Housing scarcity is linked to higher rents and house prices, but it’s rarely connected to the cost and reach of safety net programs — and it should be. Kevin Corinth joins to share his research on how increasing housing production in supply-constrained cities can help the government serve many more households with rent assistance.
Housing scarcity is linked to higher rents and house prices, but it’s rarely connected to the cost and reach of safety net programs — and it should be. Kevin Corinth joins to share his research on how increasing housing production in supply-constrained cities can help the government serve many more households with rent assistance.
Hello, this is the UCLA housing voice podcast. And I'm your host, Shane Phillips. This episode, we're joined by Kevin Corinth, who's sharing his research on the connection between housing production, and the cost and reach of safety net programs, and specifically the Housing Choice Voucher rent assistance program. In a way, this is a very straightforward episode and topic. When rents go up, we have to spend more money on rent assistance for each household, which means we either increase our budgets or help fewer people, probably more often, the latter. The flip side to that is if we lower rents than we can help more people. We know there's a strong link between housing supply and the pace at which rents rise. So if we can estimate even roughly how much a given rate of housing production slows rent growth, then we can also get a sense for how many more people we could serve with our existing rent assistance budgets, or with larger budgets for that matter. I say this is straightforward, but it's a connection that policymakers rarely seem to make explicit. So we wanted to have Kevin on to do just that. The housing boys Podcast is a production of the UCLA Louis Center for Regional Policy Studies. With production support from Claudia Bustamante, Jason Sutedja, and Divine Muttoni. Send your questions and comments to shanephillips@ucla.edu, and send a link to the show to a friend or colleague who might find it interesting. We really appreciate it. With that, let's get to our conversation with Kevin Corinth. Kevin Corinth is a senior fellow at the American Enterprise Institute and Deputy Director of AIS center on opportunity and social mobility. And he's here to share his research on the potential cost savings to federal rent assistance programs that could be realized from increased housing production in our most restrictive cities. This is right at the nexus of a lot of our interests here at the Lewis Center. And it even has a Los Angeles angle. So Kevin, thank you for joining us, and welcome to the Housing Voice Podcast. Thanks. Great to be on. And my co host today is Mike lens. Hey, Mike.
Michael Lens:Hello, Shane. Welcome, Kevin. I'm excited to talk about this paper today.
Shane Phillips:So we always ask our guests to give us a tour of a city or a neighborhood that they know. Well. Kevin, I think you are going to talk about Capitol Hill in DC for us today where you're taking us.
Kevin Corinth:Yes, and I hope that's not a repeat for your listeners, but by default. And so I've been living here for the past nine years in the Capitol Hill neighborhood of of Washington. And it's the only one I can really remember that'll do. So I've been living in. I've been living on Capitol Hill, with my family with young kids so so that'll shade some of the perspectives I have. But it's really a we have a scooting culture in DC in Capitol Hill, so in kids that have kids riding their bikes like I did when I was growing up in the in the Midwest, here, it's scooters, I think, because the sidewalks are a bit narrower, and the roads are a little bit less safe for kids on on bikes. But there's just a lot of playgrounds, I mean, there's three playgrounds literally within one block of of our house, our kids walk half a block to school, it's really a nice place for families. If you're going to come here on a tour with young kids, I would suggest bringing scooters, and you can scoot to the Capitol building, very lenient, at least nowadays, after being shut down for a while in terms of being able to scoot around in front of the Capitol Hill in front of the Capitol building. You can go get some pizza and have a nice picnic on the Supreme Court lawn, which I still think is one of the neatest things one can do. And if you get tired of what Capitol Hill and DC has to offer, you can take a quick walk over to Union Station where you can catch a commuter train to Baltimore, we have some family favorites like the Maryland Science Center, take the Amtrak a couple hours to Philadelphia or even four hours to to New York City. Again, growing up in Minnesota. The idea of having major cities with such quick access was a foreign concept to me. So I find it really neat that even though I have an amazing neighborhood to be in, it's very easy to get some of these other great, great cities.
Michael Lens:We are 67% Minnesotans on this today. I had no idea.
Shane Phillips:I was gonna say that. As a lifelong West Coast resident, the idea of having cities very close to you like that. It's also a little bit foreign and but very appealing.
Michael Lens:Yeah, it's six hours from you know, where I grew up in St. Paul to Chicago. That's not it sounds silly. Or four hours I guess. My scooter story very recent. i My youngest boy is 12. And he had a couple kids over and they like to disappear and play Pokemon Go on their, on their iPhones, and all their kids like is all this 40 do that. And at some point, I got a notification that my son had paid bird scooters. I was like, oh, like go down the street. And there are three boys just like flying down the street on scooters. I was like, oh, and then of course, one of them's like, Oh, my mom lets me do this all the time. And you know, then I'm, you know, the bad guy. Like caan't be that Dad. Yeah, like, but none of you guys asked me to do this. You could suddenly be on motorized.
Shane Phillips:And you paid for all of them? Probably.
Michael Lens:I don't think so. Because they're on phones with you know, accounts. It's it's wild time what a world. So Kevin, you know, your kids will be will be well trained in electric scooter, it will be it will be old hat for them be easy.
Kevin Corinth:I hope so. Yeah, for now at age four, seven, and nine. And I can't imagine putting them on those electric scooters. So it is a purely old school manual scooters for now.
Michael Lens:Yes, keep it that way as long as you can.
