COVID brought expansions of social welfare programs and increased flexibility. But many of the changes expired. Now Congress is considering a bit of a revival of the child tax credit expansion, but recipients of traditional welfare programs won’t see equivalent gains. Did policymakers learn the right lessons from the successes and failures of COVID-era expansions? Carolyn Barnes finds that remote appointments helped recipients but that some program changes confused them. She says we’re back to a period of retrenchment but administrators are trying to adapt when they have incentives to do so. Mariely Lopez-Santana finds that support for the child tax credit expansion was not as high as for other programs because families were not perceived as that deserving. Even recipients were not converted to program advocates.

Guests: Carolyn Barnes, University of Chicago; Mariely Lopez-Santana, George Mason University

Studies: “I don’t know nothing about that“; “Assessing public support for social policy in times of crisis.”


Matt Grossmann: Lessons from the COVID Era Welfare Expansion this week on the Science of Politics for the Niskanen Center. I’m Matt Grossmann.

After COVID struck, bipartisan policymakers expanded social welfare programs, increased flexibility, reduced burdens, and tried a unique expansion of the child tax credit. But then many of the programs and changes expired. Now Congress is considering a bit of a revival of the child tax credit expansion, but recipients of traditional welfare programs won’t see equivalent gains.

Did policymakers learn the right lessons from the successes and failures of the COVID era expansion? Are we returning to the same patterns or actually reducing burdens?

This week I talked to Carolyn Barnes at the University of Chicago about her multiple articles, interviewing recipients of the Women, Infants, and Children’s Program, SNAP food stamps, and Medicaid. She finds that remote appointments helped, but some program changes confused recipients. She says we’re back to a period of retrenchment, but administrators are trying to adapt when they have incentives to do so.

I also talked to Mariely Lopez-Santana of George Mason University about her article with Lucas Nunez and Daniel Beland, assessing public support for social policy in times of crisis. She finds that support for the child tax credit expansion was not as high as for other programs because families were not perceived as that deserving. And then, even recipients were not converted to program advocates.

Let’s start with Barnes, who studies how to reduce administrative burdens in social welfare programs.

Let’s start with the good news. You found that the COVID era forced administrators to allow remote appointments for access to the Women, Infants, and Children program. What did they do, and why did it work out?

Carolyn Barnes: I study barriers to accessing assistance programs, which public administration scholars and political scientists are conceptualizing as administrative burden. And there are three costs: learning costs, which is whether or not you know that a program exists and how to apply; compliance costs, which is basically the cost of following all the rules of the program, submitting the appropriate documents, filling out the applications correctly, responding to the demands of caseworkers, and attending appointments and interviews; and then there’s psychological costs, which is the stress and stigma you might experience when you’re trying to do all of those things.

And there’s another cost, which is my cost, called redemption cost, which focuses on what happens outside of your bureaucratic encounter or what you might experience in an agency. It’s the challenges of learning how to use the benefits you receive. This is after you successfully applied for say, Medicaid, now you have to take your Medicaid benefit and find a doctor who will take Medicaid and who will ideally give you a great experience with your Medicaid benefit.

With WIC, as a scholar of safety net programs and someone who develops and uses this framework, I like to examine programs that aren’t typically focal in the social policy literature. We know a lot about cash assistance. We’re learning more about Medicaid, especially since the Affordable Care Act was passed. But we don’t know a lot about how WIC and SNAP work on the ground, and we don’t know how WIC, SNAP, and Medicaid and some of these programs work together on the ground for families that are typically experiencing them simultaneously. The goal for examining WIC and administrative burden in WIC is to essentially give a really rich, thick description of what it’s like for people who are trying to access benefits. And I do that through qualitative research.

Prior to COVID, to receive WIC, you had to do at least four appointments. Two were required in person. You had your initial certification appointment, which is your initial intake, where they determine whether or not you’re eligible, and they set you up with benefits. And then you had a re-cert appointment, later on in the year, halfway through the year, where you had to go in in person and have the same sort of assessment done to determine what your food package would be and where you lie on the nutritional risk of the program. They assess nutritional risk along with income eligibility to determine whether or not you can get the benefits. In some states and in some counties, you had to come in person for the other two appointments. For issue benefits or benefit issuance, however you want to call it, where you would come in and the processing clerk or whoever – they call it different things – but the processing clerk would load your benefits on a card. Some states and some counties do that remotely, and some require you to come in to do that.

