The Daily - More Money Was Supposed to Help Poor Kids. So Why Didn’t It?

Episode Date: August 6, 2025

For many, the logic seemed unassailable: Giving poor families money would measurably improve the lives of their children. And so a few years ago, social scientists set out to test whether that assumption was right.The results of the experiment have shocked them.Guest: Jason DeParle, a Times reporter who covers poverty in the United States.Background reading: A rigorous experiment appears to show that monthly checks intended to help disadvantaged children did little for their well-being.For more information on today’s episode, visit nytimes.com/thedaily. Transcripts of each episode will be made available by the next workday. Photo: Andrew Seng for The New York Times Unlock full access to New York Times podcasts and explore everything from politics to pop culture. Subscribe today at nytimes.com/podcasts or on Apple Podcasts and Spotify.

From the New York Times, I'm Natalie Kittrow-F. This is the Daily. For many, the logic seemed unassailable. Giving poor families' money would measurably improve the lives of their children. And so, a few years ago, social scientists set out to test whether that assumption was right. But the results of that test have shocked them. Today, my colleague Jason DeParl on how a groundbreaking experiment has undercut deeply held assumptions about how to end the cycle of childhood poverty. It's Wednesday, August 6th.

Jason, you've been reporting on poverty and the efforts to address it for. decades. You've written books on the subject. So tell me about this study that has really shaken the policy world that you cover. It's long been clear that children in affluent families do better than their low-income peers on measures of cognitive development and behavior. The question has always been why. Is it the money itself that makes a difference, or is it associated factors like Maybe their parents have more education. Maybe they live in better neighborhoods. Maybe they have access to better schools.

Maybe they have access to better child care. There's lots of indirect evidence that suggests poverty itself causes negative child outcomes. But there hasn't been a study that isolates the effect of cash itself. And this is a big policy question because most advanced countries do provide some sort of unconditional cash aid to parents raising children. The United States is unusual in that it does not. And the Democrats and progressives are eager to change that. So this question has a high public policy potential relevance.

So a group of prominent researchers set out to try to isolate the effect of money alone on children's development. Right. And the question these researchers are trying to answer is essentially if we give money to families with no strings attached, will their kids benefit from that specifically in terms of child development? That's the goal here. Right. The hopes of the proponents of unconditional cash aid is that it could help level the playing field, that it could help give low-income children some of the opportunities that high-income children have. so it has potentially profound consequences. And what they find? This group of researchers found that after four years,

there were no detectable differences between the children that got the aid and the children that didn't get the aid. No detectable differences at all? Correct. The money, as far as the researchers could tell, has not made a difference. Which was such a surprise,

the researchers themselves are quite divided over what to make of it. Okay, that's fascinating. Let's get into the study itself. How was it designed? The study's called Baby's First Years. It came together about a decade ago when a group of prominent researchers set out to try to isolate the effect of money alone on children's development.

It was quite an elaborate effort. They raised $22 million, some of it from the government, some of it from private foundations. They and a team of assistants recruited 1,000 mothers from 12 hospitals in four cities. They got them ready when the mothers were giving birth and then followed them for four years on all kinds of measures. They had neuroscientists involved, had economists involved, they had ethnographers involved, they had psychologists involved. They looked at every aspect of child development. and they divided them at random into two groups. One group got $333 a month in cash stipends, unconditional.

The other got a nominal $20 just for participating in the study. A random control study is considered a kind of gold standard in scientific experimentation because it attempts to weed out the other forms of influences on the results. This gives you a kind of scientific result. the way drug trials work, right, where you give half the group a placebo and half the group the drug. Right. The idea of randomized control

is that you really isolate the effect of the intervention itself. It's like if this is the only thing you're changing in this scenario, is that making a difference? Exactly. That was the hope. Can we pause here just for a sec? Because I have to ask about the dollar amount itself. because by my math, it's $330-ish dollars a month, about $4,000 a year.

