some sense Denmark is a laboratory for an American. In many ways you've equalized schooling expenditure with much greater degree than what we see in the United States. In many ways you have a healthcare system that's far more generous than anything we have in the United States. We're moving in that direction, but it's still the case that in terms of many social services, Denmark is way ahead of the United States and yet we still see inequality. So what this is is a very useful exercise at least for this America and I hope other Americans as well to understand well if we can sort of equalize these things between Denmark and the US and say well we can eliminate and we still find inequality, we eliminate then we have to look maybe elsewhere or have to understand more deeply. And I would just reinforce what Rasmus is saying. I mean the greatest danger in social policy is to start addressing a problem based on a correlation. That's taken the world down many bad trails and we definitely want to avoid that. So we really are trying to investigate why these origin, what's the origin and what if anything can be done. I would say when Rasmus was talking about the genetic explanation, the reason why he mentioned it and the reason why we even put it on the table is that 100 years ago, if you had looked at those gaps, the social Darwinist explanation would have been dumb kids are born to dumb parents and the relationships that we saw between education of the mother and education of the child and outcome of the child is simply genetic. And of course, there is a genetic basis and that's being investigated and with new data, very rich data, some of the very exciting projects being done here at ARHOS. But nonetheless, there's a huge role for home environment and here I want to talk about what we know in the United States and what has been investigated around the world. And so we know and then we are continuing to learn about the various channels of family influence and by family I mean more also than just the immediate parents. It's the neighborhood, the environment, the community and of course, public policies. So we have many of these different channels if you look at them, parental investments, of course the heritability component, public policies, schooling, neighborhood and peers. So these influences of course have been on the table. We know that they're important but we haven't had and we're still now acquiring information about the quantitative importance and more importantly we're in the life cycle we can effectively intervene. So what we want to talk about today is some information about skill formation. So Carson was right, we're talking about skill formation but not just skill formation because we are going to get to things like the tax transfer policy, the incentives that are built into the welfare state, the incentives and the disincentives and what we can say. So I'm gonna present some American evidence and then hand back over to Rasmus on the Danish evidence. But let me just give you a couple of principles that have emerged from this literature. And this is a literature that has national and scope, international and scope I should say that we understand what's called the dynamics of skill formation and here I'm gonna get a little technical but it's very simple idea that I'm gonna say, there's two ideas, notions of complementarity and I mentioned social Darwinism a minute ago let me return to it. There's a phenomenon in social life and in fact in much of political economy when we look at economic evaluation of investments that we generally find very high returns for more capable people. That if you look for example in the United States who benefits the most from going to college? You'll find that the more able people cognitively and the more socially emotionally well equipped more motivated, more conscientious people generally have the highest economic return and social return to education. And so that's sometimes called the Matthew effect from the New Testament and the Bible to those who have more is given. There's a sense in which when we invest in more capable people we generally get bigger benefits. So that would suggest then wait a minute if we have a genetic or a notion that skills are formed at birth, a purely genetic explanation about how skills are formed, then we more or less should have a multi-class society and we should stratify society at birth. That was the policy of eugenics I would say 100, 115 years ago, 120 years ago or so. But on the other hand, so this is a fact people don't like to hear this fact but it is a fact. But then the question is, does this necessarily say that we should invest only in the most capable over the whole life cycle? And the answer depends where in the life cycle. And the change between 100 years ago and today is the fact that we no longer think that these skills even things like IQ, certainly aspects of personality and health, these whole vector that constitute life skills that these skills in fact are highly malleable. The role of the heritability is there but it's not fully determinative. And so the other part of the story is what's called dynamic complementarity. And that is the notion that if we invest today especially in young children who are known to be value malleable, young children are emerging. They're young, fluid, capable of many things. If we invest today in the early years we get a huge return downstream. Why? Precisely because of this effect of static complementarity. By raising the base and making the children more capable in the early years and where the investment returns are quite high. Anyway, I'll show you how high in a minute. What we do is we increase the productivity of investment downstream in the life cycle. So it's that life cycle dynamics that is really the new ingredient here. So our understanding of