Katherine Michelmore On Effects of the Expanded Child Tax Credit on Housing Stability for Low-Income Families

Siers-Poisson [00:00:00] 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. For this episode, we’re going to be talking with Dr. Katherine Michelmore about the paper that she coauthored with Natasha Pilkauskas and Nicole Kovski titled “The Effects of the 2021 Child Tax Credit on Housing Affordability and the Living Arrangements of Families with Low Incomes”. You can find a link to the paper in the program note for this episode. Katherine Michelmore is an associate professor in the Gerald R. Ford School of Public Policy at the University of Michigan, she’s also an IRP affiliate. Michelmore is a leading scholar and educator on the social safety net, education policy, labor economics and economic demography. Kathy, thanks for joining us today.

Michelmore [00:00:56] Thanks for having me. I’m glad to be here.

Siers-Poisson [00:00:58] I’d like to start with a brief overview of the CTC (Child Tax Credit) before the pandemic expansion in March of 2021. Who was eligible then and what did they receive?

Michelmore [00:01:08] Sure. The credits been around actually since the late 1990s and historically it’s been a tax credit targeted towards mostly middle-class filers. So it’s had a phase in structure, kind of like the Earned Income Tax Credit, if your listeners are familiar with that credit. So, the higher your earnings are, the larger your benefit would be. And then it has a longer plateau and then that used to be phased out around $150,000 for a married couple. You know, there was a big reform to the credit during the Trump administration that really extended the plateau region so all but the richest children essentially in the U.S. were eligible for the full credit. But on the other end of the spectrum, some work that I’ve done and lots of others have shown is that the poorest one third of children in the U.S. were historically not eligible for the benefit. And this is for a couple of different reasons. First, as I mentioned, there’s a phase in and so you can’t receive the full credit, it depends on how many kids you have until you’re earning about, you know, somewhere between $20,000 and $50,000 a year. And unlike some of the other tax credits, we have that kind of start phasing in at the first dollar of earnings, the CTC has also historically had a fixed earnings minimum. And so as of right before the expansion in 2021 households had to have at least $2,500 of earnings in order to receive even a dollar of the CTC benefit. And that might not seem like a lot of money, but studies have shown that about 5 or 6 million children actually are completely ineligible for the credit because they live in households that don’t meet that income requirement.

Siers-Poisson [00:02:35] So what changed then in March of 2021?

Michelmore [00:02:38] So there were a lot of really big changes to the credit in 2021. First they removed that phase-in structure in the minimum earnings requirement and they made the credit fully refundable, which essentially means that even if you don’t have any tax liability, you can still claim the credit as part of your tax refund. That was another key reason why historically some families weren’t eligible for the full benefit, is that it wasn’t a fully refundable credit. And so that was one big change, is that now even families with no earnings could receive the full credit. Another big change was that the size of the credit was increased dramatically. It used to be $2,000 per child in 2021 is to expand it to $3,000 per child. And for the first time, there was a larger benefit introduced for children under the age of six. So kids under the age of six could get $3,600 a year. And then finally, one of the biggest changes, it’s more about the distribution of the benefit, is that instead of getting the credit at tax filing time as one big lump sum for that year only, families could receive half of the credit in monthly installments for the last six months of the year from July to December of 2021. And then they would claim the rest of the credit when they file their taxes in early 2022.

Siers-Poisson [00:03:46] In this paper, you’re specifically looking at that expanded child tax credit and housing. I’d like to just pause a bit and ask you to give us some background on the challenges in housing stability and affordability that low-income families often face.

Michelmore [00:04:02] Yeah, it’s a great question. So housing instability and affordability have been growing problems in the US. We think that low-income households are particularly vulnerable to issues of affordability. Some recent estimates from the U.S. Census Bureau for instance, just this month released their poverty statistics numbers, and they shared a statistic that about half of all renters in the U.S. are what we call at cost burden. And that means that they’re dedicating at least 30% of their monthly income towards housing costs. That’s kind of the threshold we’ve used historically to denote an amount that’s kind of I don’t know if too high is the right word, but that’s more difficult for families to meet that requirement. And of course COVID I think exacerbated some of these concerns and some of these trends. And, you know, I saw some recent estimates well, some estimates from the time that suggested that about 25% of the lowest income families were behind on their rent. So this is this is definitely a problem that was a problem before Covid and I think Covid exacerbated some of these concerns.