That's just not that much. Let me step back for just a second and explain the two theories about why cash aid would help child development. One is an economic theory. It lets families buy more stuff. The kids get better food. They can live in better neighborhoods. They have more toys or books, and therefore they grow and they thrive. The second, and they're not mutually exclusive, is a stress model.

There's lots of evidence that the stress levels in low-income families are just really high. You know, people worrying about utility cutoffs, evictions, and the anxiety gets translated onto the kids, and this affects their child development. I think it's important to remember that all the women, all the mothers in this study were below the federal poverty line. So a $4,000 a year increase to their income was about an age. 18% increase. It's not transformative, but they thought it would be enough to be a meaningful test of what difference cash aid could make. At the same time, I would say there were some concerns among some advisors to the group that maybe the sum wasn't going to be large enough to produce the results they expected. So there was two pathways through which the increased

cash aid could have helped children. And to the credit, the researchers pre-registered seven hypotheses about what effect the cash aid would have. There were measures of children's vocabulary, their executive function, their pre-literacy skills, their spatial perception. They had mothers rank them on assessments of social and emotional behavior. And how do you get at the question of, you know, is someone developing cognitive Like, how do you actually tell if that's happening? This is where I thought they were really innovative and energetic. They went as far as measuring brainwave activity in infants by going to their homes and giving them electroencephalograms,

which are tests where you attach electrodes to infants' heads and try to detect brain activity and see if there are increases in brainwaves that might predict future cognitive development. They went from that to self-reports on the mothers on their children's health and behavior to cognitive tests, just a wide range of measures trying to look at all elements of the family's life. And why are they looking at children at this age, you know, from birth? What's important about that? I think there's two reasons. one, there's just a realm of evidence of the importance of the formative years that so much of what

happens in the first few years sets the trajectory for a child's development. And also it mirrors this policy debate about cash aid. If you're going to give cash aid, I think the theory is you can have a bigger impact if you target younger children than if you simply give it adults. Right. And the idea, again, is that if you give moms cash, that in and of itself could lead to improvements on any one of these metrics or on various. That's the idea. That's certainly what the advocates of unconditional cash aid hope, but again, because it's unconditional, they can spend it however they want. There's certainly people who oppose the idea that worry giving families money with no strings

attached will hurt them, that the mothers will misuse the money. They'll spend it on drugs or alcohol or it'll incentivize them to quit their jobs and the families will actually become poorer. But after four years of cash payments, the researchers found the mothers didn't misuse the money. They didn't spend more on alcohol or drugs. They didn't spend more on cigarettes. In fact, they did spend more on their kids. They spent $68 a month more on toys and books and child care and services for their kids. And they spent a little more time with their kids. None of that spending and none of that time affected the main yardsticks of child development in the ways they predicted. It didn't change the brainwave activity in the children.

It didn't make their mothers rank them higher on assessments of social or emotional behavior. It didn't improve their vocabulary or their executive function or their pre-literacy skills or their spatial perception. And the mothers didn't work less. used the money responsibly, but it didn't affect the kids in the way the researchers had predicted. The mothers are using this money on their kids, but the researchers find that that spending has no effect. It doesn't cause an improvement on any of the seven metrics. That is just pretty remarkable. Yes, and it's important to remember the researchers didn't predict it would necessarily change all seven. I think what's particularly striking is,

that it didn't change any of the seven. So then the question becomes why? We'll be right back. Jason, $22 million, years-long study, very rigorous study, no impact is. found on any of the things that they're expecting to see an impact on. You said it.

I'm dying to know why. The researchers themselves aren't sure why, and they disagree among themselves about whether they can generalize from these results to the broader question of whether cash aid helps kids. One possibility is that the pandemic interfered. The pandemic occurred when the kids were about a year old. the pandemic could have interfered in at least two ways. One, it just disrupted everybody's life.