But with COVID, the federal government allowed states to waive those appointment requirements. You didn’t have to come in. They didn’t have to weigh you and poke you and prod you to figure out if you are nutritionally at risk. Instead, you would report information over the phone, and they would find different ways to verify that information. Maybe they’d call your doctor’s office, under your consent, to figure out what the baby weighed the last time you took them to their appointment, what you weighed in your last appointment to figure out your nutritional risk and to prescribe the food package. You didn’t have to do that. You didn’t have to go through the in-person appointment during COVID. Instead, that happened over the phone, and recipients loved it.

I was fortunate enough to do some early data collection. We were already in the field on a SNAP, WIC, and Medicaid project in North Carolina and in two other states. When the pandemic hit, we were able to continue to collect data on beneficiaries. They loved it. They found it to be very convenient because they don’t have to pay the high compliance costs of going to an appointment that’s typically multi-step and extensive. What would typically take an hour, maybe an hour and a half in a good efficient clinic is now 15 minutes over the phone.

Matt Grossmann: So we can have successful policy changes that reduce administrative burdens. But you’ve also found that new policies sometimes bring with them learning costs for participants that can deter participation. What do those costs look like? How often are people able to overcome them? And how do things look today?

Carolyn Barnes: Early on in the pandemic, and by early I mean the end of March, we were doing interviews in real time as these COVID era allowances were being implemented. We started asking people about these policy changes. We’d ask, “Well, did you know that your recertification deadline was extended for SNAP?” You essentially got a year of eligibility. “Did you know that you don’t have to come in and do interviews for SNAP?” “Did you know that you can use SNAP benefits online?” “Did you know that for WIC, the food allowances have become more flexible?”

We asked them all these different things, and nobody knew. No one knew. They were learning through us, which is in some way a public service. But it’s also, well good God, what’s going on here that these big ticket changes that are designed to make people’s lives way easier and essentially reduced administrative burdens significantly for beneficiaries and new applicants, how come no one knows about this?

I think what we were able to highlight is the importance of understanding implementation and information sharing. A lot of times we have, and you know this as a political scientist, there are really well thought-out, great policy ideas and legislation, but they get bungled when we think about actual implementation, and even more bungled, when you think of it, from where I sit, which is frontline implementation. So what’s going on in offices?

We have a series of papers, some of which you featured here today, but some of which are being written, and we’re doing analysis on. We’re trying to figure out what the heck happened. One of the answers is simply that in many ways, COVID policies or these waivers were essentially an unfunded administrative mandate. I think what we didn’t anticipate, or what federal policy makers didn’t anticipate, was the fact that there’d be so much demand for benefits. When you have demand increase and significant policy change for street-level bureaucrats to understand and implement, there’s going to be gaps. There are going to be folks that are excluded, there are going to be folks that don’t know, there are going to be hiccups in the implementation process.

Early on, within the first six months of the pandemic, that’s what we found. We found that people just were totally unaware of these COVID provisions, and they didn’t know how to ask about them. That was what was more troubling. “I don’t actually know. I thought my worker would tell me about that. I don’t know how that works.” It’s not even that they didn’t know. It’s that they actually didn’t know how to take advantage of these new policies. Yeah, that’s what we found early on.

We continued to ask similar questions throughout the pandemic and in the post-pandemic recovery. We do find that people learned. There is this policy-learning thing that happens, because they weren’t short- lived in the sense that they were only online or only actively implemented for a couple months. They were around for about a year and a half. So people learned. People learned what to expect, people learned, they learned, especially when we think about increased benefits, like the pandemic EBT and the emergency allotments with SNAP, people learned and adjusted and got accustomed to a different kind of safety net that was far more generous and far more accessible over time.

Matt Grossmann: One of the other papers that’s already made it out is a comparison of the burdens across WIC, SNAP, and Medicaid during COVID. You were able to interview beneficiaries of all of them. What did you find? What were the differences? Are they able to learn from each other in reducing those burdens?

Carolyn Barnes: It’s a similar story to what I just told, that the issue is implementation, not the policy itself. When we were doing the interviews, we would find people that had applied for benefits and couldn’t access them, even though they tried really hard. For example, because we are interested in families that would have connections to multiple programs at once, sometimes you’d get somebody that managed to get on Medicaid but couldn’t get on WIC or couldn’t get on SNAP, or managed to get on WIC but couldn’t necessarily get on SNAP and was in the process of getting on Medicaid. We were able to find folks that were applicants who hadn’t yet heard back or applicants who never heard back. I think that was the through line for the findings on that paper, that for some programs, in particular SNAP, for SNAP in North Carolina at least, we found a lot of folks who were eligible or who thought they were eligible for benefits, that applied and just never got any response from workers or from the agency.