Siers-Poisson [00:05:01] You say in your paper that housing is an important factor to consider specifically for child well-being and development? Can you give us a little more information on that?

Michelmore [00:05:10] Yeah, I mean it’s difficult to really pinpoint a causal relationship here. But we know from descriptive studies that children that live in families that are denoted as having a household crowding, where there’s kind of more people in the household than rooms in the house. These are associated with negative academic and social outcomes for children. There’s also, you know, evidence from experiments moving children to two different neighborhoods that that kids do really benefit from being in more stable housing and in better neighborhoods. So housing is really crucial for children’s well-being.

Siers-Poisson [00:05:39] Kathy, who is in your data sample and where did you get the data that you’re basing this research on?

Michelmore [00:05:45] I think this is something that sets our work apart from some of the other work in this area is that we are really lucky to have this relationship with providers. A company created this app Propel, which is an app that helps food stamp users track their benefits. So this might seem a little bit strange, but prior to this app, food stamp users have a really hard time knowing how much they have in their benefits without having to go through a couple hoops to figure that out. This app essentially lets these food stamp users track their benefits in real time. So the majority of the people in our sample are people that are currently on food stamps. Essentially Propel will push out a monthly household survey to some subset of their users to ask them questions about their economic well-being, some questions about material hardship and so on and so forth. And importantly, while this is a bit of an unknown quantity, this dataset, we do note that about 25% of food stamp users use this app so it’s quite a large population. And actually, many people continue to use the app even after they have gone off of food stamps. That helps them track other benefits and there’s often coupons and things that get pushed through the app so, a lot of low-income families still rely on the app. Again, this is kind of food stamp users, think about those who are whose income is largely below 130% of the federal poverty line. I think the average monthly income in our in our dataset was about $500 a month so it’s a pretty low-income disadvantage sample. But we’ll really compare the characteristics of our sample to other kind of nationally representative samples. We see that it looks broadly representative of low-income populations that are in the American Community Survey and things like that. I think another thing that sets our work apart is that, you know, because we have this unique data set, we were fortunate enough to be able to add some questions to the survey. So we actually have a lot more detailed information about some of these markers than it’s available in other surveys. A lot of the prior work, for instance, relies on the Census Pulse survey, which is, you know, a great real-time data source but doesn’t have some of the rich information that we have. For instance, they don’t ask people if they moved and why they moved, there’s less information about who’s living in the household. And while it might seem like a minor detail we actually really appreciate that people self-report how much they own back owed rent. Whereas in other data sets they often just ask a question of “Are you up to date on your rent?” and so this allows us to, again getting in the weeds a little bit here, to be able to differentiate whether the credit might be beneficial in both kind of eliminating debt that people might have for back owed rent. But also, if it’s not enough to help them eliminate that debt, whether they’re less in debt. So it was important to us to be able to look at both, whether they had any back owed rent and how much back owed rent that they had.

Siers-Poisson [00:08:20] We’ve touched on some of these aspects of the impact of the expanded child tax credit on housing for low-income families. But I want to look a little more deeply at a few of those. Let’s start with moves that are driven by difficulties in affordability. Can you give us a picture of what that might look like in the life of a low-income family?

Michelmore [00:08:40] Yeah. What we have in our survey is we have information, essentially our survey asks individuals if they think that their housing situation might change the next month. And then if they answer in the affirmative or give some chance that that might happen, they’ll ask families kind of the reasons for those moves. You know, we rely on this kind of unique data set to look into these issues. But some of our common nationally representative surveys, like the current Population Survey, will also ask questions about why families move. What we’re really trying to get at here is moves that are driven by some kind of indicator of affordability or are costs being prohibitive for families, and so oftentimes other surveys will ask questions like this as well, where they’ll give us a set of reasons why families might move. And so we were particularly interested in understanding whether individuals were saying that they were going to have to move, essentially because they were worried that they couldn’t afford their rent or their mortgage.