Shane Phillips:So the article that we're talking about today was published in July in the journal of urban economics with your co author, Amelia Irvine. And it's titled The effect of relaxing housing market regulations on federal rent assistance programs. Kevin and Amelia, you start off this study by making note of the fact that when rents go up, the cost of rent assistance programs, like Housing Choice, vouchers also increases, which means we either spend more to help the same number of people or we help fewer people, despite spending the same. This happens because housing vouchers subsidized rents on the private market, the government does not set voucher rents directly or the rents that the units are rented for. The next point in the article is that it's pretty straightforward, and is that rents tend to rise more slowly in places that build more housing on a per capita basis. This is sometimes disputed in policy circles, but the empirical research is pretty unequivocal. All else equal more homes equals slower price and rent growth. The next step in the study is estimating how much slower rents would grow under different rates of homebuilding. And that is really the challenge and the heart of this analysis. Once you have those figures, though, it is pretty easy to get a rough idea of how much further our rent assistance dollars could go under various simulations. To tease some headline findings, Kevin, you estimate that if over a decade, the Los Angeles metro area produce new housing at the same rate as the 90th percentile metro area, meaning it would be in the top 10% For per capita, homebuilding nationwide, rents here would be about 18% lower, and the federal government would save $353 million on its rent assistance program. Or it could actually continue spending the same amount and give vouchers to about 24% More very low income households. Those are big numbers, they will justifiably, I think, raise some eyebrows, and we're going to talk about them.
Michael Lens:But you know, raise some hopes,
Shane Phillips:Or some hopes, yes. But I looked up some Zillow data from for the LA Metro area, rents over the last eight years increased by 57%. This is not inflation adjusted, but most of that time, there wasn't a whole lot of inflation. So you know, maybe an 18% reduction over a decade isn't as far fetched as it might sound to some of us who have just dealt with forever increasing prices. So that is my introduction to this. Kevin, let's hear yours here. Tell us what interested you about this nexus between housing production, private market rents and the cost of federal rent assistance programs. I feel like you don't often see these things linked together. And that's part of why we were really interested in having you on, you've usually got the market and housing production and rents on one end or one area of academia. And you've got the government and rent assistance and other subsidy programs in some other areas. It's a totally different kind of analysis. And they're treated as pretty distinct subjects most of the time. No,
Kevin Corinth:I agree completely. I mean, my research has actually traditionally focused more on the government assistance side of things, especially for homelessness and rental assistance programs, and then how these programs fit within the broader safety net in the United States. But as more research has come out, uncovering the really wide swath of negative consequences of overly restrictive housing regulations, it's hard just not to see the connection to everything else. To me, the most pernicious effects are less access to opportunity. So fewer high paying jobs for adults, lower quality schools for kids, just less healthy. munities in general, for families, it's also meant more segregation of people across socio economic lines. And as our paper points out, it means a safety net that works less efficiently. So at the end of the day, I think these topics should be merged together, you know, advocates for more housing subsidies and advocates for more housing supply think share many of the same important values, including concerns about family's ability to to afford housing, opportunities for kids and less skilled workers to move up. And then most fundamentally inclusion in society. There are certainly some disagreements about the ways to achieve these goals, the policy, but I think there should be areas of agreement as well. And I hope that our paper is just one example that shines a light on some of these interconnections. And when we recognize some of these shared values, there are ways to achieve the same underlying goals with with better policy.
Shane Phillips:And to set the stage a little bit here, just on the program's themselves, can you give us a quick overview of the scale of the various federal rent assistance programs? I think at HUD, the US Department of Housing and Urban Development, you have that really the big two are housing choice vouchers, where you have both tenant based vouchers that follow tenants, and they can take them from unit to unit and project based vouchers that are tied to a specific unit or building. And then you have public housing, which has been around quite a bit longer, but has been kind of dwindling as a source of housing supply. Then you also have the Low Income Housing Tax Credit, which is not a HUD program, but it does fund the construction and rehabilitation of housing that's then rented to low income households at below market prices. And, you know, historically, this has produced about 100,000 units a year, which is pretty significant. How much do we spend on these various programs? And how many people do they help? And are there any other federal housing assistance programs, especially on the rental side, that you think are worth mentioning in the same breath as these others?
Kevin Corinth:Sure, I'd be happy to provide that overall context. And I think it helps to put these housing programs in the context of other safety net programs that we have. So you listed the main rental assistance programs, the main HUD ones Housing Choice, vouchers, project based assistance, and public housing. In addition to Lai tech, overall, we spend about $50 billion a year on those programs, which is a lot when you compare that to other safety net programs. For example, we spend only about $30 billion on cash welfare. And at least before the pandemic, we spent about $60 billion on on snap the Supplemental Nutrition Assistance Program, which was formerly known as food stamps. So the key difference, though, between snap in housing assistance, is that snap provides benefits to anyone who is eligible. Housing Assistance is different than that it's not an entitlement. So even if you qualify based on your income, you may not actually get the assistance. In fact, only about one in four eligible families actually receives housing assistance, right. So And one reason for that is that rent is expensive, especially in supply constrained areas. So the average cost to serve a family in Los Angeles for one month is about $1,500, or around $18,000 for a full year. By contrast, the maximum annual SNAP benefit for a family of three is about$9,000. So about half of that. So we're serving a lot less people with housing assistance, but we're providing a much higher average benefit for those who get it. Now in terms of specific programs. Within the housing assistance realm, the Housing Choice, vouchers are the biggest. So they voucher serve about 2.2 million families each year at a cost of about $23 billion. So vouchers are vouchers that the tenants can take out into the market and lease up with a private landlord. Then we move into project based assistance where these are the government directly funding specific landlords and specific units, who are then renting out those units to recipient families. And project based assistance serves about 1.2 million families at a cost of $12 million a year. And finally, we have public housing, which is housing owned and operated by the government that serves right around 1 million families a year at a cost of$7.4 billion basically to operate and maintain those units, although doesn't incur
Shane Phillips:It doesn't take into account the cost to build them and also the probably not the backlog on renovations and maintenance and stuff as well.
Kevin Corinth:Exactly. And so those are the main ones. I mean, there's lots of smaller actually lots and lots of smaller programs, which actually some co authors and I have been previously tried to start to quantify but there's a US to partment of agricultural rural rental subsidy program, there's homeless assistance programs, which are important. But that's more like $5 billion a year. So I think we've we've characterized the main HUD ones in livestock are really where the major money is.
Shane Phillips:And light tech is also in the realm of $20 billion a year. Is that about right? It's a tax expenditure. So it's funded in a very different way than just like a line item in the federal budget. But is that roughly what it ends up costing effective?