We found that in other states too. We had interviews in Kentucky, and we had interviews in Pennsylvania. One respondent described it as submitting paperwork to a black hole. I don’t know where it went. I have no way of engaging these workers because at that time also, it was a part of a COVID era practice. A lot of work was done remotely. And when you have remote interviews instead of in-person interviews, you’re relying on the worker to pick up the phone, to pick up the phone and respond to you. Losing the opportunity to engage people in person or to submit the paperwork in person for a lot of participants was viewed as sort of an exclusionary practice. So all of that. “They’ve only created one way that I can engage them. There’s only one way for me to submit my application, and I know workers are overwhelmed”, or “I know workers are…”

In some instances, people reported that they felt like workers were willfully ignoring them. “You can’t get my worker’s attention to submit this paperwork, which means I’m not going to hear back, and I’m not going to get on benefits, or I’m going to experience a disruption in benefits.”

We find that for WIC and Medicaid, people have pretty good experiences with getting on the program. I mentioned earlier that folks really did like their remote WIC appointments. And to a degree, people liked the fact that they didn’t have to come in for their SNAP interview. They also liked the fact that they could report themselves. They could report their income so they had self income attestation, meaning they didn’t have to submit a bunch of pay stubs and all of that stuff. That eased access for WIC and Medicaid.

But SNAP, SNAP was the tricky program. A good portion of our sample reported negative experiences with SNAP and trying to access SNAP during COVID. They attributed that to the sheer demand for the program and the limited ways of reaching caseworkers.

We have the other side of the story. We have interviewed workers across all three states that work in these programs. So we know the story of why they were inaccessible, and it really is this inattention to the details of implementation and how demand can outstrip capacity to deliver benefits.

Matt Grossmann: So you had been studying these three programs prior to COVID as well, so you’ve been emphasizing kind of the implementation of the changes. But is there anything about the structure of these three programs that makes them different, have just different levels of administrative burdens? And did the COVID experience just sort of reinforce their prior differences, or did it kind of equalize them?

Carolyn Barnes: So we’ve been thinking about, or I’ve been thinking about this intersection across SNAP, WIC and Medicaid, in part because they’re the three programs that have grown since welfare reform and the three programs that families with kids under five, so young kids, those are the three programs that families are likely to be connected to at one point in time. So if you walk down the street and you see a mom with a preschooler, a low-income mom, not that you would know she was low-income. But if you walk down the street and you encounter a mom with preschoolers, she’s probably on all three of these programs. So what we learned from interviewing both workers and beneficiaries prior to COVID is that there are some really key differences in the priorities of each program and how they’re delivered. So for the beneficiary side, one interviewee summed it up like this.

She said it’s like night and day. So when she engages these programs, they’re very distinct and she noted that she enjoyed her WIC experiences relative to SNAP and Medicaid, and there’s some reasons why. So from a WIC administrator’s perspective, as I mentioned earlier, the program is structured to emphasize caseload expansion and retention. So it’s funded through a block grant, which typically has the opposite effect. So we see this in TANF. The block granting of cash assistance is what led to welfare reform and people getting kicked off the roles. WIC is block granted, but the emphasis is on keeping people on the roll. So the program is under enrolled. The coverage rate or the percentage of folks who were eligible and participating is around 50%. It hovers between 50 and 55%. So it’s underutilized and it’s a block grant program that rewards retention and expansion. So every year, your local WIC director is looking at her numbers, and if she doesn’t meet 95% of her caseload from the previous year, she loses administrative funds.

So the money follows the case. The same is the case for childcare, but that’s another conversation for another day. But in WIC, the money follows the case and there are financial incentives to expand the caseload or at least keep as many people on the program as possible so that you don’t lose your administrative dollars, you don’t have to fire anybody, or you don’t have to under invest in training or pull back on resources you would give WIC beneficiaries. With SNAP and Medicaid, the rewards are tied to efficiency and accuracy. And inefficient and inaccurate delivery of benefits results in sanctions. So with SNAP and Medicaid, there are bonuses you get if your error rate is really low. So you’ve accurately processed a case to where you’re not overpaying someone for benefits that they’re not eligible for, or underpaying folks, people aren’t getting enough benefits that they’re eligible for.