Siers-Poisson [00:09:32] Because I can imagine for some families, maybe not as much for low income, but a move could be a positive thing. They’re moving to a better neighborhood, to a larger apartment, to a place with better schools. But it doesn’t sound like that’s what’s happening here.

Michelmore [00:09:45] Yeah, exactly. And, that was the reason why we wanted to focus specifically on this reason why people said that they were moving. We wanted to specifically be able to differentiate between people who are making what we think might be kind of viewed as more positive moves. You know, we know from some other work that’s been done on the expanded CTC that some families will use a credit to kind of put a down payment on a new apartment or move to a better neighborhood and we would certainly view that, I think, as a positive outcome. But we were really interested in whether the affordability piece was the reason why people were moving, which we might think is not so good of an outcome. You know, it certainly could be the case that someone was renting an apartment that was too expensive and so maybe the money helped them put a down payment at a cheaper place so perhaps that could be a positive outcome. But in general, we think it’s concerning if a family is forced to move because they don’t believe that they can afford their rent.

Siers-Poisson [00:10:33] You also look at whether these families owed back rent or mortgage payments, and if so, how much they owed. How common a condition is it that folks like those that you were looking at do have those payments that are hanging over their heads?

Michelmore [00:10:48] Yeah, it’s actually incredibly common in our sample. It was something like over half of our sample said that they had some kind of back owed rent or mortgage. And this was much higher than some other estimates in the literature. I saw some estimates based on the census pulse data, for instance, during this time period that even among low-income households, only about 20% to 25% of families were behind on their rent. This just, I think, speaks to the really economically disadvantaged nature of the sample that we’re looking at is that the majority of people in the survey said that they were behind on their rent.

Siers-Poisson [00:11:22] Another aspect that you looked into was living arrangements, and that includes whether a parent is living with a partner, whether they and their children are doubling up and also the whole household’s size. Why are those important factors for these families?

Michelmore [00:11:38] Yeah, it’s a good question. So I’ll maybe I’ll start with the doubling up one. This is a marker for whether you’re living with somebody else in your household that’s not part of your nuclear family. And there’s kind of a mixed take on how we view doubling up. It’s generally considered not a great outcome for children mostly, I think because it’s often thought of as like a precursor to homelessness if someone is kind of sleeping on your couch or something like that. But absolutely, there are certainly arrangements that could be very beneficial for children if they’re living with a parent and a grandparent, that grandparent might be there to help support raising the child. But overall, we tend to think that people do better when they live more independently. The living with a partner indicator, we don’t actually have a lot of great information on, we don’t know if this is a marriage, we don’t know if it’s a cohabiting union. We don’t know, for instance, if the partner is the parent of the children that are in the household, but we were just curious. Essentially, we wanted to kind of map out who’s living in the household besides the parent and the children to get a sense of what those living arrangements look like. And in terms of household size, again, this I think relates to our understanding from previous work that household crowding is not a great thing for kids. In our measure, we don’t have a great way to concretely measure crowding because we don’t know how big their houses are, but we just think in general a marker for how many people are in the household would give us some suggestive evidence on whether the children are experiencing household crowding.

Siers-Poisson [00:12:58] All right. Let’s dig into some of the results. Let’s start with housing affordability, what did you find?

Michelmore [00:13:06] We looked at a couple of different markers of housing affordability. We looked at, as I mentioned, this indicator for whether families owed any past due rent and how much rent that they owed. And we also, as another indicator of affordability, use this question about whether individuals said that they might move because they’re worried they can’t afford their rent. Across the board, we found that families that were eligible for larger CTC benefits were less likely to be behind on their rent or mortgage, and they owed less on their back owed debt. We think that’s coming from those people not having debt at all or being able to pay off all that debt, but also could be picking up on people like I mentioned, who might be less in debt but still might carry some debt. And then finally, we did also see some evidence that individuals reported that they were less likely to say that they might move because they couldn’t afford their rent. So all of those things we thought signified that people were better able to afford their housing.