Kevin Corinth:Exactly in terms of the tax revenue foregone, because these are tax credits that are offered through a roundabout way to developers, in return for a promise to lease out 30% of units at a certain rate to certain types of tenants? But yes, that's approximately the cost of in terms of tax revenue foregone, which is a real cost.
Shane Phillips:Yeah, yeah. So when I introduced you, I gave a very simplistic overview of your model for this problem. And just to summarize that, again, it's essentially, when rents go up rental assistance programs cost more or serve fewer people. At the same time, building more housing keeps rents from growing up as quickly. So if we can estimate how much different rates of homebuilding will lower rents, different increased rates of home building will lower rents, we can also estimate the benefit to rent assistance programs and their recipients. Before we get into the methodological details here, is there anything else we should know about your approach at this more conceptual level? How you approach this question?
Kevin Corinth:I think you summed it up very well, earlier. I mean, the key to all of this is that the government wants each family to pay only 30% of its income on rent. So that means that when the market rent goes up, the government has to absorb the entire cost increase. Since the family's contribution won't change, it's fixed at that 30% amount, this can lead to very big effects on rent increases on how many families are actually served. So maybe just a quick example, to help illustrate this and why the effects can be pretty big. So so say, every family can afford to contribute about $900 per month toward their rent. If the market rent were $1,000, then it cost the government just $100 per month to serve each family. So now say that the market rent rises from $1,000 to $1,100 per month, which is just a 10% increase in market rents. Well, now the government has to pay actually $200 per month for every family, the $1,100 market rent minus the $900 contribution that the family makes. So that's doubling the cost for the government to serve these families. And so the government can serve only half as many families due to just a 10% increase in rents. So this simple logic is really the basis of the model. And again, it tells you why we simulate some pretty substantial effects that people should be hopeful about rather than too skeptical. There are some other factors that we include in the model, such as that the fact that wages might adjust as more households join an area in which could increase labor supply and reduce, reduce wages and offset some of these government costs. We do account for other things. But those factors don't really change the underlying story, that regulations that control housing supply are a really big deal for rental assistance programs.
Shane Phillips:And I mentioned that the assertion that more supply lowers rents is a little bit contested, again, not so much in the literature. So are there just a few, you know, studies we could maybe point folks toward, we can include these in the show notes, Mike, feel free to chime in here too?
Kevin Corinth:Sure. I mean, one of the more recent ones is by Raven Malloy and some some co authors, they actually looked at the effect of regulatory stringency on market rents. And admittedly, they find a smaller effect of regulations on market rents than you find for home prices in terms of purchasing new homes, but nonetheless, a relatively important effect. That was pretty substantial, at least for those who had the areas with the most stringent regulations, and actually, some effects that are in line with what we we simulate in our model.
Michael Lens:Yeah. And I think, you know, we've had discussions, certainly in on this podcast about about the effects of land use regulations on prices. And no, no, never, no, never discuss that. And, you know, and you know, we've couple of star on our work on that confirming some of these connections that Kevin is summarizing. And you know, I think all say rather than rattling off a list studies is like, where most listeners might get hung up on is the effects on a neighborhood versus the effects on like an entire regional housing market, something as large as Los Angeles or Los Angeles. county or Southern California, you know commuting zone or something like at the neighborhood level we're already getting, you know, I think plenty of studies and evidence that like even at the neighborhood level, more housing supply does help more than it hinders housing affordability. at the regional level, this has long been fairly uncontroversial, you know, to the extent to which you believe good economic research, these connections are pretty clear that you need more housing supply to avoid out of control increases in either rent or housing prices. And, you know, over the latter half of the 20th century, in the early 21st, we have gone in the other direction in terms of making it more challenging to build in more places.
Shane Phillips:Alright, so let me try to summarize the research question here. Let's just go with La only hear even though you did look at some other metro areas, if the LA Metro built housing at a rate equal to the 90th percentile metro area for 10 years, how much lower would rents be at the end of that 10 year period? Is that generally correct?
Kevin Corinth:Yeah, exactly. And we're not saying that rents are going to go down from where they are now. It's not saying that, Oh, if rents are, say,$10,000 a month today, they'll be actually less than $2,000.10 years from now. We're just saying that relative to what they otherwise would have been 10 years from now, they'll actually be about 18%. Lower.
Shane Phillips:Okay. Okay. So yeah, that is a very big difference. And I think really important to clarify, because, yeah, to say that if we had started this increased homebuilding, 10 years ago, they would be 18%. lower today than they were 10 years ago. That is a very bold statement that I think, you know, we haven't really seen many places, even those that do build at that level, their prices have not gone down, but their prices have climbed more slowly than they have in places like Los Angeles. So good to get that clarification out of the way. So now let's get into how you actually estimate the impact of increased homebuilding on rents. And again, we can use your analysis of Los Angeles as an example here. As I said, under your preferred model, and we'll talk about what that was, you find that rents are 18.1%, lower at the end of a decade of sort of supercharged homebuilding. You also ran these numbers for 10 other metro areas that have similarly restrictive housing development regulations, and limited production. But LA is pretty representative of the results overall, other than being a larger city than most of these other places. Can you walk us through that process, starting with how you decided the amount of housing production to model in the various scenarios in your study, and then how you translated those production rates into changes in rent? I think there are a lot of assumptions that go into this because you're you're modeling a counterfactual. It's a world that could have been, but never was. And so I think it's really important that we're clear about what those assumptions are, and that we try to stress test them a bit.
