Jesse Rothstein On Ways To Reduce Intergenerational Poverty

Siers-Poisson [00:00:07] Hello, and thanks for joining us for the Poverty Research and Policy podcast from the Institute for Research on Poverty at the University of Wisconsin, Madison. I’m Judith Siers-Poisson. Experiencing poverty in childhood can hinder a person’s opportunities throughout their lifetime and those of their children and grandchildren as well. The National Academies of Sciences, Engineering and Medicine recently released a report titled “Reducing Intergenerational Poverty.” Today, we’re going to be talking with a member of the committee that produced the report. We’ll look at the research and findings and what policy approaches may help to interrupt the cycle of intergenerational poverty and why that matters. Dr Jesse Rothstein is a professor of public policy and economics at the University of California, Berkeley, and the faculty director of the California Policy Lab. Previously, he was a senior economist at the U.S. Council of Economic Advisers and then chief economist at the U.S. Department of Labor. Jesse, thanks for joining us today.  

Rothstein [00:01:06] Thank you for having me.  

Siers-Poisson [00:01:08] The NAS report you worked on was looking at causes of and possible solutions for intergenerational poverty. Can you define what intergenerational poverty is for us?  

Rothstein [00:01:19] Yes, I’m happy to. This was actually a big topic of several of our early meetings, trying to figure out exactly what counted as intergenerational poverty. But the definition that we use is intergenerational poverty is when a child who grows up in poverty, when they become an adult, they remain in poverty. And that’s quite common. Around a third of children who grow up in poverty are in poverty as adults, which is much higher than share, obviously, in the overall population.  

Siers-Poisson [00:01:46] And how are you measuring intergenerational poverty? What does that mean for when someone was a child and then when they’re an adult?  

Rothstein [00:01:53] Yeah. So the ideal case is if we have data that we can apply the kind of current best poverty measures which we took to be the Supplemental Poverty Measure defined by the Census Bureau. If we could estimate that for households when children were young and then estimate that same exact measure for the children’s households when they’re adults, the difficulty is that data collection has changed over time. That requires going back 30 or 40 years and being able to link data from back then to now and apply the same definitions when the definitions have changed over time. There are concepts that are important to the Supplemental Poverty Measure that we weren’t even collecting 40 years ago, and so we had to adopt various approximations to try to get at it. The one metric we can use is how many children grew up in households where their parents were in the bottom 20% of the income distribution. Of those, how many are still in the bottom 20% of the income distribution as adults? That’s not exactly the same as the Supplemental Poverty Measure, but it comes pretty close.  

Siers-Poisson [00:02:58] Before we get into the real meat of the work that this committee did, can you give us a description of the work that the committee was charged with, and by whom?  

Rothstein [00:03:08] Yes. So, we were a committee of the National Academies of Sciences, and we were charged by Congress, which commissioned this report, to write what the National Academy calls a consensus report, trying to reflect the consensus of academic experts, largely academic experts, but just experts more broadly as well, on what are the causes and correlates of intergenerational poverty. What are the key drivers of intergenerational poverty? What are programs and interventions that have been shown to reduce intergenerational poverty? And how do all of these vary across racial and ethnic groups?  

Siers-Poisson [00:03:49] There was a previous NAS committee report that looked at reducing child poverty by 50%, and I believe the time frame was ten years. How does this report that you worked on fit in with that one?  

Rothstein [00:04:02] Yes, that’s a very good question. There, there obviously closely related. But I think we conceived of reducing the intergenerational poverty as, somewhat narrower type of, of intervention that we were interested in than just intervention through reduced poverty overall. We were interested in the dynamics of child development of kind of children’s life cycles that poverty makes worse and that then contribute to the child being in poverty as an adult. And so when we thought about programs to reduce intergenerational poverty, we really focused on programs that affect those developmental dynamics. So just to take an example, an intervention that’s to eliminate poverty for all parents today, and continued to eliminate poverty for all children going forward, we’re obviously eliminating intergenerational poverty, but we didn’t consider that an intervention that was aimed at intergenerational poverty specifically. That’s just an intervention that reduces poverty overall. And so we focused on these kind of long term impacts that if you could treat the parents when the kids are young, even if you never did anything further with the children, the children would be less likely to be in poverty as adults. That’s an important topic. It’s obviously a narrower topic than the interventions, than just reduced child poverty at a point in time.  