ability and abilities, what they are, how important they are and how malleable they are especially in the early years is extremely important. And so we have both the static complementarity and the dynamic complementarity and there's no necessary contradiction. The early years are essentially creating the skill base that make it better and make better or later investments. And it's also the case that young people are far more malleable than old people. You can't teach an old dog new tricks, you can but it's harder to do so. And so we have to really understand something which neuroscience, developmental psychology, many different sources of knowledge, many different fields have converged to the importance of understanding the early years and the malleability of the early years. And so one notion here is that skills beget skills. And it's not just that they passively beget skills that the early years essentially create. And these skills are multiple in nature, social and emotional skills, sometimes called soft skills. Health, very important and also producing cognitive skills. Unhealthy children, missed school, can't concentrate, poor diet. There's even now a microbiome basis for understanding how cognition can be stimulated and of course cognitive skill creates greater understanding of health practices. So we have now a deeper understanding of this vector of skills and the fact that they can be shaped. And so here's a curve with which I'm associated and I'll just explain the logic of this curve. It captures this idea of static and dynamic complementarity. So if you look at it, you can see that in the earliest years you get a very high return. It's a thought experiment that I'm making here. And that is in the very early years precisely because it creates the base for all subsequent investment and makes later investments so productive. Precisely because of that, what I call static complementarity, later on in life. That actually it turns out that you get the highest return because it percolates over the rest of the life cycle. It builds the base that makes the next step and the next step and the next step more productive. It's not to say the job training or schooling are ineffective in any fundamental sense. But what it is to say if you have a unit, first unit of value to be invested, that the earliest years would have the highest return precisely because they are producing the base for everything that follows. And so that's the logic of this diagram. A hypothetical experiment. And let me give you some of the evidence about this. And let me make a point though, which I think is on the agenda now in Denmark. At least this is my perception. Rasmus can correct me. In fact, this is really Rasmus' slide, but I will use it anyway in the interest of continuity. And that is, I think many people would agree that early childhood investment leads to better outcomes. In fact, your preschool system and your schooling system in the early years is far better developed than anything we have in the United States. It's a model. That's what Hillary Clinton was often extolling in her, not just in her presidential run, but for many years before. People understand that in Denmark and it's on the public agenda. It's embodied in public policy. But there's a second point which is that Denmark does invest heavily in education. You can see this graph by this table I put up earlier. And therefore, it must be the case that Denmark has higher mobility because Denmark invest heavily. And that doesn't follow. And you just saw that we just showed you that in fact, there is a substantial, some notion that family background plays a huge role. So we wanna really understand the mechanisms. And here I'm gonna review some lessons from interventions around the world, not just in the United States, but predominantly in the United States. We have data now from Jamaica, from India, from Columbia, as well as the United States, the United Kingdom, and Ireland. So what we understand, and we've come to understand, and this is something that is really basic to our evaluation of public policy, that much of the early work on schooling and understanding schooling talked about test scores, achievement test scores, PISA scores. OECD has pushed PISA in the last 20 years. And if you go to China, I go to Shanghai, everybody in China is boasting about the great Shanghai test scores on PISA. That's give you the mark. The whole books are written about test scores and national progress. But what we've come to understand is that these test scores measure only a tiny fraction of the skills required for success in life. Now I'll give you an example of this. There's a project that I've been working on for the last 10 years or so. I actually started out as a passive observer. Right now I'm actually analyzing data on individuals from the so-called Perry Preschool Project who are now in their mid-50s. So we've actually followed people in this intervention program for the last 50 years or so. The program was targeted to disadvantaged children. And a recurrent theme, by the way, of these evaluations, of these studies has been that the most effective programs are those that target disadvantage. And this is a recurrent theme. But nonetheless, this program targets disadvantaged children. What's interesting about it is the kids were enrolled on the basis of having low IQ scores and having a family socioeconomic status that was quite low. Many of them were single-parent families. However, this study was randomly, was evaluated by a randomized controlled trial. And so participation was determined by the toss of a coin. That was in 1962. And we followed these kids ever since. And so the program itself starts at age three. It lasts two years and it's fairly modest in terms of about two and a half hours a day during the school year. But the key idea