Kevin Corinth:Sure, I first want to completely agree with you about the caveats of an exercise like this that simulates the effects of a hypothetical policy, which does in fact, require a lot of assumptions. I've increasingly come to the mind that we can't constrain ourselves to ask only about the effects of policies that actually took place, and for which we have a way to cleanly identify their effects. policymakers don't have the luxury of of only considering policies with mounds of directly applicable evidence from least recently enacted policies. They also need estimates of how future policies would affect outcomes that they care about. Of course, it's important to rely on assumptions that are backed by the academic literature as much as possible, and to show how results would change under different assumptions. But I would argue that even when we can't provide an exact estimate, providing an expected range of magnitudes is still valuable, and really just necessary for rational policymaking decisions. So with that off my chest, I would be happy to get into some of the crucial assumptions, which I think are important to interrogate a little bit. So let's start, as you mentioned, with the level of housing production that we assume. So as you noted, for our baseline estimates, we assume that Los Angeles could build as many new units per year on a percentage basis as the 90th percentile metropolitan area. That means that instead of expanding its housing stock by about 0.6% per year, which is the past decade trend, it would expand by 1.6% per year. Now, Is that realistic? I mean, I guess no, in the sense that I don't think Los Angeles is actually going to do that. But would that building rate be plausible if it pursued aggressive reform of Building and land use regulations that put it on par with less supply constrained areas. And I think the answer to that question is yes. So by definition 10% of metro areas built at a faster rate than the 1.6% rate that we assumed for Los Angeles. So places that built faster than that 1.6% rates include places like Austin, Texas, Orlando, and Salt Lake City. Now it's, it's true that it may be easier to build in some of those areas than Los Angeles, because they might have more undeveloped land. But some of those areas like salt lake city also face severe constraints on buildings, such as big mountains and lakes. So of course, many regulations restrict how densely housing can be built in already developed land, as well. And so undeveloped land, is it necessary for substantial expansions of the housing stock?
Shane Phillips:Yeah, and I, you know, I think it's probably worth noting here that in the Los Angeles metro area, there are about 4.7 million homes, and 13 or 14 million people. And so a 1.6% growth rate would be about 75,000 units a year, which, frankly, you know, it's a lot more than we're building, the whole state has only been building about 100,000 units or less per year for quite a while. So this would be a big increase from where we're at. But you know, 75,000, into 4.7 million. It's not stupendous. And the city of LA Actually, according to their revised housing element is supposed to plan for about 56,000 units per year over an eight year period, just in the city, a city of 4 million people in a region of 13 or 14 million. So if anything, we are actually calling for in our own, you know, laws in our housing element that has been adopted by the city and kind of pushed and authorized by state and regional governments, that we should be building actually a lot more than 1.6% per year, at least for the for the next decade or so. Okay, so that's the homebuilding rate. I will say you did also look at what the impacts would be at the 75th and 95th percentile, is that correct? That's right. And so the next really important consideration here is the price elasticity of demand. So you chose a figure of point seven, five. And again, you considered a range of values here, but tell us what the price elasticity of demand represents, in the least economist terms possible. And again, kind of justify that point seven, five figure for us.
Kevin Corinth:All right, I'm gonna try my best here. I am a University of Chicago trained economist. So I see everything and supply and demand, I will try to get off my econ,
Shane Phillips:I will say I have looked at a lot of supply and demand curves in my life. And it is still not intuitive to me, I don't know that it ever will be the way my brain works
Michael Lens:Four micro courses before any of it made sense, but good luck, Kevin.
Kevin Corinth:So let me try here. So so as you mentioned, first, you need to know how much more we're going to build. And we're saying it's this 1.6% rate of building, which would increase the housing stock over the course of 10 years. So now we have to translate that increased amount of building into how much rents will fall relative to that counterfactual baseline. So to do so, we need to know how prices and quantities are linked together. And to do so we rely on studies that have looked at how the demand for housing changes when price has changed. So at a very basic level, if demand for housing is not very sensitive to price, because people are just going to live in Los Angeles, no matter how much it costs, them building, a lot of new housing rule would reduce the price by a lot. There's just not a lot of demand for the new housing, so prices have to fall a lot to fill up all those homes. On the other hand, if demand for housing is very sensitive to price, because people will move to LA only if housing gets a little bit cheaper, then building a lot of new housing will not reduce the price as much. There's just a lot of new people who will buy or lease up the new housing at just a slightly lower price, because they are so price sensitive. Now our baseline assumption is that consumers are somewhere in between these two extreme scenarios. Basically, we assume that for every 10% Decrease in rent, there's a 7.5% increase in families who are willing to rent homes in Los Angeles. So well, estimates vary a lot across studies in terms of what this price elasticity is, but that 7.5% or 10% or 5%. Our reading of the literature is that this is a central estimate. If you think that consumers are actually less responsive to changes in rent than we state, then rents would fall by even more than our baseline simulations suggests. If you think that they're more responsive than rents would fall by Less than we do show in the paper, how results would change with different elasticities. And it does change substantially. But at the end of the day, you do get substantial cost savings from deregulation and more buildings, sort of, regardless of which elasticity you choose.
Shane Phillips:Mike is there. Like, I feel like we could use more clarification on this, I don't know exactly the right direction to go. I think part of it is like, I'm much more familiar with the price elasticity of supply you are. And so thinking about this, in terms of demand is is a is a challenge for me.
Michael Lens:One reason why the demand response is interesting to me is because I'm wondering how much you're accounting for population growth from outside Los Angeles, or household formation from within Los Angeles, that might respond to this decrease in prices, is that related to this discussion of elasticity that you gave us right now, in which you're the two kind of extremes that you've mapped out is one way I would put that as like, there are 100 people with offers from from UCLA and USC and an aerospace industry that is located in Los Angeles 100 people outside, and they're only going to move in, if the, you know, cost of a two bedroom apartment goes from 2000 to 1500, or something. But then in Scenario B, those 100 people take the job and come in, and they don't even care what the rent is because they have to move to Los Angeles to have to have this job. So is is part of that demand kind of this, this story of this elasticity of demand is part of that this story of people moving in and forming households and how responsive they are to those prices.