Siers-Poisson [00:05:25] As you mentioned, the committee looked at several key drivers of intergenerational poverty—I believe there were seven—and found that there were five, that there was direct evidence that some success could be made on. Could you share briefly what those five were that you saw those positive impacts for? And then we’ll dig into parts of the report that you worked on very closely.  

Rothstein [00:05:46] Sure. So, we can use various words, but one of those categories of drivers is things having to do with education and the education system. Another is health and the health care system. Another is family income and parental employment and earnings. So everything to do with how much the family earns over the child’s childhood. Another is housing, residential mobility, neighborhood conditions. And the last one is neighborhood safety and the criminal justice system. Again, because we were focused on this, this fairly narrow definition of things that reduced intergenerational poverty, we really were only able to look at programs that had been in place for a long time, such that we can see children who were exposed to them as children, and then how those children are doing as adults. And that meant that there are lots of interventions that are probably effective, but that we just don’t have enough of an evidence base about to be able to conclude that they’re effective. I think there may have been other areas where there are successful interventions, but they just didn’t meet our evidence standard.  

Siers-Poisson [00:06:55] One of the aspects of the report that you worked on specifically was K-12 education. And how did you find that K-12 education is different for low-income children than higher-income kids?  

Rothstein [00:07:10] Yeah. So I think there are many, many ways it’s different. Low-income children tend to go to very different schools than higher-income students, those schools are different in lots of ways. They are often less resourced. They’re often in buildings that are falling apart. Teachers tend to have less experience, sometimes less formal education as well. There tends to be more turnover of teachers. There’s more disruptive behavior in the school. The students come to school with less support and less focus on schooling. And that has spillover to other children. So there’s just any, just a large number of ways that the schools are different. And that’s just at the elementary level. All of this can magnify as children get older. Children who grow up in poverty tend to be much less likely to go to the most selective colleges, for example, than are children who, grew up in, in higher-income families. And all of that leads to very different educational outcomes. And then I guess the other thing I would mention that I think is important is that schools are not isolated from other components of children’s lives. And so lots of what affects how much children learn in school is things going on in their lives outside of school. If their parents are able to help them with their homework, are able to give them a quiet place to study, are able to kind of read to them before school or during school during the school years. All of these are ways that the educational process is going to be quite different for children growing up in poverty.  

Siers-Poisson [00:08:42] There are also a lot of intersections between people who are living in poverty and people who are subject to structural racism and institutional racism. How does that intersection play out in the K-12 arena?  

Rothstein [00:09:00] Yeah, that’s a very good question. Another way in which schools attended by children in poverty are often different is they’re often pretty have I mean, all schools in the United States are pretty heavily racially segregated. So if you’re thinking about Black or Hispanic children growing up in poverty, they tend to go to schools that have very few White students in them. And there’s substantial evidence that segregation itself has a direct effect on children’s outcomes. And so that’s another mechanism by which schooling ends up being different. Broadly, structural racism contributes to that segregation and also contributes to a number of ways that schools are different. And there’s evidence that children are disciplined differently depending on that race for similar kinds of infractions. I think that’s related.  

Siers-Poisson [00:09:51] Looking at the challenges that children who are living in poverty are facing and putting it through that lens of K-12 education, what do those obstacles mean for them in different outcomes as adults, which, again, if they can’t overcome those obstacles, they’re going to be more likely to be living in poverty as an adult and raising children in poverty.  

Rothstein [00:10:13] Right. So I think the simplest measures we have of educational outcomes are how many years of education you get if you graduate from college or you drop out of high school, and test scores, measures of student achievement. And on both of those children growing up in poverty end up quite a bit behind. Children who grow up in poverty, they’re less likely to go to college. They have fewer years of completed education. And also at every stage of the educational process, they have lower test scores.  

Siers-Poisson [00:10:43] And just to extrapolate, I’m sure that those impacts then have effects on how successfully they are employed and able to support themselves and their families.  

Rothstein [00:10:53] Exactly. Employment rates are measure those most directly by educational attainment. So that’s kind of the strongest channel. Employment rates are higher for people who graduated college than for people who went to college and didn’t graduate. They’re higher for people with some college, than they are for people who just stopped at high school. And they’re higher for people who just stopped at high school than for people who didn’t finish high school. Earnings are also higher. So the average earnings for a relatively young worker with a bachelor’s degree are not quite double, but quite a bit higher than somebody who finished with a just a high school degree.  