was really having a program that supported children's cognitive and social and emotional development. So there was a lot of interaction. And the key, I think, to the program is that children and the childcare workers interacted closely. There was interaction with the parents. And the children were encouraged, not just to learn how to do math or learn how to read, but they learned how to carry out tasks, to plan tasks, to review tasks, and then to collectively work together on various projects. And these were followed over many years, as I said. And so it starts at age three. And if you look at this, and we follow these people into adulthood, we can actually compute an economic rate of return. This is after tax. So this is the way a good economist would do a good cost-benefit analysis. So taking a public program, saying yes, when you raise taxes, you distort incentives, you get less work effort, all that's true, allow for all those distortions, and you still get a 10% rate of return per annum, which is higher than the US stock market between 1946 and say 2008 or so. And if you look at what the actual outcomes were in this treatment, you can see that if you looked at the IQ scores, the IQ scores of the children in the treatment group rose enormously. And in the 1960s, when cognitive psychology was at its high point, and everybody said, ah, you know, the be all and end all is cognition, just notion of IQ, the first three years of the program reviewed as a wild success, okay? You can see that when children were six years old, their IQs in the treatment group much higher than the control group. But as you went out to age 10, there was no difference. And so by many standards, many people, people like Arthur Jensen, for example, but many others, made the remark that in fact these programs were failure. And the failure was precisely because it didn't boost IQ. And that was the goal, that was the sole goal of these programs. And yet, as I said, there was a very high rate of return. And what was the channel through which it produced its beneficial results? It produced its beneficial results through basically improved self-control, through less aggression, through children's better enhanced ability to stay on task and to work with others and to be motivated. And without going into all the details, the mechanism of success was that it promoted both on the part of parents and children, different expectations about what was possible. And this is true, every one of the programs that have been studied and where there's been a success, this has been observed as a byproduct. And not just a byproduct, probably the central reason for the success of the program. The children felt they were better able to belong to school, things they can change things. If you think about following plans, the ability to stay on task, to socialize with others, I won't get into the details of the treatment effects, but these social and emotional skill enhancements constitute the lion's share of the treatment effect. And more than that, then just essentially boosting the children's perceptions in adult life, it also boosts the home life. So what happens is that if you ask the question of the parents, one year into the program, do you believe, what is your belief in the importance of parenting on raising your child, the children in the treatment group, the red line here, are far stronger about their belief that they can do something about enhancing the lives of their own children. And if you look at warmth in the family, if you look at so-called authoritarianism, allowing the child, just forcing the child along a certain mold, much less of that. And so there really is substantial improvement in parental and child environments. And it's the latter, the parental environment, which frequently gets overlooked. The fact that we turn the parents on and they become engaged in the life of the child is probably the most important engine of this program. And it's something that gets neglected. We think of building centers, building schools, but the one lesson that's emerged from all activity is that parents are far more important than schools in educating children. It sounds contradictory, but in fact, a parent, a strong home life is the core of a successful educational program. Now I'll show you some interesting results. We just recently have analyzed the data on these same people at age 55. And so without going into great details, we actually do see effects on intelligence, add to 55. This is adult measures of intelligence. We see great enhancements in personality. I could bombard you with lots of statistics. But the other thing that's interesting too, though, and something completely unexpected was the benefit in health. Because you see people think to themselves, oh wait, school, cognition, PISA scores, IQ. Now suddenly we're changing the home life. We're changing the way that children can control their own lives, think about what's possible, and also what the parents think is possible. That translates into health. How? I mean, you're not directly, Perry was not a direct health intervention. But through measures of self-control, through measures of following diet, exercise, understanding various risks, and avoiding those risks, then what you find is really an enhanced ability. So cortisol is a measure of stress. We know cholesterol. And across a number of items, even including in the United States context, participation with health insurance. And there are strong reductions in crime, stronger for females, by the way, than males, although the dollar value is greater for males, employment effects and the like. But what's something that people have forgotten, at least haven't been able to look at and hadn't realized, is that if we look at the children of the Perry children. So now these children are 55. They have children. We