Siers-Poisson [00:13:59] In general, do we know what low-income families were spending that expanded tax credit on? Was housing one of the biggest things that they were prioritizing?

Michelmore [00:14:10] Yeah, we asked our survey respondents this question in just like an open-ended way and we let people kind of respond and the vast majority of people in our survey said that they spent it on “bills.” So that’s kind of hard to differentiate exactly what those bills mean. We think some of that could be things like back owed rent, could be credit card bills or utility bills. And from other data sets housing is a significant expenditure that others have shown that people spent the credit on, I think housing was a big one, the other big one is food.

Siers-Poisson [00:14:42] Let’s move on to living arrangements. What about changes in whether a parent was living with a partner or not?

Michelmore [00:14:50] We looked at a number of different things and one thing I think that was nice about this survey is we can almost in very real time ask respondents whether or not they experience a change in their living arrangement. Which is often not available in other surveys or we might know that on a year to year basis. But in this case, we knew in real time, did your living situation change in the last month? And so in our study, we found that families that were eligible for larger CTC benefits were more likely to report that their living situation had changed. And the magnitude of that response was almost exactly the same and in opposite direction to the effect we found on living with a partner. So let me back up there and say essentially what we showed was that families that were eligible for larger CTC benefits were less likely to be living with a partner. So we think that when people were reporting that their living situation had changed, most of that seemed to be driven by them being less likely to be living with a romantic partner.

Siers-Poisson [00:15:42] And in your paper you note that often lower-income folks might make decisions about sharing a household based maybe more on economic needs or a crisis that one or more of them are facing than, “we’re at a stage in our relationship where moving in together feels right.”

Michelmore [00:16:03] Yeah, I think that’s right. As I said, it’s one challenge that we have is we can’t determine the nature of these relationships in the sense that we can’t distinguish between marriages and cohabitations. We know, for instance, from a large literature on cohabitation that a lot of times couples will move in together for financial reasons, for cost savings and often very quickly, especially in lower income and lower educated populations, that these couples often kind of progress very quickly. And it’s not necessarily, as you said, because of a progression of the relationship. We also know during COVID there were, you know, lots of either anecdotes and other statistics on couples getting kind of trapped, living in a house, you know, for lack of a better word, like trapped, living with a partner that they may have kind of ended romantically but didn’t have an easy way to get out of that relationship, maybe for cost reasons or because of the pandemic. We’re really not entirely sure what the nature is of those relationships. But we think, you know, finances certainly may have played a role in that.

Siers-Poisson [00:16:59] I found the changes you saw in doubling up really interesting because it seemed like it depended on how much the people were making, which direction the doubling up might go.

Michelmore [00:17:11] In our main results, we didn’t actually find a significant relationship between the CTC benefit and doubling up. And as I mentioned earlier, you know, people double up for a variety of different reasons. It could be for health reasons to care for an elderly parent, or it could be for financial reasons. It could be to bring in a grandparent who might help with taking care of the children in the household. It’s more of a complicated arrangement than just a purely financial one. And so we didn’t see anything overall in the sample. And so in some of our subgroup analyzes where we split the sample by household income. We did actually see some indication of a change in doubling up that actually went a little bit in the direction we didn’t expect. So we saw actually that, I use the term higher income here, it’s not totally accurate here, that these are families that have earnings of at least $500 a month those families actually seem to be more likely to be doubling up during the time that the CTC was expanded. You know, again, it’s really hard for us to know exactly what’s going on there. One hypothesis we had was that, you know, one thing we can’t tell is whether they’re doubling up in someone else’s home or whether they’re inviting others to live in the home with them. Our hypothesis was that this might be driven by more of the latter, that it might be an indication that these families with more resources are able to be, you know, more financially stable and able to take in other members of their network that might also be experiencing difficulties affording rent. We know that low-income families are embedded in networks of other low income folks and so it’s not surprising, perhaps, that if you get resources to one member of that community, that those resources are shared in the broader community.

Siers-Poisson [00:18:44] And finally, did you observe any other changes in conditions, for instance, household composition or that overall household size that you mentioned earlier?