Kevin Corinth:It is and I think that was a very much better way of putting it than I than I did. So So think of Los Angeles housing as any other good like bananas, right? Like the potential consumers of Los Angeles housing. It's not just people who live in Los Angeles, who may form smaller households to smaller families, or, or people who may buy second homes, but it's really everybody in the United States. And so as the price of Los Angeles housing goes down, that means there's going to be more demand for Los Angeles housing from not only people already in within Los Angeles, but also people from outside of Los Angeles. And if there's a lot of people who are really willing to come in when the price falls just a little bit, then prices aren't going to fall very much when you expand housing supply or why. But if there are a lot of people in Los Angeles there, and they're going to be there, no matter what, as you said, then it's really hard to attract more people in and so you're going to have to see a very big decline in print supply
Michael Lens:is going to have a big effect, right, that supply is gonna have a big effect, because there's not all these super price responsive people who are going to jump in and then, you know, kind of create this demand demand versus supply problem all over again, is one way to do
Kevin Corinth:Yeah, exactly. Right.
Shane Phillips:Okay, that yeah, that is that is helpful. And, you know, I think it's also just worth saying here. Some people might hear this and think, well, a lot of people do want to live in LA, it's got great weather, and great, you know, amenities and tons of jobs and everything. And so a lot of people, you know, really do want to live here. And I think people might hear that and think well, then, you know, what's the point of building a lot of housing if people are just going to come in and move here, and it's not really going to have an effect on prices. And I think if the proposal were actually just like, only Los Angeles should build a lot of housing, and everyone else should just keep doing what they're doing, that would be bad. I think, to some extent that prediction might come true. I think even if that were the case, it's a little overstated, because not everyone wants to be in Los Angeles, just talk to people around the country. But like that point aside, what we would really hope to see is all the places that people really want to live building a lot. And then people really have many choices. And so no single location is going to draw them all in because yeah, there's a whole mix of things people are looking for beyond the price. Even if they're just kind of limiting themselves to the high cost coastal city. There's a lot of decisions to make about where you want to live for for jobs and you know, the culture and a million other things.
Michael Lens:I want to I want to give Kevin a chance to explain to you know, answer the questions we keep throwing at him. But, you know, I think that's a good a very important point chain, right is like, Los Angeles is not in a vacuum. It's in a you know, massive country of 300 million people and lots of metropolitan areas and there's competition for jobs and housing and people that and workers that kind of involve all of these other places. But when thinking about like, Los Angeles is intrinsic, you know, attraction, which we all hate the traffic, we all love the weather. There's obviously something that brings a lot of people here, because there are a lot of people here, right? Like there is a high baseline attraction to Los Angeles. But like, I think what Kevin would say, I don't know why I'm answering this question for him, is like, what really matters is at the margins, right? Like, there's some people who kinda want to LA and there's some people that would move to LA if it were just cheaper. And so these are the this is the subset of people that we're really talking about. And correct me if I'm wrong, Kevin, that the numbers are bigger in the aggregate in Los Angeles versus Kansas City. But like the bait, you know, the kind of those marginal decisions are similar everywhere, to some extent, is that right?
Kevin Corinth:Yeah, exactly. You do want to look at the margin. And and I think your point is exactly right about sort of, if everybody does this, if everyone sort of relaxes their supply constraints, you're going to see a much bigger decrease in prices everywhere. So even if it's just Los Angeles builds more, and people will move in, you'll still see some decrease in prices, but people will move in and which will mitigate that to some extent. But if every other large metro area, like San Francisco and New York City and Washington DC, also engage in aggressive deregulation, then people now have lots of options. It's not that just the price of Fuji apples went down, but red delicious apples went down as well.
Shane Phillips:And so I think the last assumption here that I want to ask a question about is this cross price elasticity? And, you know, if I'm understanding this one correctly, and I it's entirely possible, I'm not, if this is probably the part I understood the least, I'll just say you assigned a or assumed a cross price elasticity of zero. I took that to mean that you're assuming owner occupied housing, any owner occupant housing that is built will have no impact on rents? Is that correct?
Kevin Corinth:Yeah, that's right. And that sort of makes a pretty conservative assumption. Yeah. Yeah. So So you know, if Los Angeles were to build a lot more, they're going to build some more rental housing, and they're going to build some more owner occupied housing. We're basically saying, forget about all the owner occupied housing, if only the renter occupied housing is what affects rents. This is what you'll see, you might actually see something, something else you might see that when you build more owner occupied housing, people sort of move from the renter market and say, now I can afford to buy a home. And there they may move from the rent rental market into the owner occupied market. Yeah,
Shane Phillips:yeah, that was really what I was getting. Because and I think that's probably likely, it's maybe hard to model and safe to assume that the zero impact, but it seems like some impact is likely. And just so the listeners know, in the study, you assume that about a third of this production, this 1.6% annual growth is in owner occupied housing. And so you're just kind of dismissing the impact of that entirely just to be safe. And so there are, you know, assumptions made here like 90th, percentile, homebuilding point seven 5%, elasticity of demand, these things can be debated up or down, but you can't go any lower than the owner occupied housing having zero impact. So you can only kind of go up on the actual effect on rents from what you're assuming. Right, right. Okay. So with all of that explained, tell us how much you estimate this increased production would lower rents in LA, and what impact that would have on federal rent assistance programs, both in terms of or kind of, either in terms of money saved, or the number of households assisted? So let's start with LA. And maybe if you can just share the the total amount for all 11 metro areas that you looked at as well for all those figures? Sure. So as each metro area, just all of them
Kevin Corinth:Will do. So I think, as we we mentioned, you know, we estimate that market rents would fall by 18% in Los Angeles after 10 years of increased building again, relative to the baseline rent increase level. So given that decrease in rent, it's relatively straightforward to simulate how much the government would save since as I mentioned before, all of the reduction in market rent for subsidized households accrues to the government. There are a couple of caveats which I won't go into, including that wages can adjust that people may accrues Some of the savings but for the most part, then that's That's right. So in terms of results, so the effect on federal rental assistance programs for Los Angeles would be substantial. We estimate an annual savings of$353 million, which is 19% of current total spending on vouchers and project based assistance in the Los Angeles metropolitan area. If those funds were reinvested in serving additional families, that would mean 24% More families could be served, I should be clear that this would come nowhere close to filling the unmet need for rental assistance in Los Angeles, and the majority of eligible families would continue to be turned away. But nonetheless, 24% is an important increase and means 1000s More families would get access to assistance who weren't otherwise getting it in terms of all areas. So Los Angeles is not the only place in the country that has substantial supply constraints on building new housing, a paper by Ed Glaeser and Joe Jurco, from 2018, classified about 11 metropolitan areas that were severely supply constrained, where prices were really driven up a lot more than they should have been, because of how hard it is to build new housing and the restrictions on land use. So if you take all 11 of these areas, and that includes places like San Francisco, and Seattle, and Denver, and New York City, Washington, DC, and Boston, at all see the paper, total savings would amount to $1.8 billion annually, which could be used to serve 5% More families. So I think you can be underwhelmed by$1.8 billion, as we mentioned before, we spend $50 billion annually on rental assistance programs. So $1.8 billion, is not enormous. But remember that we're only talking about 11, metropolitan areas that are going to see any effect at all, because these are the only places with the most severe supply constraints. I'd also suggest that, you know, that's important money. So HUD spends about $5 billion a year on homeless assistance. So, so $1.8 billion is important. And it would be important to the families who who would receive it. But again, we're not talking here about solving the entire shortage of rental assistance. shortfall. Yeah,
Shane Phillips:Well, then what are we even doing here? Could you say a bit about the the sensitivity of these results to different assumptions in the model? Because, you know, depending on what these elasticities and homebuilding rates are, there's a pretty wide variation, right?