Families came home from work, or in many cases, poor families weren't able to come home from work. They were still out there trying to make a living. They had lots of stress. People's living arrangements changed. It's just a really bad environment to try to isolate something like the effect of a $300 a month stipend. Nobody wants to do a social experiment in the midst of a pandemic. The researchers obviously couldn't have known this going into it. But a second way in which the pandemic could have skewed the results is that it triggered lots of other forms of government aid, which the families in the control group also received.

So it made the size of the experimental payments even smaller relative to the other aid families we're getting. It may be one thing to get a $333 payment when you're not getting other forms of aid. it's another thing when you're getting it on top of thousands of dollars a year in other forms of emergency aid. So maybe just the government spending overwhelmed the effect of these experimental stipends. Yeah, the idea here is that basically if you have a government that's giving out a lot of money anyway, you blunt the impact of this other cash payment. Like, it's not going to be in and of itself that important in that context. Correct.

And another way in which the pandemic might have shaped the results. is that it triggered high inflation, which eroded the value of the cash stipend over time. Initially, it was an 18% bump in income for the families that received the cash aid, but that steadily shrank with the impact of inflation. And a final caveat is that the experiment's not over. These results measure the impact of four years of payments. The payments went on for six years, so at some point we'll get data on two more years of results, and it's possible the picture will change.

But four years of results is a lot of results and some evidence shows that child interventions actually fade with time rather than increase. But no, we don't know for sure. Okay. While all of these extenuating circumstances, caveats seem completely legitimate, maybe the pandemic and inflation obscured or blunted some of the effect. But if unrestricted cash, had the benefits that people were kind of hoping it might have, don't you think you would see it show up somewhere? Well, absolutely one of the possible explanations for why the aid didn't have the expected impact is that it was wrong to expect it from the beginning. And the researchers themselves are quite divided over what to make of it. One caution to me, you can't make anything

of it. It's just a Rorschach test. People will project their own preferences. upon it. And another, quite a prominent researcher named Greg Duncan, who's a economist at the University of California Irvine, told me that these results have prompted him to change his views. He has been surprised and he thinks they can't simply be dismissed away as the product of low stipends or pandemic interference. Greg is one of the most prominent researchers in the field. He's spent decades studying the impact of government aid on children. He still thinks the government aid does a lot to boost children's development, but these results have made him question the impact of unconditional cash payments. He's the only one who so far has come to that

conclusion. It sounds like he's questioning his own assumptions, which obviously is hard to do. Why do you think he's alone in that? I think there's a legitimate, substantive, academic disagreement over how to interpret a complicated body of evidence. I think there's also some concern among the group about publicizing these data for fear that others will use them as an argument or a pretext to cut government aid. Because the results are coming at a time, a pretty fraught time for government aid itself. I mean, the Trump administration has been going about making some pretty extreme cuts to the social safety net recently. Yeah, the idea of providing cash guarantees to young children is a top social policy priority

for the Democratic Party. So there's that context, but then there's also the much larger context of the Trump administration cutting back significantly on all sorts of government aid and some of the researchers thinking that, gee, this isn't the time to publicize potentially ambivalent results from one cash experiment during a pandemic. to give you an example of how the context has changed, they did publish year one results, which showed some tentative improvements into children's brain activity in a way that could predict better cognitive development. That paper was published in a journal, and they were

happy to discuss that. It seemed very promising. It was also a different time politically. In this context, they published the results quietly, and most of the researchers declined to discuss them. What do you make of the reluctance to talk about what was found here? I mean, do you think it could raise questions that some of the researchers may be underplaying the significance of what they found? Certainly when I was bouncing the results off of people, I got that question. Why are they publishing quietly? Are they burning away from their results? I don't think that's quite right.

They did publish them. They're not hiding them. I think some of the reluctance involves talking about a paper that hasn't yet been peer-reviewed, meaning it hasn't been fully vetted by other researchers and published in an academic journal. I think probably the more fundamental reluctance to talk about it is that they are ambivalent and they do contradict this earlier finding that I think a lot of the researchers really found hopeful that cash aid could really make a difference in children's lives. I think it's hard for all of us to really rethink something we fundamentally believe in.

