PALLAVI MURTHY – OCTOBER 2, 2023
Interview conducted by Pallavi Murthy
Professor Hilary Hoynes is a Professor Economic and Public Policy at Goldman School of Public Policy at UC Berkeley. Her research focuses on the social safety net, poverty, and the effects of government transfers on low income families. This interview took place in April 2022.
Murthy: A lot of what you research is oriented towards the social safety net and child poverty. What drew you to these subjects?
Hoynes: I think the main reason is my mom. She actually passed away recently. She was a very early user of computers and data, really ahead of her time. And she worked where I grew up in Madison, Wisconsin, at a very famous and still existing research center, called the Institute for Research on Poverty, which is the place in the United States where for 60 years they have been getting government grants and funding research to study poverty in America, pretty much since we started measuring poverty in the early sixties which was around when I was born. It wasn’t in the front of my head. It wasn’t something I experienced. I grew up in a privileged setting in a very middle class city, not a lot of rich people but also not a lot of low income people either. But there were always people around at my house, grad students at parties, and I didnt think I was very aware of it at the time, but clearly I was, because it was very formative in thinking about the kinds of work I wanted to do. My first job after college before grad school was at a public policy consulting firm where I did a lot of work on safety net and tax policy, so I think I really fine tuned my interest in that setting.
Murthy: You wrote and signed a letter supporting the expansion of the child tax credit, pointing out that it reduces childhood poverty. What have you observed since its expiration?
Hoynes: So Much. There’s been a lot of incredible research, very little which I’ve done, but many others have done, that have been real time surveillance of the efficacy of the one year expansion and following what has happened since its decline. I think we know more than we normally know when policy changes. Usually we have to wait a few years for data to come out and come up with a research design and study the policy. It was very interesting during COVID in part because the census introduced these new data sources, the census Household Pulse Survey. They had people in the field biweekly, collecting data that was very important for tracking the pandemic, mostly about the economic impact of the pandemic. Because of that data, with a lot more timely release and time granularity to study, we learned a lot of things. As soon as those expanded tax credit checks started dropping, which happened July 15th, you saw dramatic reductions in food insecurity among households with children, but not households without children, a kind of nice comparison proof that we’re kind of simultaneously experiencing the challenges of the economic landscape in the pandemic. You got one group that gets this new cash infusions and the other doesn’t, and you saw a lower food insecurity rate that persisted through the fall, and you started to see the food insecurity rates come back up again. Then there is other data that comes out of people who are analyzing data from the Chase Institute. This has data for a lot of people who are banked, which is a select sample. Not everyone has financial bank accounts, but for those that do, you can sort of see how the household balance sheet is looking when that initial cash infusion went away. It is not a good dataset to target the most disadvantaged Americans, which is where my interests lie, but it sort of concurs with this observation with having less resources, less cushion and that sort of thing. As I like to say, solving child poverty isn’t rocket science, you just got to spend the money to do it. You see that it makes a difference, you can track those differences in poverty almost in real time.
I never really thought about how COVID-19 changed how much data we have.
Part of it is what’s been happening in the background, over years and years, not just during COVID, is that more people are getting access to more creative use of different administrative data sources, that either come from states or federal governments. I have a study that tracks how states respond to COVID appearing in terms of SNAP benefits, food stamps. You can get data across states that are almost real time, how many more people are accessing this benefit, how much more is it costing, how much money is going to and to whom, and who seems to be left out. That’s one source of admin data, and of course there is the Census Pulse data. But there’s also data that comes from different third party private organizations that have allowed us to track what’s happening in the labor market. In general, monopolies are not good. One way that they are good is that they control a large share of the market in terms of data. For example, there is a company where their business model is all about doing electronic check in and checkout of workers who are paid hourly. Back in my day, there was a physical paper card that would be put in a machine. Now it’s done electronically. There is one company that pretty much does that, so if research gets access to one company’s data, you are now seeing something super granular about what’s happening with respect to employment, watching from day to day and week to week watching what’s happening to people in restaurants and other kinds of settings. There’s a lot of very creative use of data that maybe didn’t start with COVID, but I think we really saw, because everybody turned their attention to it with all this force. We saw what we had managed to accumulate over time in addition to these new data sets like the Household Census Pulse that were more reactive to the moment. It’s pretty cool actually.