And there’s also these timeliness requirements. There’s a percentage of your caseload that should be processed within 30 days, 45 days, 60 days, so 30 days for SNAP, 45 and 60 days for Medicaid. So workers have a completely different orientation to how they’re supposed to engage clients. Another thing with WIC is that the program is designed to be long-term. So the idea is that you have up to five years of eligibility and there are these routine interactions that you have with workers over time. So I mentioned the four appointments earlier.

So if two of those appointments are in person and you have two other touch points, that’s 20 appointments over time, possibly more if you enroll as a pregnant woman. That’s 20 appointments collectively over time that you have with your WIC office and they want to be nice to you. So you end up forging these more socially supportive ties and interactions with your WIC workers in comparison with Medicaid and SNAP where the worker wants to get you in and out. So they’re not going to have personal conversations with you beyond the contours of eligibility and determining were there any changes. Are you still eligible for this program? In fact, they’re going to try to limit those interactions with you and SNAP and Medicaid beneficiaries who are also receiving WIC. They notice a difference.

Matt Grossmann: So these programs are racialized, at least in the public imagination and probably in some of the experiences of recipients and administrators. So how much does race matter in the way that these programs are administered and perceived? And we might imagine that COVID could have exacerbated that or could have made it less salient if we were in a time when there was more of a perception that we were all going through the same thing at once. So how much did race matter? Does race matter all the time? And what happened in COVID to increase or decrease that salience?

Carolyn Barnes: That’s a complicated question only because I think race always matters. And I’m not just saying that because it’s baked into social policy. The social welfare state. So if you think about the origins of the social welfare state and the systematic exclusion of Black Americans in particular from benefits, the good ones that are… The good ones, I shouldn’t say that. Benefits that are universal like social security, minimum wage, and benefits that are more targeted like cash assistance, federalism has always undermined access to those programs, both historically and in present day. That’s a well-established, I think interdisciplinary finding that we have about the role of federalism and exclusion.

Matt Grossmann: So the pandemic did precipitate an increase in social welfare expansion and an easing of requirements in total. But since then, there’s been a political pendulum swing back, including winding down a lot of those expansions and easings and a renewed push for, quote unquote, “work requirements” for some of these social welfare programs. So how did that come about? And do we see that as just kind of another in a long series of swing backs against social welfare benefits or a unique pattern that occurred in response to the COVID expansions?

Carolyn Barnes: So when I teach social policy, I frame the evolution in contemporary contours of the welfare state as a story of expansion and contraction, where expansion is brought about through crisis, and there’s this inevitable contraction or retrenchment that happens. Because the origins of these programs are primarily political and they do rest on notions of deservingness. So prior to the Progressive Era, when communities are trying to figure out how to care for the poor, these notions of who’s deserving and who’s not, whether or not we should have work requirements, how hard should we make it to access something like coal, or the in-kind aid of the day? That’s why we have poor houses, we want to deter people. So the welfare state in its infancy prior to a national or federal adoption, that’s really echoing what we see now. The state is a tool of deterrence.

Deterrence from what? Public dependency. So we don’t want taxpayers’ dollars being used to support people who don’t want to help themselves. That’s literally it. And we see that now. It’s the same argument. Nothing has changed at all. Americans have a challenge with helping the poor in whatever way you want to frame that out, whatever the target population looks like. Americans have a hard time with redistribution, and we know this in the literature. There are some conditions where they support it and other conditions where they don’t. So this pendulum swinging back reflects the historical narrative of how the safety net is developed over time. So we get a federal safety net with The Great Depression. And even then, some of the early programs around work relief, you see the deservingness narrative there. Again, as I mentioned earlier, you see the exclusion of racially marginalized populations from social insurance.

So we get a federal safety net in light of The Great Depression. It’s in its infancy. It gets some steam in terms of constituencies that support dimensions of the policy, so like social security and the like. And then we get another sort of introduction of new kinds of benefits with The Great Society. So we get social services-oriented component of the welfare state in light of racial unrest. So the way that we have to calm racial unrest and urban unrest is to invest in Head Start. It’s to provide in-kind aid programs like public health insurance and food stamps. It’s to do community action programs. It’s to do job training programs. That’s how we’re going to solve this crisis of unrest and racism, a tool to solve or to address racism and inequality. And then we have this long, long, long period of retrenchment where there’s this dismantling of New Deal error and Great Society programs or fundamental restructuring of those programs with this whole deservingness notion in mind.