And all the more so in this case when there's confounding circumstances like the pandemic and the experiment isn't fully complete. That makes a lot of sense. I guess I just think there will be some people listening to this who would say, you know, I get that scientists have hopes for their research, but they shouldn't let that get in the way of being open to and really engaging with the results, even if the results aren't what you would want. And so, I mean, on those results, Jason,

you've been looking at this stuff for a really long time. What do you think this tells us about unconditional cash transfers? How should we interpret this? I want to be clear. These results did not bear out Progressives' hope that cash aid will unequivocally help children, But they didn't bear out conservatives' fears that cash aid would hurt children either. They didn't really provide full support for either side.

I mean, the mothers didn't misspend the money as some conservatives feared, and they didn't stop working. You don't have to believe that cash aid will make a measurable cognitive difference in a child's development in order to support cash aid for kids. You might just think it's a good thing to do, particularly in a world of such extreme inequality. And we don't impose a cognitive test on beneficiaries for most forms of government aid, right? There's other reasons to support giving poor families money. And the ethnographers involved in the study, the people who went out and interviewed the mothers, said that they heard mothers saying that the money was meaningful to them. They asked mothers to take pictures of what they did with it.

And one of the mothers sent in a picture of a new winter coat that she felt so proud that she had been able to give her kid. Now, that didn't translate into quantifiable changes in these benchmarks, as they predicted. But it may be meaningful nonetheless. You know, there's other ways to measure meaning in a child's life. What do you think this means for the field overall? For people who are trying to find interventions to try to address this core problem, that poor kids have worse outcomes than their more affluent peers.

Like, if you're a person trying to figure out what will improve early childhood development, what do you do now that you have this study that's saying that, look, no strings, cash, at least until now, hasn't had that effect. I mean, the question of whether poor kids have a fair shot at American life is just such a central question to American identity and to, you know, who we are as a people. I think the world's hungry for a solution. And for me, having reported on this issue for decades, I'd say the results have just deepened the mystery of why some kids flourish and others don't. I mean, there's a part of me that would have said, of course, the cash aid would make a difference.

Giving poor families more money is going to help poor families. That's self-evident. And then there's part of me that thinks, why would you think a few thousand dollars a year would make a meaningful difference in a family's life when poor kids are up against so many things from vast economic inequality to violent neighborhoods, to bad schools, to differences in parental education

and ability to help them. You could make a list of 50 things. I feel like as long as I've thought about this issue, I'm no closer to knowing the answer. and this study only deepens my puzzlement. Jason, thank you so much. Thank you, Natalie. Here's what else you need to know today.

On Tuesday, the Republican-run House Oversight Committee subpoenaed the Justice Department for its files on Jeffrey Epstein and his longtime associate, Galane Maxwell. The committee also sent subpoenas to 10 former Democratic and Republican officials it wants to depose in relation to the case, including Bill and Hillary Clinton. The committee's Republican chairman was required to send the subpoenas after Democrats forced a vote on the issue last month. If the Justice Department doesn't deliver the documents by the August 19th deadline set by the committee, it could set up a showdown between the Trump administration and Congress

on an issue that's provoked outrage within the MAGA movement. Maxwell, who's serving a 20-year prison sentence for sexually exploiting and abusing teenage girls, asked a federal judge on Tuesday to deny the administration's request to unseal to grand jury transcripts from the investigation against her and Epstein. Maxwell's lawyers said she'd never seen the material herself and that unsealing it would be a violation of grand jury secrecy. Today's episode was produced by Olivia Nat, Mary Wilson, and Jessica Chung. It was edited by Mark George with help from Lisa Chow and Lexi Diao. Contains original music by Pat McCusker and Rowan Nemistow.

And was engineered by Alyssa Moxley. Our theme music is by Jim Brunberg and Ben Landsberg of Wonderly. That's it for the daily. I'm Natalie Kitrawef. See you tomorrow.