Siers-Poisson [00:11:32] The good news is that the committee was able to identify some policies and programs that are shown to help to reduce this impact of intergenerational poverty. This one is not surprising, but I’d like to know how you saw it implemented. And that’s increased spending for the poorest of school districts.  

Rothstein [00:11:53] Yeah. In some ways, I think for people who have been part of this literature for a long time, that’s maybe a surprising conclusion, because I think there was a lot of for, for many years, the kind of conventional wisdom among people who studied this was that extra school spending didn’t lead to better student outcomes. But there’s increased evidence in the last decade or so from good, causally identified studies that policy to direct more resources to the lowest income school districts do lead to better student outcomes, better test scores, better educational attainment, and ultimately to better labor market outcomes for children who are exposed. And so that led us to conclude that there was strong evidence on these intergenerational effects.  

Siers-Poisson [00:12:39] Another policy—or practice maybe is a better way to put it—that can help reduce the impacts of intergenerational poverty in the K-12 setting is increased teacher workforce diversity. How does that affect the situation?  

Rothstein [00:12:55] Yeah. So I think the key evidence we have here is about students exposure to teachers who are like them. I think for the kind of reasons of structural racism that we talked about, that are there are good reasons to think that having, for example, White teachers in all-Black schools, that may not lead to as good outcomes for students as having diverse teachers. And so, in particular, I think there’s good, strong evidence that Black teachers raise high school graduation, college enrollment of Black students. Right now, teachers in Black schools are disproportionately White. And I think that given that evidence, I think we have a basis for concluding that having more Black teachers, particularly in Black schools, would lead to better outcomes for students.  

Siers-Poisson [00:13:44] When we touched on institutional racism earlier, you mentioned how school discipline can really vary by the race of the student. And one of the findings of the committee was that the reduction of exclusionary school discipline makes a difference. What does that look like in real life practice?  

Rothstein [00:14:02] Yeah. So I think exclusionary school discipline, I think, is complicated words for suspending kids when they get in trouble. And I think there’s reason to think that particularly in high-poverty schools, there’s an increased readiness to suspend kids or to even expel them, to keep them out of school for a while. And that’s obviously not good for the student’s likelihood of successfully finishing. You’re telling them, we’re not going to teach you during your short period or a long period, and then they don’t get taught. And I think there’s, you know, there’s reasons why you might want to suspend kids, that you obviously want the school to be safe for the other students. But our read of the evidence was that there’s good evidence that there’s kind of over exclusion of students, particularly in high-poverty settings, and that reducing that can raise, school completion and reduce criminal justice, involvement.  

Siers-Poisson [00:14:55] It seems like if there are students who are already kind of tenuously attached to the school, to the education, even, like you said, a short suspension, not an expulsion necessarily, could really be that last break of that connection.  

Rothstein [00:15:10] That’s, I think, the idea and that’s what the evidence seems to be supporting, that for students who are on the verge of dropping out. You want to be pulling them in, not pushing them out.  

Siers-Poisson [00:15:20] The fourth finding on the issue of K-12 education I found especially interesting, and that was increased access to ethnic studies courses. Tell us more about what that looks like. And have you seen good examples of that happening in certain school districts, for instance?  

Rothstein [00:15:39] Yeah. So that was based largely on a study of San Francisco, where they were able to show that when students were exposed to ethnic studies courses that increased their high school graduation. The evidence is fairly new. We don’t have a ton of direct evidence on exactly what the mechanisms are, but I think there’s at least a plausible story about why that could be true, that it’s basically letting students feel that the courses they’re taking are relevant to them, letting them feel included in the process, giving them something that really interests them and engages them. I think all of those are going to lead to better student engagement and retention.  

Siers-Poisson [00:16:15] It seems like that aspect would kind of go hand in hand with that increased teacher workforce diversity that we talked about.  

Rothstein [00:16:22] I think that’s right. They’re both kind of similar in spirit in their in their mechanisms, in making school a place that is less a foreign experience for students who sometimes feel like it is.  

Siers-Poisson [00:16:36] I want to turn to another part of the NAS report on reducing intergenerational poverty that you worked on, specifically Jesse. And that’s the impact on poverty and intergenerational poverty of unconditional cash transfers like the Earned Income Tax Credit. Can you give us a really brief outline of the history of the EITC?  