We did too much for too long, and it’s costing the taxpayers a lot of money. Roles are too high. So we have to reform welfare or have to add work requirements. The pendulum swinging back to this notion of deservingness and this pattern of retrenchment and underfunding the welfare state. So this is inevitable is what I’m saying. Now the question is what will we pick up from the pandemic? What COVID era expansions will we continue in light of this broader pattern of retrenchment? There’s no new normal. It’s just marginal changes to the welfare state or marginal changes to safety net programs that will adapt as we move towards a path of retrenchment.

Matt Grossmann: So a lot of your research uses interviews with program recipients and administrators. Tell us about that kind of method relative to surveys or administrative data that are used in the same kind of field. Do you feel like we’re getting a consistent picture across those different research methods or are different methods giving us different answers?

Carolyn Barnes: So I have taken up qualitative methods as the way to study the safety net because what we knew about the safety net was based on really one program and not these other programs that are super important and have expanded its welfare reform. So the goal of qualitative methods from where I sit, is to provide rich descriptions of things we don’t know about and to do some really good theorizing. So if I’m going into policy context that we don’t really know about, odds are there are going to be new and different things that are happening in those contexts that we as academics could learn from to develop our own new updated picture of how these programs work. So that’s my goal. We actually don’t have very good data or hunches around how these programs work. Not the administrative data on spell length for Medicaid or SNAP, not that kind of stuff.

But literally what is it like for someone to apply? What do workers have to do? That kind of process. People don’t actually know how long it takes to process a SNAP case, but I know that because I interviewed workers and I asked, “How long does it take you to do this?” So, and you need that level of information to develop interventions that can hopefully reduce burdens for the participant and the worker. So the administrative data will tell us broader trends and they’ll tell us about outcomes we care about, but they don’t tell us about the process that leads to the outcomes and you can’t assume… Which is what we’ve done. You can’t necessarily assume that one program works the same as the others. And my job as a qualitative researcher is to provide granular details about how these programs work so that we can develop actionable interventions to reduce burdens and to ensure that the program’s actually doing what they say they’re supposed to do.

Matt Grossmann: So one story of administrative burdens is that this is sort of unintentional. And if researchers bring it to the attention of policymakers and administrators like you’re doing, then they might change behavior to reduce those administrative burdens. Another story, and obviously it’s probably some of both, but another story is that administrative burdens are part of this larger back and forth political process that you mentioned. They’re part of retrenchment. Some people want to impose more administrative burdens to reduce the number of recipients, and the amount spent on these programs, and to restrict who has access. So where are we in that? To what extent is the renewed interest in administrative burdens among researchers and frontline administrators, actually likely to reduce them versus, everyone’s caught in this broader political system where they’re being imposed?

Carolyn Barnes: Oh, that’s a really good question. I’m in the camp that it’s not unintentional. I don’t think any of this is benign at all. In part because, if you get something like work requirements that’s so politically contested, why wouldn’t eligibility requirements be politically contested? Why wouldn’t the length of an application be politically contested? I think that, again, if you go back to the origins of these programs, and how it’s designed to exclude people who are deemed undeserving, it’s hard to divorce politics from the milieu or the minutiae of administration or implementation. I do think that the Biden Administration did something really cool in that, they looked at the administrative state as something that they could shift, as opposed to going through legislation to change things.

And one thing that I do know, is that states and counties, even if to a degree, they are captured by local political interest. States and counties do care about what the Federal government says. And unlike opting out of a benefit like the pandemic EBT, it’s hard to opt out of how many days you have to process a case, or the kinds of information you have to collect on a client, or mandates to update your system that hasn’t been updated in 20 years. So those things seem a little more apolitical. The means at which you’re implementing these program roles seem a little more apolitical. And I think it’s been cool to watch the Biden Administration figure out ways to circumvent the intentional nature of burdens, if that makes sense.

Matt Grossmann: So we have a long way to go. But the COVID-era child tax credit was supposed to be an innovative new way to make a lasting popular program. Now let’s turn to Lopez-Santana to find out why that didn’t happen. So, what were the major findings and takeaways from your new article on support for the child tax credit?

Mariely Lopez-Santana: So first of all, thank you so much for the invitation. I am really happy to be able to share the findings of my article. So first of all, I think it’s important to say that there’s very little research on the child tax credit. Most of the literature on the US welfare state that has dealt with tax credit has concentrated on the earned income tax credit. So. I think the paper makes an important contribution there in the sense that, it provides some information and some data on public perceptions, how people use that money, et cetera, et cetera. But when it comes to the article by itself, I think the major findings, first of all, were that when it comes to how you said that Americans perceive met families as a group in need, they don’t rank as high, or as high as other groups, including the age, the working poor, but especially the disabled.