can ask how well do their parents do, their children do, compared to the children of those in the control group. And we can see across high school graduation, in terms of their health, in terms of their employment, participation in school, arrest, and many other dimensions. I won't go through that. You find substantial improvements. Those are the red bars here, you see. So that's above and beyond the gain that would be at the base. So there's another project. I won't bombard you with projects, but there was another project that started even earlier. And the reason why I mentioned this project is this project starts at eight weeks of birth, much more similar in some ways to when the Danish healthcare, the childcare system starts. And it's more full day. It's more full time. And it puts together features of a program which sometimes get neglected. It provides both childcare, enriched childcare, as well as stimulation. And so what it is, it's providing very rich environments for children. And there are some health checkups given. So it starts at eight weeks of birth. And one thing that's interesting is even though I showed you a fade out of IQ, at least through 21, for the Perry preschool children, for this group, when you start early enough, you can boost IQ. And it's a permanent boost. It lasts at least through the time we've met it. We're now following these people now until the mid-40s. They don't have those results today. They're being collected right now. But what you find is enhanced engagement, greater, stronger parenting practices, exactly like we saw with Perry. Positive effect on female behavior and mental health in the sense that women now can go back to school. A lot of these are high school dropouts. Mothers are high school dropouts. They can essentially go to school. They're mental health. They have higher earnings. And so it enhances the home environment. It enhances the nature of parenting. And it also reduces both child outcomes. So I won't go through their strong effects on IQ, lasting effects. Here's the health effect that I was mentioning though. We took physician visits of people in these healthcare programs at age 35. And if you look at measures of blood pressure, pre-hypertension, a cholesterol, precursors for things like diabetes, you find that the control group, trust me, the treatment group, as far I had the control group and generally statistically significant. And here we can look at substantial lifetime benefits across these different dimensions in terms of labor income of the mother, labor income of the child, reduction in the value of crime committed and the quality of life measured in health survey. And here when we include the cost of health, we get rates of return about 13 and almost 14% per annum. And the same kind of mechanisms that work for the sake of brevity, I won't go through all this, but in fact, parenting engagement. And there were no direct home visits. That's the other part that many people forget. When you think about parenting and improving parenting, many people say, oh, you need to go visit the parent. Well, maybe, but there was no visit of that sort. But what happened is the kids came home and they were highly energized. And as a result, the parents were energized to engage the parent. So let me, I guess I should turn this back to Rasmus and get back to Denmark, which we will in a second. But let me just talk about trying to distill what the lessons are about program effectiveness. I've already told you more or less, but I think what we found and what has been found is that if you strip down these programs to their essence, this is more of a conjecture than a formal proof. But I think it's a conjecture that is bearing fruit. It's been at least found in many other studies. If you ask, well, what's the most important ingredient in these early childhood programs, it seems that the most important ingredient is engaging the parents, getting the parents active. And so there's a study that was done in Jamaica some 20 years ago, 25 years ago now, 30 years ago actually, in the slums of Jamaica. And it consisted of one activity and one activity alone, sending visitors out to mothers in disadvantaged homes and teaching them how to engage effectively with their child. And what it did, what was the engagement? It consisted of having the kids and the mother make a drawing book, tell stories, engage each other. The home visitors came out one hour a month for about three, two or three years. That's all, that's the only engagement. And yet the parents were actively engaged. And if you see what it was, very cheap because, and low cost and effective. And in fact, right now I am implementing a version of this program. We have implemented, I should say a version of this program in Western China, in Gansu, one of the poorest areas of Western China, rural China, very poor. But this is a very sturdy type of program that gets at the essence, I think, of what's going on here in these programs. And it's cheap, easily replicated, and it gets to the vital ingredient of teacher of mother-child interaction. So huge effects, I won't go through all those. But what has been a consistent finding? If we look at school effectiveness now, not just looking at early childhood, if we look at school effectiveness, the University of Chicago has been engaged in a 25-year experiment about how to increase the quality of public schools in the United States. And here, a charter school was run as a prototype for what could be done. And there's something called the University Charter School. And here I'm just showing you that the dark bars are those of participants and the light bars, those of people who didn't participate. And these are studies through the fifth grade. But what was the central ingredient? It was mentoring. It was an enhanced form of parenting. Supplement