Michelmore [00:18:53] So we did see a decrease in household size as well that we think is again, we think it’s partly driven by those partners that were no longer living there, but the magnitude of the effect was larger than that. So it could be an indicator for instance, it could be a partner that had their own children that were living in the household. So it might be a suggestion that it could be the partner leaving the household and they may have brought any children that they had with them out of the household as well. But unfortunately, we don’t know each person living in the household. We know how many people live there and we know kind of the set of people living in the household, but beyond that, we don’t. It’s a little tricky for us to get exactly at who was moving out. But as I mentioned before, we don’t find an overall effect on decreasing and doubling up. We don’t think it’s kind of either an unrelated person or maybe the extended family moving out. We think that’s in part driven by these partners.

Siers-Poisson [00:19:40] We’ve been talking about the data set of these SNAP users who are sharing information about the child tax credit in their households and their housing stability. I’d like to talk about whether you saw any significant differences, especially by race or ethnicity, and also those average income levels that did come into play a little bit, you were saying, with the doubling up.

Michelmore [00:20:05] Yeah, we did see some interesting differences in the pattern, again, it was somewhat unexpected. We often found significant effects on the housing affordability outcomes for one subgroup, and then we’d find effects on the living arrangements for a different subgroup. So I’ll start with the income ones first those might be a little bit more intuitive. For instance, we found that the lower income folks in our dataset, so these are folks that have earnings less than $500 a month or in some cases no earnings at all. We really found that the credit was beneficial in reducing their back owed rent, whereas we saw more of the effect on living arrangement changes among those higher income folks. We saw that the increase in doubling up was happening on the higher income folks, but also the reduction in the likelihood of living with a partner that was primarily driven by those higher income folks. I think this was an unexpected finding and we got a little bit of, question when we’re going through the review process about this. And one metric that we thought was kind of helpful that we determined is that when we looked at the income of those households that had a partner versus those that didn’t have a partner, the households that had a partner had about $500 a month higher income than those that didn’t. And so maybe not coincidentally, we found in our sample that the average benefit level for the CTC was about $500. We were thinking that, you know, again, these are low-income households so $500 a month that they’re getting from the CTC is not an insignificant share of their household budget. A monthly credit of $500 could very well replace the income of another person living in the household. And that was something that we thought was somewhat interesting. In terms of race and ethnicity, we saw that again these the changes in living arrangements were primarily driven by white and Hispanic families, we didn’t see much happening for black families. In part we thought that might be because black families were less likely to be living with a partner in the first place, living with a partner was much more common among white and Hispanic families in our survey. But where we saw bigger effects for black families was again on that affordability measure, we saw that black families were much more likely to report reductions in back owed rent, and they were also less likely to report that they needed to move because of affordability reasons while the CTC was in place.

Siers-Poisson [00:22:13] In reading your paper, I wondered about how much of an impact there might have been that the expansion of the CTC was temporary, and I mean that beyond the obvious stoppage of the payments. Do you think that affected decision making by these parents?

Michelmore [00:22:28] Yeah, it’s a good question and one that we’ve heard a lot. Another way that others have phrased this is “why would anybody make such, you know, life changing decisions about their living arrangements for something that only lasted six months”. We have a couple of responses to that. First, it’s not clear when people knew that the credit was temporary, many people may have on that first payment not counted on that continuing and didn’t ever expect it to be a consistent benefit. But I will point out that there were many of us who do research in this field who were unsure whether the credit was going to be extended even up until the last few weeks of 2021. There was a lot of uncertainty about whether it was going to be renewed and not to say that everyone is following the news or the inner workings of Congress so intimately, but I think that there just wasn’t a clear idea, a clear sense of whether the credit would be made permanent. And so for that, we don’t really know how families were viewing it. We don’t know if when they were receiving that benefit, they were thinking this is something I’m going to get forever, or whether they were thinking this is a temporary thing. If we do think that they thought it was temporary, if anything, we would have expected that a permanent policy might show even larger effects because it’s a stable source of income that families could count on. But on the other hand, I think also thinking about the context of who’s in our survey and these are low-income folks who are largely making about $500 a month where the CTC average benefit was about that size as well. This is quite a large amount of money for the individuals in our sample and so we think that even if the credit were temporary, the amount of money that they were getting on a monthly basis would have been enough to make kind of big changes, perhaps for these families just because their baseline income was so low.