So that’s the first finding I think that should be emphasized. The second finding that we thought it was quite surprising is that, even after people receive the checks, five checks or six checks, they didn’t change their perceptions about how they view the families as a group in need. So I think those takeaways from the survey are quite important, because it tells us a lot about how is it that families as a group in American society are perceived generally as a group of in need or not, and also what type of benefits they deserve. Is it more universal? Is it more targeted and so forth? So, I think those are the main two findings that people should walk away with in terms of the child tax credit.

Matt Grossmann: So you have some comparisons to compare the child tax credit to other programs, and the recipients of the child tax credit to the recipients of other programs. So, walk us through what you compare it to and what you find.

Mariely Lopez-Santana: So, in the paper what we do is, that we ask a bunch of questions about, first of all, support for programs for elderly people, for the disabled, and also for families, those three groups. Then when it comes to the programs, we ask about SNAP, which is typically what we know, food stamps. We ask questions about the earned income tax credit, and then we also ask questions about the child tax credit. So, the point of reference in this paper are the disabled, the elderly people, and also working families in general.

Matt Grossmann: But you did find that these programs aren’t unpopular. You were just kind of comparing their relative levels of popularity.

Mariely Lopez-Santana: Exactly, exactly.

Matt Grossmann: So how popular are they, and how big of an emphasis?

Mariely Lopez-Santana: Right. So, hen it comes to the elderly people, they rank in the middle in comparison to the disabled. So to put in another way, the disabled rank on the top in terms of need. I believe just looking at the data from the paper, about 83% of the samples support giving aid to this population. Then when it comes to the older people, about 79% of the respondents supported giving aid to this population. Then the low income was about 63%, and then finally family with children was up at 61%. When it came to the programs, as SNAP ranked higher, that’s food stamps, higher than the other programs. So that was 63%. The earned income tax credit was 58%, and finally child tax credit 54%. So, in that sense the findings are consistent, because the child tax credit ranked the lowest. It’s true that it’s the majority of the people who supported these programs. But in comparison to the other programs, it ranked the lowest. And then also when we compare the groups, families also ranked the lowest in comparison to the other three groups.

Matt Grossmann: So your other surprising finding was that, receiving the benefit didn’t necessarily increase support for it. So, walk us through how you found that, when you were in the field and what was occurring at the time, and what you asked to determine that?

Mariely Lopez-Santana: Sure. So, the year that the child tax credit checks went out, we conducted the survey with… We used YouGov to run the survey. And we asked about, it was close to 1000 respondents the first time, a group of questions about levels of support for these programs that we provided. Also when we designed the survey, it was designed in a way that we ran an experiment. One group of people received the questions that basically, how do you support the child tax credit? Whether you support the earned income tax credit. Whether you support SNAP and so forth. And then the other group of people instead what they receive and they responded to it was, a description of the program.

So our sense also to get more data was the idea, okay, perhaps people without having any type of information about this program will respond differently, that if you simply ask them about a program. In the end it didn’t make any difference. The findings were very similar to each other. And then in the end we merged the data. Then sometime after Thanksgiving, so that was about November, after people received five or six checks, depending on timing, then we asked the questions again. So, about four or five months passed by, and then we asked the same questions with additional questions. So for example, how do you use the money, whether you were eligible, et cetera, et cetera?

And then we ran again, the experiment. And it was very surprising for us. The responses were almost exactly the same. So it presented that conundrum, right? Because we went into the field with these hypotheses about how is it that support by the government will change the perception of these programs? But that wasn’t the case.

Matt Grossmann: So, the proponents of this child tax credit expected that receiving it would lead to demands for continuing it. It was scheduled to be temporary, but the idea was, “Well, people are going to keep wanting it once they get it.” So what are some potential explanations why that didn’t turn out to be the case?

Mariely Lopez-Santana: We don’t have specific data for that. But at the end of the paper, we provide some explanations, or some potential explanations. First of all, it is the case that most people receive many different type of benefits from the government. So it was very difficult for people to really differentiate whether you were getting a stimulus check, or whether they were getting benefits from the child tax credit, especially if you got it as a direct deposit. So we thought, well, people are not able to differentiate, so why is it that they should change their perceptions? They’re just getting money from the government.