Rothstein [00:16:55] Sure. Let me start by explaining something that was really complicated in our, in our deliberations, and that was trying to understand what’s the effect on children’s outcomes, just giving their parents more money with nothing else attached to it. So that’s what’s usually referred to as unconditional cash transfers. And that evidence is quite messy. There’s evidence from different settings that leads to different conclusions. In particular, there are some studies of lottery winners where lotteries are kind of a perfect natural experiment that’s pretty much randomly assigned who wins the lottery. And you can compare the people who won the lottery and not and see if the winners’ kids do better. And maybe there’s some evidence of that. But generally, I would say the way that the evidence points to the answer, really no to that, that that doesn’t lead to, better outcomes. Now, when we come back to the Earned Income Tax Credit because that’s the policy—you know, nobody’s going to propose a policy of just handing families a lottery tickets—but the Earned Income Tax Credit is a policy that is designed to get money into the hands of low-income parents. But it’s not unconditional. You can only get the income tax credit if you work. And so it’s usually referred to in the literature as a conditional transfer. It’s only available to to families that work. It was proposed by Richard Nixon actually as president. It was gradually introduced. I think the first EITC was introduced in his presidency, but it really got going in the early 90s when it was part of Bill Clinton’s Making Work Pay program that dramatically expanded the Earned Income Tax Credit. And there are a number of studies based on that early 90s expansion looking at whether children whose families got more, did better in school, were more likely to go to college. And overall, I think the evidence points to positive effects of the EITC on intergenerational outcomes. So I mentioned that there’s at best, mixed evidence about direct unconditional cash transfers. We have positive effects of this package that Earned Income Tax Credit that both gives you money but also encourages parent parents to work. And we have evidence, strong evidence that the Earned Income Tax Credit increased parental employment on the third hand—we’re running hands—but on the third hand, there’s also studies of programs that increase parental employment without increasing family resources, in particular welfare reform programs that basically turned welfare from a transfer that families got to something that was conditional on working. Those got more parents to work, but you didn’t get more resources, particularly if you work than if you didn’t. And those also don’t seem to have positive effect on children’s outcomes as well. What we concluded is that the package of increased resources and increased parental employment produce better child outcomes, and we have good evidence of that. We don’t have good evidence of positive effect of either of the two pieces on their own, and I don’t know that we fully understand at this point how that can be. So I think this is where our distinction between there’s not good evidence and what some people might want to interpret as there’s good evidence against comes in. I would say we don’t have good evidence for direct cash transfers without any conditionality. We don’t have good evidence for pushing people into work without increased resources. But we do have good evidence for the combination of those two.  

Siers-Poisson [00:20:24] The American Rescue Plan Act of 2021, expanded EITC and other economic supports during the COVID crisis, and then most, if not all, of those expansions were allowed to lapse. What effects have been documented during and after the COVID expansion?  

Rothstein [00:20:42] So we know that that package of COVID policies really dramatically reduced child poverty in that period. I would say the kind of most impactful part of that was the expansion of the child tax credits, which prior to COVID, was not available to families without earnings. These packages both expanded the amount of money available, but also made them available even to families without earnings. I think that the logic that there were families who were pretty involuntarily without earnings during COVID, it didn’t make sense to take away their child tax credit as a result. So we really dramatically reduced child poverty. And then we took it away. And we know that child poverty has gone back up. That’s not surprising. That was the predicted effect. It’s obviously too soon to measure the effects on adult outcomes of children who are exposed to this, but I think the evidence suggests that increased resources was probably quite good for families keeping them, keeping their children out of poverty and good for the children’s development. But the fact that it was so short lived means we probably won’t get the full benefits out of it that we would if it were maintained.  

Siers-Poisson [00:21:48] I’m sure the committee was working well before COVID hit and and during the pandemic as well. So the conclusions that you were able to come to in that consensus process would not necessarily reflect the day by day impact of the pandemic. But what were some of the recommendations that the committee felt that they could make based on the history of these cash transfers and the history of the EITC?  