Kevin Corinth:Sure. I mean, I would say that, under all of the pretty reasonable assumption, the most reasonable assumptions on things like building rights and these price elasticity ease and cross price elasticities, I think we get important effects. So again, a baseline we're talking about $353 million of savings for Los Angeles, it's something like around say $200 million. In Los Angeles, if we use our most conservative assumptions, which would still mean, say more than a 10% increase in family served. So So that's, that's real money, and it's a real increase in family served. On the other end, I think you could certainly argue for sort of more liberal assumptions in terms of the elasticities and higher building rates, then you could get much higher saving something close to close to a billion dollars in savings for Los Angeles alone. So while there is a range, I think we're somewhat close to sort of the central expectation, you could see a little bit less or substantially more
Shane Phillips:Just a point of clarification on the rent reduction figure. So you said 18.1%, for the Los Angeles metro area, just to assign this around number, let's say that rents went up by 30%, over a 10 year period in the Los Angeles metro area, in the real world, like that's what actually happened, when you say, La rents went down by 18.1%. That's 18% of that 30%. It's not they went from 30 to 12. It's they went from 30 to you know, 24 or something?
Kevin Corinth:Not quite so it's it would be 18% of 130%. So it's saying so say whatever rents are in 10 years, say it's $2,000. It's going to be 18%. Less than that. $2,000. So, it's something in between what you said,
Shane Phillips:Yeah, the the 130% you're talking about is it's sort of taking the the 100% rent that you started with adding 30% So that's where the 130 so Oh, Okay, so yeah, if rents did go up 30% over that 10 year period. But then, you know, the counterfactual is you built at the 90th percentile and rents were 18%, lower. An 18% reduction on 130% Gets you pretty close back to where you started a little bit higher. But you know, that's that's significant. Yes. So another limitation that you mentioned in the article, Kevin is that as lots of new homes are built, and rents fall, at least relative to the counterfactual, many people would probably seek out higher quality housing rather than just accept lower prices and kind of stay where they're at. This would raise equilibrium rents above the levels that you estimate in the study. So some of these federal rent assistance savings would not materialize. I thought you had some interesting responses to that possibility, one that reforms the program a bit and another that just sort of accepts this new reality of people choosing to pay more for higher quality housing. So could you say a bit about both of those options?
Kevin Corinth:Yeah, definitely. So the the simulation is focused on how rents change for a fixed quality of housing. So if in fact, housing quality does rise, market rents may not fall by as much as we say in the paper. Now, from a policy perspective, there's a couple of ways to deal with this. One way is just to let the program work as it was designed, which means that tenants will get higher quality units. But we may not see quite as many new tenants getting assistance. As we said, the better option in my mind would be to peg the fair market rent to something lower in the rent distribution, perhaps say the currently the fair market rent is pegged at about the 48th percentile or just below the median of market rents. Say you instead peg that at about the 30th percentile of market rents. Now, these units would be similar quality to what they have now. But you'd allow more people to rent out units and be served by the voucher program, just given how many people receive no assistance at all, I do think it's better to serve more people rather than trying to boost the quality of housing. Fortunately, the physical quality of housing has improved a lot in recent decades. And the biggest problem now is the lack of affordability, not as much problems with the housing itself. So the goal should be to serve more people rather than providing better housing to those who are currently lucky to get anything at all.
Michael Lens:Yeah, and one relatedly one thing that I think we struggle with, in providing more housing to more people, which I agree with, you should be, are providing housing to more people not necessarily providing more housing, but providing housing to more people, is the issue of people being able to actually successfully lease up with their voucher. And so one thing that, that I think policymakers and scholars have been both puzzled by and frustrated by over the years is that no matter where you go in the country, you know, well, south of 100% of the people who receive vouchers use them. So we call that the utilization rate. And you know, that's usually something like 25 to 30% people do not use. So their vouchers that they're given in a given year, usually have six months to sign a lease, basically. And then they don't sign that lease and they forego the voucher. What happens after that depends, I think, on the housing authority, but the question I have is, do you think that a lower value voucher that might come about because housing is less expensive, might be more likely to go unused? or would like additional housing stock, make it easier to find housing and perhaps raise utilization rates? I don't know how much you all have thought about the utilization problem relative to that what you're looking at?
Kevin Corinth:Yeah, no, I think that's an important point. And it's something that we note some uncertainty about in the paper, I would suspect that as supply was expanded, and the market in Los Angeles became not quite as tight, that you'd likely see an increase in successful utilization of vouchers. The places with the lowest utilization rates are places with the tightest markets. So places like Los Angeles so I think it stands to figure that as you loosen the market, there became more vacancies as more units came online, you probably see a more successful utilization rate.