They assume that it is pandemic related, but they can’t truly differentiate one for the other. The other thing that we have been thinking a lot, and also for example, President Biden hinted at this, it was an issue of framing. How is it that the Democrats were not very effective in framing the child tax credit as support for families? And especially, how is it that families as a group have a common set of challenges? So for example, child care, if also in general, if you look at the economy and also the different trends as the United States. Low fertility rates compared to other countries, women tend to stay at home at higher rates than in other countries, for example, in other advanced democracies. So this type of framing was not very effective.

So, people can really understand, we’re in this together. And then finally of course, at the same time more or less, September, October or so, this is when Republicans really started to talk about how is it that support by the government really was linked to inflation. So, this idea that the government was giving people so much money, that then in the end, that led us to, down the road that everyone knows in terms of inflation. Whether that’s true or not, that’s a question. But I think that in turn might have affected how people perceive these different type of programs.

Matt Grossmann: So as you said, you also conducted a survey experiment, where you separated people answering just from the name of the program, versus the function of the program, and you didn’t find a big difference there. And it seems that that might cut against some of these explanations. Like, is it really framing if the program name isn’t that important? Is it really these connotations about who the recipients are, if people actually do know what the program does and just half the population really just isn’t for it?

Mariely Lopez-Santana: Right. That’s a very good point. But to some degree I think, the findings are very consistent with the literature. Who is the deserving poor? Who are the undeserving poor? What the survey and the paper adds in relation to that is this idea, okay, where is it that families fit, right? Families as a group. And one of the issues with this tax credit is that it’s universal. So, you have all type of families with all sorts of backgrounds, including high income. It’s true that a small percentage of the highest strata didn’t have access to the child tax credit, but it’s not the same to talk about middle income versus lower income, right? But going back to the point, I think the idea that the welfare literature has really emphasized how sticky the welfare state is and including the perceptions of how people understand and view certain groups in terms of the need and also how deserving they are. I think the findings are very consistent with that. For the longest time, the disabled, the elderly people, those groups are tend to be seen as the deserving group, right? The working poor, less so, right? If you refer to the poor that are not working, those truly go in the undeserving category, right?

So in that way, one implication is, well the pandemic wasn’t really a critical juncture that really changed the perceptions of people, how is it that they view these groups. And welfare policy and social policy is very sticky, it’s very difficult to change. And in the same way it’s very difficult to change how is it that people perceive the government social benefits and also these groups.

Matt Grossmann: So you say it’s difficult to change, but we just went through a pretty large expansion of the welfare state that was temporary. We expanded a lot and then we retracted a lot. So it seems like there was, and the same kinds of factors were somewhat reflected in public opinion that is there was support for expansion under COVID, and then there was support for not continuing some of those benefits afterwards. So is that just normal thermostatic politics when things increase, some people are going to go in the other direction.

Is that just the normal history of welfare politics that we get expansion and then retrenchment or was there anything unique about this period?

Mariely Lopez-Santana: Well, I think it’s important to say that the child tax credit was first introduced under Clinton, right? When he and others of course decided to end welfare as we know it, right? So then after that, and I think this is consistent with the literature, we see that this program has been in place with incremental changes in the sense that it has become increasingly more generous and more expansive in terms of covering the lowest strata of society. But when it comes to making it a child allowance, right? As it was under Biden, we saw that it was temporary, it was mainly justified because of the pandemic, right? But then after that, we just went back to the regular child tax credit that we have right now, right? So if we refer to what is going to, if it happens, right? The conversation that we’re having right now, it will be also about incremental change, right?

In the sense that it wouldn’t become a child allowance in the same way that many European countries have it. It will just make it more generous and more expensive in terms of, the problem with the child tax credit is that it hasn’t really covered the most disturbing in society, the lower strata, it has been mostly a tax credit for middle income and also with Trump, higher income population also benefited from it. But the most needy for the most part have been left behind, right? So what if we see this reform taking place is just another incremental change and still we’re talking about this will be temporary, right? It will be until 2025.

Matt Grossmann: Yeah. So as you mentioned, there is action in Congress on this, and even if it looks like Congress isn’t doing anything, there is actually a Democratic and Republican joint bill that looks like it has some real opportunity to become law that matches a re-expansion of the child tax credit with some business tax incentives, which have also expired. And so that might suggest that, like we discussed earlier that there’s kind of a trade-off between the hidden and the visible welfare state. On the one hand, maybe the visible welfare state as Biden thought leads people to know their recipients and to push for it. On the other hand, maybe that just politicizes the issue and bringing it back to what seems like it’s tax politics behind the scenes and isn’t really welfare politics, is actually helpful to getting something to re-materialize.