in kids entered this elementary school at age four years and are followed very intensely. And if children fall behind, they're mentored and there's a huge effect. So, and this has been a theme, for example, of successful adolescent training programs and the like. And of course, we know that education has multiple advantages in terms of wages. But what isn't so well understood is the benefit of education. And we're talking about now college education or graduation of high school on reducing smoking and promoting health, reducing crime and voting behavior and self-esteem. So there are many, many, many components. So, I think what I would just point out is that virtually all of these successful programs end up working through enhancing this interaction between parent and child and so forth. There's a very successful program. But here I think we should return to Denmark now with this study, please, yes. Well, thank you. So, as Jim said, now we're gonna return to Denmark. And as we do in Denmark, it's mostly through public policies and mostly through public policies for everyone. So it's not a targeted program. It's a universal daycare for everyone. It's universal public school for everyone. It's a focus goal. So what is the role of public policies in Denmark? Well, we've just seen that quality interventions are high quality interventions that target the disadvantaged children that that has large effects, not only in the short run, but also in the longer run. But as Jim also said, there seems to be a disconnect between the level of expenses and what we think of as high quality interventions in Denmark early in life. And then the differences we see later in life. So the gaps, the gaps we saw by mothers education, for example, early on. So what can explain this disconnect? And many things can do. And he will list four examples. And this is, as Jim also said, actually a part of a longer research agenda where we try to address all these questions and we try to investigate them so we don't have the answers to them. And certainly there can be more. And it's not only us, it's also, for instance, the Troi Foundation, et cetera. So there are as many as basically investigating this. So the first challenge in terms of the large investment in Denmark is basically what Jim talked about, what you can call the Matthew effect. That we basically end up giving more to those who have most already. And what we mean by that is, say, for instance, that we make education free, which isn't Denmark. So that's one barrier that's down. So there's a huge public investment in that. But if we don't take the other barrier to education, for instance, like skill, and equate that, then just by that difference, we actually invest in the skillful. And by the gaps we saw earlier on, which group is it that has the highest skills? It's those from affluent backgrounds. So by that extent, we actually equalize access later on, but we fail to make skillful mission equal early on. So that's the first issue. The second issue is that we actually know very little about how parents respond to all these public investments. So for instance, my child and my children are in their daycare today. So that's just not something that's added on to my investment in them. Of course, I'm here today, so I'm not with them. It's actually a substitute. So what parents do as respond to what the public sector does is not fully understood. Say that we increase quality of what's going on in the daycares. Say that now they read more with their children. Do parents then say, well, that's good. I'm also going to read with my child now. Or do some parents say, well, that's good. Now I don't have to do it. Is it complementary substitutes? And is that different for different types of parents? Unless we know, it's hard to say exactly how public investments increase in quality early on for all how that's going to affect inequality and social mobility. And just to illustrate that there is a large difference between how parents behave with their children across different backgrounds. This figure shows basically time use from the Danish time use survey. So parents, time spent with children, child age one to five, by different educational categories. So those who only have compulsory school, those who have high school, those who have college, or those who have a university degree. So the light gray bars, the ones on the left y-axis, you can see that that's time spent on reading per day on average. And what you can see is that there's strong monotonic, actually gradient in terms of education, and time spent reading with their children. Or we can also look at all characters in general, that's the black bars on the right y-axis, and it's basically the same relationship. Of course, there's much more to parenting than just reading with your child, but this is just an illustration of the differences that we see. And unless we understand why do we see these differences and also how these differences are a response to what's going on elsewhere in the child's life, then it's hard to address, basically, the differences we see later in life. Then of course, as I said, what we have in Denmark is the universal system. So we tend to have sort of a one-size-fits-all. But what Jim talked about was really high quality investments and in particular high quality investments for the disadvantaged. But when we do the same for everyone, it's then high quality for everyone. It's that high quality is for the median child, for the least able child, or for the most able child. Certainly it's hard to imagine that high quality can be for everyone if we do the same for all. And that's a political question. So should we go away from unity or should we stick to unity? And how does that affect mobility? And finally, we have solving. So it's not random where people live. So where