Michael Lens:Yeah, and I think another thing I would pile on here is that some of the research that we have on the value of when we change the dollar value of vouchers looking at things like this small area, fair market rent program, that you're able to use those vouchers in the higher diversity of places right where you know this, you started off talking about access to opportunity, right and issues of segregation and The fact that people with vouchers or without kind of housing subsidies, they're not able to access neighborhoods with better job and public safety and education resources that like, it stands to reason, I think as well, where if we build more housing and more places, then there's a greater diversity of places for people with subsidies to live.
Shane Phillips:But I do have the, you know, I think a valid concern here is that by lowering the rent threshold, you know, as a percentile across the whole market, so it's 40th percentile in most places. Now, if you lowered it to the 35th percentile or 30th percentile rent, are you going to kind of encourage more segregation or at least relative to the baseline or the scenario where you keep it at that 40th percentile? Mark, I think there's a real valid concern of a trade off that might occur there, which, you know, in the same way, as we don't necessarily want to think just about cost per household or other things. We also want to think about kind of the equity implications of this and whether, okay, we're helping more people, but a greater share of them are in communities with fewer resources. Like that's not the best outcome yet, either. Or it's at least like a trade off that we really would want to debate a lot.
Kevin Corinth:Yeah, I mean, I I agree with that. And I know I certainly believe the research that kind of where you grow up does matter. We've seen that with moving to opportunity with public housing demolitions, neighborhood matters, and there is reason to be concerned about where people are, are living. That said, I think the first order problem is, we have a lot of people who want these vouchers, and very few people actually get them. In terms of fairness, we should be lowering the amount so more people can have access. There's different ways to do it. I mean, one way is to decrease the amount of the voucher, say to the 30th percentile, make the 30th percentile the fair market rent. I mean, another way is to put time limits. So this is something that even like Rob Collinson, and Ingrid Gould Allen had talked about, if not everyone can have a permanent voucher, what about having everyone gets a couple years of a voucher, and maybe that helps them to stabilize, get access to a new neighborhood, and move up that way. I'm sure many listeners will say, just fund the program and everybody gets the voucher and you get it at the 40th or 50th percentile. We're good. But I think in a world of where that doesn't happen, we should think creatively about what our goals are of this program. And how can we do it in the most fair and efficient way?
Shane Phillips:Yeah, and we we actually had Rob Collinson on a while back to talk about vouchers, and maybe a hybrid here that is part of what he's proposed is this small area of fair market rents where you have different rent thresholds by zip code or other kinds of smaller geographies. But rather than it being at the 40th percentile in those geographies, maybe it's at the 30th. So people still have access to that kind of 30th. And below housing in every neighborhood, it just costs, it costs a little more in these higher cost areas. But you actually may find some savings and lower cost areas. So maybe it mostly balances out. Yeah, I do want to actually just quickly, you know, have you talk about that question of why don't we just fully fund this? Like, why are we talking about five or 10% improvements? When 75% of people who are eligible don't get anything at all out of this program? What do you say to the people who say, you know, if we just made this an entitlement if we increased federal funding, so everyone eligible could get a voucher, we wouldn't have to be pinching pennies and worrying about, you know, these these really challenging trade offs about do we help more people less? Or do we help fewer people more? What do you say to that?
Kevin Corinth:Well, I would say they should care even more about our paper. So so if you had an entitlement to housing assistance, and every eligible household received a voucher, then government savings from deregulation would be even even larger. Because the government would in that case, be paying the excessively high market rents for a lot more people. And then you could depending on what your preferences are, but say then you put those savings into the SNAP program, or cash assistance programs, or whatever you want. So you don't get around this problem by simply having an entitlement to housing assistance. Deregulation would have even more cost savings in that case. And in terms of what the preferred policy outcome is, I think it depends on a lot of value judgments about one how much assistance to provide to low income families, and then to how to provide that assistance, whether it's in the form of specific things like this amount for housing, this amount for food and nutrition, this amount for health insurance, or do you want some more flexible assistance? Do you want to put requirements on it or timeline That's or not? And I think there'll be a lot of disagreement on those questions, and rightly so. But I think there should be much more agreement on the smaller question that we posed in this paper, everybody wants people to have better access to lower cost housing. Everyone wants to serve more people with the funds that we spend, regardless of what that level is. Therefore, you know, regulation is deregulation is good for actually many, many reasons. But this paper adds one more, one more reason.
Michael Lens:To follow up on that, Kevin as well. I mean, there's a fair amount of research that would suggest pretty strongly that if we immediately served the other 75% of people that don't or that qualify for vouchers, but do not receive it right now, that landlords would eat a lot of that subsidy. And the cost of housing for everybody would just go up because there's much more money earmarked for housing. Right. So, you know, I think that's another reason why you have to link these these these two issues, land use regulation, and the value or mechanisms of housing subsidies together. Another housing subsidy that gets some attention in this paper, Kevin, is the Low Income Housing Tax Credit. We talked about that briefly at the outset. As we discussed, it's a tax credit that is the main source of funding for below market rate housing production in this country. And what you do in this paper, with regards to the Low Income Housing Tax Credit Program is that you simulate what would happen if we instead of relaxing land use regulations, we really increase the scope of that tax credit program,
Shane Phillips:Doubling the funding for it,
Michael Lens:I think it's doubling the funds. Yes, thank you, Shane, and you find that the effects are pretty minimal. You know, first, I want you to just kind of give those basic comparisons. And then I think what other folks might critique there is that the Low Income Housing Tax Credit Program is going to produce a housing subsidy for people on its own right. It's not, its main mechanism is not to make the voucher program kind of more effective for people. So, you know, what do you just kind of those two things? How do you respond to that?