Hoynes: What would you say to critics of the child tax credit, who are afraid it might be abused or disincentivize work?
You just raised the two things that are the most common things that are raised. It depends who you are talking to, but disincentivizing work is a very typical concern that comes out of folks who tend to have anti-government policies in general. And then the other piece is that “abuse” typically takes the form of something along the lines of “we can’t trust people with cash because they’re going to make bad decisions with it”. On that second point, it’s so interesting, one of the things I often say back is do we worry about giving old people social security in cash? I’ve never heard anyone worry about that. So I think in the United States, there is this legacy of racism and how it fits into people’s perceptions of who the “poor” or “low-income” population is and whether or not they can be “trusted” with more unconditional benefits. I don’t know any other way to explain the fact there has been one part of the population for who we have always just given cash and never really thought anything of it, and another part of the population where we want to either have a lot of conditionality, like “you only get this benefit if you do x, y, and z,” or we give you the benefit in kind like “food” and “food voucher”, because then we know you’re going to spend it on food.
On the criticism about how do we know if it will be abused, I usually just go to the fact that we actually have a lot of studies about what happens when people get cash assistance including studies that have tracked people’s use of the child tax credit during this time due to the Census Pulse data which basically shows that people are spending it on basic needs like most of us spend on income on. Things that showed up were food. Particularly because the benefits started in the summer, a lot of things were about back to school for the kids, clothing, school supplies, lunch boxes, paying bills and things of that sort. I think the evidence is pretty clear that this idea that people are going to spend it on wrong things is not a very common actual lived experience.
What’s kind of interesting about the child tax credit is that it’s almost a theorem that you cannot develop a plan that gives assistance to people that doesn’t distort work in some way. If I give you income, you are going to take more leisure. That is a fundamental aspect of our assumptions of models. If I give you more income, you are going to consume more of all normal goods. If you think leisure is a normal good, you would consume more leisure. And so you first have to confront the backdrop that no matter what we do, if we think that we are concerned about providing some kind of insurance or some kind of investment in children, you’re going to have to pay the price somehow. The child tax credit is basically the least distortive way you can do it, because it is designed in a way that’s a pure income effect. The reason that’s the case is that it used to be that with all traditional cash welfare in the United States, we would give you some kind of floor and then we would tax the benefit away as the income goes up. The idea is that we would want to target it at the poorest. When you design the safety net that way, you are not only imposing an income effect but lowering people’s wage. Suppose I give you a fellowship this fall. I’m going to give you three thousand dollars, but for every ten dollars that you earn I’m going to cut that benefit by five dollars. I‘ve effectively just lowered your net wage from ten dollars to five dollars and you might say “Ugh, maybe I won’t take that job at the coffee shop because my time feels really important to my studies.” So it’s just what we think would be normal economic behavior. The child tax credit is basically the constant amount of three thousand dollars per child to your income, and your income is like a hundred thousand dollars per year. You might say it is not very tragerted, and we could have a conversation about that. But there is not really a way to design a protection system that is less distortive than one that doesn’t phase out. It’s universal basic income for kids, and you’re going to get some income effect on unemployment, but it’s about the minimum you can design.
Murthy: One of the areas you research is the SNAP program(Supplemental Nutrition Assistance Program). Can you talk about how SNAP and health outcomes for children are related?