parents choose to live. But also where the most qualified pedagogues are teachers' work. So they're better schools than others. And certainly many parents look to that in terms of whether they live with their children. So now we're gonna look at some evidence for a couple of these points for Denmark. First, let's go back to where we started. The role of redistribution and income mobility. And just to give a very, very crude example of how that affects income mobility. So this figure shows the fraction of children that are doing better than their parents in terms of income. So who has an income level today that is higher than what their parents had when they were in equal age? Across parents' income distribution. So children who grew up in the poorest family, that's to the right, sorry, to the left. And children who grew up in the richest family, that's to the right. And what we see is children who grew up in the poorest family is both for those who were born in the 70s and 80s. Around 75% are actually doing better today than their parents were. And then if we go to the richest family, we see that almost none are doing better than their parents. So that's mean reversion. But there's a very, very large proportion that's doing better. But still, some, even from those from the poorest families, I actually have a lower income today than what their parents are. And this is, however, in terms of market income. So gross income excluding public transfers. So basically what you earn from jobs, from capital income, et cetera. So now we're gonna add transfer income. So one part of redistribution to this. And look how it changes in particular in the bottom. Actually, there's no change in the top. But the bottom, if you just jump back and forth, you can see it's basically lifted. So now almost all are doing better than the parents of the poor. So those who grow up in the lowest income families, they're all doing better today than the parents. And it's basically for a large proportion explained basically by redistribution. So redistribution is important and really important when you think of inequality and income ability. But what about skill formation and investment? And these challenges that talked about earlier on. There's, of course, a large literature that studies early lifetime intervention. And Jim talked about some of it. But when we think about the universal programs, the literature is much sparser. So it's summarized in a long good L. But I think that the prime example for Scandinavia is hardness and moisture. So they study the expansion of daycare. So from selected to universal in Norway in the 70s. And they look at the effects there. And I'm gonna show you an example of those just in a second. A recent study from Germany investigates the same things, and they find quite similar results. So I think that this table basically gives the main points from the Norwegian study. So what they do is they study what is the effect of having access to universal daycare on education down in the longer run. So in column one we see the average effect on education. Years of education increased by 0.35. So that amounts to roughly four months extra education by having access to daycare at age three and four. But what I want you to focus on is actually column four and five. So if you look at here, these are the effects by mothers education. So there are no effects for mothers or for children whose mothers have high school or higher. All the effects are driven for children whose mothers have low education. And this is really the notion that early lifetime investment can boost children skills. And it's particularly relevant for the disadvantaged children. And indeed, when we look at distribution of test scores or piece of test scores in Denmark, as Jim said, we see similar things. So this is the distribution of piece of test scores for Denmark, the light rate. And then the dark red is for the US. And what we basically can see is that the distribution is basically compressed. It's shifted upwards in Denmark. But so there's basically a large mass of low-skilled children in the US that are simply not there in Denmark. So there seems to be a large skill compression at the lower tail in Denmark, both when we look at math and when we look at reading. However, there are some challenges to this. Another challenge is basically who participates, because what we talked about now is access to universal data. But access is not the same as take-up. So what the German study, so Dusmann and Allergy finds is that, well, those who are least likely to participate are actually the ones who have the highest gains. Another way to say it is those who are least likely to participate are actually those with disadvantaged background. And it's the same problem we're discussing today, in particular when we're discussing the problem in terms of immigrants and their children, that those who have the highest gains are actually the ones who are least likely to participate. So, and that, again, would go down the Matthew fake rose. Then we would be giving more to those who have more in the beginning. So that's a serious problem. Nevertheless, when we look at Denmark and when we look at the expenditures, not only in levels, but also in terms of distributions. So this shows the expenditures per child in municipalities in Denmark. So that's a dashed figure. And then in the US, we see the distribution is just much, much tighter in Denmark and similar for schools. Yet, the gaps in family structures, well, they're quite similar. Okay, so now I'm gonna talk about three short things. The first is sorting. That's the first challenge I'm gonna talk about. Okay, so it's not random what people did. So now we're gonna look at sorting across the income level. So do rich people live together or do rich people live across all neighborhoods? And