Rothstein [00:22:15] So, as you say, we weren’t kind of looking at day-to-day policies. And again, because we were focused on intergenerational impacts, we really needed 30- or 40-year studies. And so it’ll be a while before we can get those kinds of that kind of evidence for the COVID policies. But we did conclude that there’s good evidence that would support expansions of the Earned Income Tax Credit, that expand the amount of dollars available to families without reducing the incentive to be employed. And we don’t think that the evidence points to one exact to exactly one model for that. There are a number of ways you can do that. But the key is that you expand resources to families that are working, and you can in some of them, you can expand resources to families that aren’t working, but you just can’t do it by more than you’re expanding the resources to families that are. So that there remains extra, extra credit that you get if you if you are working. And we lay out a few different examples of that in the report, but they’re just examples, not the only ways to do it.  

Siers-Poisson [00:23:16] We focused in on the K-12 education and EITC aspects because you worked most closely on that. This is a very, very far ranging and in-depth report of several hundred pages, which we will have a link to that in the program notes for this episode. Are there any of the other key overall policy and practice implications of the committee’s work that you’d like to highlight, Jesse?  

Rothstein [00:23:40] I mean, I learned a lot about a lot of the other ideas from this. I, you know, I think we had, good strong evidence of that intergenerational effects of criminal justice practices, in particular incarceration of children is a really bad idea. And I think there’s evidence against that. I think we had good evidence on health and housing policies. I think there’s lots of evidence across, you know, a range of domains that there are programs that can can improve outcomes. I guess the one thing I would highlight is that the nature of this deep dive into the evidence that we did really forced us into kind of being scavengers in the in the research literature that we had to kind of limit ourselves to things for which the right kind of evidence was available. And that means that we shouldn’t interpret our list of interventions that have a strong evidence base as a comprehensive list of what could work. I think we really need a lot more research to really try to construct that comprehensive list. And the report closes with a set of recommendations about research and data needs that would enable us to kind of really accelerate the pace of knowledge around these topics. One set of research priorities might be that we should think about all the programs now that people might be interested in 40 years from now, and start doing experiments on other effects that are obvious limits to that strategy. It’s good if you can do it, but there are limits. We also need to create data infrastructure that allows us to now go back and think, what were the interventions that happened in the past that we don’t know about the effects of, and how can we study them? And the big limitation there is that we just have so little intergenerational linkage data in the United States that it’s very hard to tell of today’s young adults: what interventions were they exposed to as children? And I think there’s a lot we could do to improve a data infrastructure that would really dramatically accelerate the pace of knowledge.  

Siers-Poisson [00:25:44] So, as you said, when we were starting out, this report by the National Academies of Sciences, Engineering and Medicine was commissioned by Congress, and you and your fellow committee members put a lot of time and effort into producing this report. What happens now? How does it get used?  

Rothstein [00:26:02] I think that’s a topic for policymakers. I think we’ve given pointers for directions that they might think about. I think the nature of this report relative to others, for example, the report on reducing child poverty, that you talked about, the nature of this is that it is not quite as direct in pointing to “you should pass this legislation to do this if you want to reduce intergenerational poverty.” It more points to areas where we know that there is there is room for more action. And then I think we hope that we’ve made it accessible for policymakers, that they understand what the evidence supports and doesn’t support at this point, and that they can run with that and start to implement some of these policies. The National Academies do not endorse policies. Our report does not in any way conclude that you should implement these. It concludes that these are policies that have a strong evidence base for reducing intergenerational poverty. And there are obviously other considerations that one would want to keep in mind in deciding whether to implement any one of these policies. So we have to do we have to be careful not to not to make recommendations that aren’t supported by the evidence.  

Siers-Poisson [00:27:10] Well, Jesse, thank you so much for joining us today and sharing the work of the committee on this really important issue of reducing intergenerational poverty.  

Rothstein [00:27:19] Thank you. Thank you for having me.  

Siers-Poisson [00:27:22] Thanks so much to Doctor Jesse Rothstein. He joined us to discuss the recent National Academies of Sciences, Engineering, and Medicine report titled Reducing Intergenerational Poverty. You can find a link to the report in the program note for this episode. The production of this podcast was supported in part by funding from the U.S. Department of Health and Human Services. Office of the Assistant Secretary for Planning and Evaluation. By views expressed by our speakers, don’t necessarily represent the opinions or policies of that office or of any other sponsor, including the University of Wisconsin-Madison. Music for the episode is by Poi Dog Pondering. Thanks for listening.