Hoynes: One of the things that I think is so important that we don’t talk about enough when we talk about these policies is not just,okay, we know that this isn’t going to lower poverty today, but what difference does it make for kids in the long run? What we know is that poverty is bad for kids in the long run. And that’s something that in a correlational way, we have seen for a long time. If you just track kids over time and compare kids who grow up with more resources versus those with less resources, you’re going to see, on average, that the kids with more resources are going to end up with more education and earning more. But that’s correlational, you might say there are a lot of things going on there. The kids with more resources go to better schools. It’s not income per say but the neighborhood they live in. There could be a lot of higher pollution rates in neighborhoods with low income. Unfortunately in America that would be correlated with higher and lower resourced families that could all point in the same direction. But we got this new generation of research that I’ve contributed to with my work on SNAP that basically aims to identify the causal effect of research per say, separate from all the ecosystems that it is correlated with. That’s important because one of the reasons we want to argue for the child tax credit is that it is actually a good investment for society because it’s just like investing in kids going to school. Why do we do that? Because it actually makes a difference in the long run. Maybe we think it’s just ethically part of society, but it also pays society back because you have higher earnings and are less likely to be incarcerated and lots of other things. And it turns out that the same thing is true about poverty. And the work that I’ve done on the SNAP program shows that when kids have access to SNAP, particularly before age five, they have better health in adulthood, they live longer, they have higher rates of going to college, are less likely to be poor themselves in adulthood, live in better neighborhoods, and for black men, are less likely to be incarcerated in adulthood. And those things are obviously good for the kids, good for the families, and good for the communities. But from a very hard nose cost benefit analysis, it’s actually good for society in terms of what it brings back. Part of the costs of the programs are paid back in the long run. I also think that that’s important because getting back to your question about employment which of course comes up all the time, the two things you raised are exactly the things that people who are against these policies raise.I like to come back and point out and say, let’s remember we can’t just talk about cost programs without talking about benefits of the program. For a long time, our quantification of the benefits of these programs and hence the cost of poverty, we haven’t been doing our job on. There wasn’t a lot to talk about. Correlations, but not really anything convincing. Now we’re at the point where we can be a little bit more convincing and say this is an investment. It’s my hope that that makes it easier to find some place of agreement. It’s like “Hey remember, this is about the kids!”. And I don’t think we had that before. It was really about the moms, it was like “Hey we do this, moms are not going to work and they are going to do the wrong thing”, and it was like an echo chamber with nothing reminding us that this is about kids. This is both my hat as an economic researcher but also somebody in the policy school where I care a lot about how to be successful with how to do things that I think are good policy.
Murthy: I think that argument is particularly convincing because it does not rely on people’s altruism, it is just better for society in general.
Hoynes: Exactly. Why do we fund public preschools? Because we think they generate returns. Kids come to school more school ready, and then they do better in school. We bought into that, into this investment framing. We have not bought into that, generally about cash and the social safety net. I think we have made some progress.
Murthy: Since we talked about COVID-19 earlier, can you talk about the transfers and the expansions of SNAP have been effective, but at the same time we have seen a lot of economic hardships in many families. How did the government transfers and the economic crisis go together?
Hoynes: I would tell you as a scholar of the social safety net that the United States did much better than what I would have expected from our treatment of prior downturns. I think the reason, to get back to my earlier point about the legacy of racism in America, has something to with the fact that it was a very shared experience. There was a lot of hardship across a lot of people and I think that made it easier to have a robust response. That’s my hypothesis, I don’t know if it is true. Measured poverty rates actually were lower in 2020 than in 2019. That tells you that it is not rocket science! We spent a lot of money, we had those stimulus payments, we topped up unemployment, we did a lot. Now people are complaining that we did too much and it’s causing inflation. That’s another story. We had a robust response, but at the same time, the safety net doesn’t catch everybody. I think there are two reasons why we saw a lot of hardships despite the fact that we spent as much as we did. The two reasons are not everybody gets access to the safety net because either they are excluded because they’re undocumented, and many of our programs don’t cover undocumented or you’re excluded by the burdens of the program, making it hard for people to get unemployment. Maybe they weren’t taxpayers so they didn’t get that check automatically. So there are two different buckets. And then the third buck is that it took us a while to get this stuff out. So there were a lot of months of hardship to get this money flowing. There were a lot of other reasons people were anxious, but in terms of the federal response, I think it was a timing issue, not everybody is covered, and even those that are covered didn’t get all the benefits they deserved because of various kinds of administrative hassles.
Murthy: Thank you for your time!
Featured Image Source: National Bureau of Economic Research
Disclaimer: The views published in this journal are those of the individual authors or speakers and do not necessarily reflect the position or policy of Berkeley Economic Review staff, the Undergraduate Economics Association, the UC Berkeley Economics Department and faculty, or the University of California, Berkeley in general.