similar with poor people, do they live across all neighborhoods or do they live concentrated on a scale from zero to one? So zero is no sorting. One is sorting such that all the rich people live together and all the poor people live together. Okay, so basically what we're gonna look at is the trend. So this is the trend in where people live by the income percentile in 94, 2004 and 2014. So basically what we see is that in 2004 and 2004, the middle class were quite equally distributed across different neighborhoods. So this is for the greater Copenhagen area. So they were well represented across different neighborhoods. So poor people were concentrated to a larger degree and rich people in particular were also concentrated more. But what we see actually to 2014 is that there is an increased concentration of the poorest in the greater Copenhagen area. And this is actually a trend that resembles other countries. For instance, also a problem we see in the US. Of course the levels may be different than in the US, but this is an increasing problem. So it's not random where people live. It's not random who goes to the best schools and who don't. Okay, then we get to the trade-offs between the redistribution and then skill attainment. Of course, most people may know that the returns to education, the returns to skill are lower in Denmark. And so I'm not gonna go into detail about that. So this is most relevant for those who actually have the lowest levels of skills. In particular because the redistribution really kicks in there, down there. So that's where the returns to skills can be changed a lot. This figure shows uptake in social assistance around a 25 year cut-off. So until recently, social assistance increased by 55% when you turn 25. So now we see by age, the fraction that I enroll that take up social assistance and then by the education. So those who have compulsory schooling only are those who have high school or higher. And basically what you can see for those with a low education, that there is a strong jump in take up in social assistance the minute they turn 25. And this is even in equilibrium. So this is not a reform. So this is basically even that everyone can anticipate that they turn 25 at some point. We still see this last day's continuity. Whereas for those with higher education, we see nothing. And similar we can look at the fraction enrolled in education around the same cut-off. So for those with low education, we see that their education enrollment declines when social assistance goes up. Whereas for those with higher years of schooling, there's basically nothing. So there is a trade-off here. So we can't say how to balance that trade-off, but that trade-off certainly exists. So the last thing I'm gonna talk about is basically the Matthew effect. So who can utilize from the progress that they've seen from the increased availability of services and quality services. Okay, so just let's take a brief step back here. This figure shows educational ability, the education IDE that Jim started with in Denmark for all cohorts born from 1910 until today. So as we see for those born in pre-World War II, there's a strong educational dependency between generations. So educational mobility is low. Then there is a huge increase in educational ability for those born past World War II. So the cohorts born between 40 and 60. So the coefficient between parents and children's education decline. And that's really driven by the large expansion of education in the lower tail. So all basically go from having zero, I was having seven years of education, to having nine years of education. So a huge expansion, but what we see afterwards is that educational mobility declines and it declines rapidly again. And that's basically because educational attainment continues to increase, but it increases in the other tail. So now it's all the college degrees, university degrees, PhDs that are driving the increased educational attainment. And who gets the university degrees? Well, we actually started with that, right? So who gets the college degrees? It's basically this figure, right? That those who get college degrees by their parents' education, the relative difference is stable, but in absolute terms, it's actually increasing. So one group is running away from the other group. And this trickles down the system in terms of how we think of inequality and mobility. So the last thing I'm gonna say is basically, if we look at education support. So this is, I guess, a prime example of where we think that we actually pay into those who are in need. But this basically shows the black part here, what amount you get in education support, totally from 18 to 25 by your parents' income percentile. And what we see is those who get most, actually those from the richest percent. Everyone from the 60th percent on downward get less than half than the richest percent on average. So this is actually a reverse redistribution in terms of education support. So because I think that the core and actually going back to Carsten's presentation, that in many of these aspects, there are trade-offs and there are political trade-offs, there are subjective trade-offs. So that can be more than one truth. So inequality is really at the core of many political debates. So we can't say what the right level of redistribution are as researchers. They're ambiguous and there's objective meanings about that. Of course we can highlight which trade-offs there are, but others have to decide on what the right trade-off, how they should be balanced. However, there are some things that are not subjective. That equality of opportunity in all of life aspects, that that basically requires that the fostering of skill formation is equal for all. And that's basically the point. And I guess that I used my time and more. Thank you for your time.