Siers-Poisson [00:24:08] Kathy, we’ve covered a lot in this conversation, and you and your coauthors analyzed a lot of different outcomes. Can you give us a bottom-line summary of your findings?

Michelmore [00:24:17] I think overall, we found that the expanded Child Tax Credit in 2021 led to improvements in housing and living arrangements. Primarily, if we even look at the metrics of housing affordability, we see clear evidence that the families that were eligible for larger credits were less likely to owe back owed rent. Their debt was lowered, and they were less likely to report that they would need to move because of concerns about affordability. The changing of living arrangements of course, it’s harder for us to make kind of value judgments about what’s going on there, but we generally think that if we’re providing families with more resources, we’re trusting them to make the decisions about what’s best for them. And so overall, we think that our findings are suggestive that the credit was leading to improvements in both housing affordability and housing stability.

Siers-Poisson [00:25:01] What do you see as potential policy and practice implications of this research and your findings?

Michelmore [00:25:07] I think there’s a number of things here and I’ll kind of tie this to some research that I think we need more evidence on. But one thing I think that’s really unique about this policy was that it was distributed monthly. And so, just hearing from qualitative data and we asked some families some questions about how they felt about the credit, it was very clear that this the monthly distribution was hugely beneficial for families in terms of providing some stability and having some income that they could count on from month to month. So I think that’s a big open question for us. It’s the first time we’ve really done something on that kind of scale of changing an annual lump sum tax credit to a monthly credit. We clearly need more evidence on the tradeoffs between doing something on a month-to-month basis versus a lump sum payment, I think from qualitative data. Some folks really like getting a lump sum benefit, we know that from some work that IRP affiliates have done on the Earned Income Tax credit, for instance, that people like getting those large lump sums. But we also, on the flip side, heard lots of evidence from people that they really like those monthly benefits. So I think it’s just something to consider, there might be an option going forward to let families opt into or out of one of these types of arrangements. I think that the nice thing and I don’t know how thoughtful this was in the in the legislative process, but I think one nice artifact of the structure was that families got a little bit of both worlds. That they got half of the benefit in six monthly increments and then they could still get a lump sum at tax time. That was an interesting aspect of this policy.

Siers-Poisson [00:26:37] And as we wrap up, what further research would you like to do or see done on this topic?

Michelmore [00:26:43] There’s always room for more research. One thing that I just alluded to about this difference between the monthly and lump sum nature of the credit, I think would be really interesting to push a little bit more on. In the broader research community, there’s this debate about whether it’s better to give families kind of a regular payment or a lump sum payment even, you know, in the context of benefits that are distributed monthly like food stamps. There’s discussion of how families run out of those benefits by the end of the month so that now families in this demographic, this low-income population, have a difficult time making ends meet. And so any degree of lumpiness in distribution I think is challenging and how to strike the right balance of how frequently to distribute those credits is a big challenge. But I think more broadly than that, at the end of the day, this is a pretty recent expansion that happened in 2021. And there was a flood of evidence on some of the markers that we could get immediate outcomes on like food insecurity and housing. But we don’t really know the impact on kids themselves, I think all of the things that we have learned about have positive implications for children: we think reducing food insecurity is good for kids, improving housing stability is good for kids. But the end of the day, we don’t yet know a whole lot about the benefits to the children themselves outside of a couple few small studies. So I’m really interested to see going forward what evidence comes to bear on how this expansion improved or affected the outcomes of children across health, education, poverty and so forth.

Siers-Poisson [00:28:06] Kathy, thank you so much for taking the time to discuss your research with us. It’s really interesting and important.

Michelmore [00:28:12] Thank you. It’s been such a pleasure. Thanks for having me.

Siers-Poisson [00:28:15] Thanks so much to Dr. Katherine Michelmore for talking with us about her research on how the expanded child tax credit affected housing stability and affordability for low-income families. You can find a link to the paper in the program notes 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. But 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.