Mariely Lopez-Santana: Yeah, that’s correct, right? If we see what’s happening right now, we don’t really see that it is being saw as welfare reform. It’s that it is being sold as a tax credit, right? It’s a tax reform. So in that way, it’s a way of not pushing each other’s buttons in that way, right? And also it is being sold as a trade-off, right? We are expanding to tackle this population. However, the risk that also benefit from the tax hidden welfare state, right? For example, when it comes to tax credits for mortgages, and if you have an enterprise when it comes to healthcare for your employees, you’re going to get those type of things at the same time, right? So it’s that trade-off, but it is not being sold as welfare reform.

It has been sold as perhaps having that potential to reduce poverty, right? Because I think the message that after the child tax credit expired, poverty went up quite significantly, right? It is bad for politics, right? So I think it’s that’s the main reason, right? Why I said that they’ve been able to do that without major hesitation, is because it’s been sold as a type of benefit that most people do not understand and don’t care about, even if many people benefit from it. Because the data shows that the EITC and the child tax credit are the most beneficial program, social programs in terms of reducing poverty. So that’s a huge takeaway I think.

Matt Grossmann: So you also study European welfare states and politics, and I know that these have traditionally been a part of what talks about how the American system is different, that it’s more hidden, but also that it’s less generous to the poor. So how did your findings here fit into kind of this wider American exceptionalism on social welfare states? Do you think you would’ve found the same kinds of stuff everywhere or did this repeat the American patterns?

Mariely Lopez-Santana: No, the US is quite unique in that regard. For example, as I noted in the US, we don’t have major family policy, right? In the European sense. For example, in Europe you have many countries that have these so- called baby checks, right? You get benefits for having child, something similar to the child tax credit. You get child allowances as well. So if you decide that you’re going to have a family, there’s going to be some type of social assistance that is associated to having kids. In the United States, we know that that’s not the case, right? So that’s a major difference. And if you look at the data, the United States always see the data. It’s important to say, right? The US ranks the lowest in terms of percentage of GDP spent on family policy.

Matt Grossmann: So when these programs were being designed and first implemented, there was some hope among supporters of a larger welfare state that this would be a kind of a ratcheting up opportunity. They thought that they had learned the lessons from previous stimulus projects that they had. They obtained support under Republican president and large bipartisan majorities to pass several major bills that included social welfare expansions. And yet we got all the same politics back. We soon got complaints about the effect on work, on the effect on inflation, on the same values that are always used to favor welfare state retrenchment. And many of the programs or program expansions did not last. So what do you think the legacy of this will be? And next time there is an opportunity of this sort, will anything be different?

Mariely Lopez-Santana: Yeah. And the paper is interesting that way, right? So it is a way of how it said that you can publish journal findings because we went into this spec in certain things and we were quite surprised with the findings, right? In the sense that families did not get as much support. Also, after people got the checks, perceptions didn’t change, right? So in that way, what is it that we can learn about this? Well, first of all, one lesson for us is that we overestimated the idea that there will be a policy feedback, right? And then actually once you start to look into the literature, the literature is quite clear about how is the policy feedback doesn’t happen automatically, that it might take years or decades to be able to be to observe and make those policy feedbacks really tangible, right? So there are some books that talk about 20, 30 years, right? So perhaps in 20, 30 years we will see that policy feedback about the child tax credit, right? We don’t really know.

Also, perhaps right now what we’re seeing is a policy feedback from the Clinton years, right? So that might be one observation. If Democrats especially, right? Want to pass welfare reform, perhaps the road of less resistance is to do it through the tax system, right? It has been proven that you can actually reach that population and reduce poverty without being so visible, right? So that might be an important lesson for Democrats.

Matt Grossmann: There’s a lot more to learn. The science of politics is available biweekly from the Niskanen Center. I’m your host, Matt Grossmann. If you like this discussion, here are the episodes you should check out next, all linked on our website. How administrative burdens undermine public programs. Can Democrats design social programs that survive? How Obamacare and Medicaid drive voting? Does anyone speak for the poor in Congress? And why rising inequality doesn’t stimulate political action? Thanks to Carolyn Barnes and Mariely López-Santana for joining me. Please check out their work and then listen in next time.