Kevin Corinth:Sure. So let me preface this by saying that I don't think our simulation should shift your thinking much at all on whether lie Tech is a good way to provide affordable housing. The point we make in the paper is simply that because life tech crowds out to some substantial extent market rate housing, expanding ly tax subsidized housing is not another way to reduce the costs of our rental housing assistance programs and supply constrained areas. Okay, of course, live tech wasn't intended to reduce the cost of our other programs. But it's making that point that said, I will not defend lie tech on other grounds, I think there are reasons to worry about relying on lie tech to increase affordable housing, especially that much of the subsidy is captured by parties, other than the tenants. And it's also not well targeted to the neediest families, unless it gets layered on with vouchers and other forms of assistance, which further increases its costs. If we're going to help families better afford housing, I think it's best to provide the assistance directly to the families while building lots of market rate housing, without lots of excessive burdens so that families have as much choice as possible. With this malls a costly government as possible.
Michael Lens:Yeah, I think most people probably aren't super aware of the crowd out effects of of live tech, you know, and you know, some people don't care because they don't really want to see a lot of freight and housing production. And they are happy to have that replacement. But it's definitely something that more people need to be aware of.
Shane Phillips:I do think a lot of the critique of light tech rests on that crowd out question. Yeah. And the estimates are that have a very wide range. One of the more recent studies, which is itself from back in 2010, it estimated that the crowd out, at least at kind of the local level was approaching 100%. So for every light tech affordable housing unit, you're building, you're building roughly one less market rate unit, and so that the overall supply benefit is not really there at all. I suspect that's an overestimate to some extent. But you know, an argument I make in favor of light tech is that unlike vouchers, it is directly increasing the supply, but the more crowd out there is, which is just something we don't know very well, but it certainly exists. But the more crowd out there is, the less the supply is actually increasing through that program. So I think maybe for our last question, Here are questions. We should talk a little bit about policy? And a big question that I had. And this is really a recurring question for me is whether at these higher rates of production or this higher production goal, developers would consistently at, you know, for a sustained period, build, and build and build and build, if prices were maybe not only stabilizing, but in some cases falling. What would it take for us to actually, you know, this is effectively increasing in Los Angeles, this would mean increasing our production by about two and a half times. So like, how do we get there? How do we know that developers as prices start to stabilize and maybe in some neighborhoods even fall, that they won't say, you know, we own a lot of the housing here, there's a lot of competition, we're just going to stop building kind of let prices climb a little bit or recover before we continue. And so when we're so almost entirely dependent on on the market to produce this housing, is there a risk of the market not delivering? Or if you if you don't find that concern, convincing? What do you say to people who really do sincerely have that concern?
Kevin Corinth:Yeah, so I think I have a theoretical response and an empirical response, per se, my, my theoretical response is that I think developers are greedy, they want to make profits. And there's a lot of bad way. So in a place like Los Angeles, prices are way higher than the cost to construct a house, most of the high price is due to the land prices. If you start to allow people to build much more densely, you're going to have a lot of houses on any given acre of land, and the land is going to become a shrinking amount of the total price of the given parcel that contains the housing unit. So I just don't believe that developers once they now are able to build, say, for plexes, or multifamily apartment buildings are just going to say, No, we're not going to do it. We don't want to get those profits. And so I do think you would sustain building for a sustained period of time, in places like Los Angeles, where supply is really is severely constrained, and impro. Home prices are so high empirically. You always want to test your theory with empirical data, I think we can point to examples of places that have sustained housing construction over long periods of time. So Atlanta, for instance, had not had nearly the high housing prices that Los Angeles had. But over the couple of decades before the Great Recession in the housing bubble, but even before the housing bubble started, they were maintaining housing production of around three or 4% every year. And this is a place where home prices were only just a little bit higher than the price to build to construct a home. So you can see that even when profit margins are not that large, developers are willing to build in a sustained basis. And I would believe that if Los Angeles really did allow for that kind of unencumbered building, that you really would see sustained building, because the demand is really strong, and it does exist.
Shane Phillips:I do think there's maybe the folks who live in places like Los Angeles have gotten used to a world in which there aren't a ton of developers, there's just, you know, a relatively small number of very large developers. And so it feels like it's not an oligopoly, certainly not a monopoly. But it's like closer to that than it should be probably it'd be great if we had more developers who have more kind of idiosyncrasies and like they're going to compete on different metrics, and just the competition generally would be would be more fierce. And so having the policies in place that maybe you're not just promoting the 50 unit and 100 unit apartment and condo buildings, but are allowing that sort of missing middle, whether it's a four unit apartment or a 10 unit building, those kinds of things where you can have more people entering the market as contractors and developers might be really key to that, to having that increased competition, and just maybe a couple of data points in favor of that. I look to, you know, Tokyo as a place where it doesn't really have single family zoning, and we think of it as this place with all these towers. But in fact, a lot of the housing construction is in sort of smaller scale, multifamily, like under 10 units, I think often even single family but just on very small lots and multi storey and that kind of thing. And we talked with Ryan Greenway McGreevy, several months ago about Auckland's up zoning where 75% of the city now allows multifamily, but I just saw something only a few weeks ago where when their production increased a lot. It was mostly in sort of the townhome style development It wasn't in the larger apartment buildings. And so again, maybe making room for that kind of development really needs to be a priority, and one that we've kind of overlooked thus far.
Michael Lens:So finally, 90 minutes in, the people can cross off missing middle on their bingo card. We got there
Shane Phillips:That brought up zoning. I'll say it. That's the winning score every other episode, missing middle
Kevin Corinth:And light touch density.
Michael Lens:Oh, that's good.
Shane Phillips:So let's just make this the last question. Kevin, you focused in this paper on the impact of increased supply to rent assistance programs. It's a very, I think, deliberately narrow focus. But is there just anything you want to say about kind of the broader impacts here? And what we're maybe missing by just looking at how housing supply is affecting you know, how much we spend, or how many people we help with housing choice vouchers? Yeah, I know, I shouldn't do this. But I would just say our paper is very, very narrow, in terms of thinking about all of the major benefits of relaxing restrictions on housing. Again, my field has not been on housing supply. It's been on safety net programs. But it's just overwhelming how much research has come out on the excessive costs of regulations across society from lowering access to high wage jobs for adults, reducing the ability of kids to go to higher quality schools, reducing integration across socio economic lines, stunting economic growth, and imposing environmental costs because people have to drive so far and to get to work. I increasingly think that the state of our housing regulations is a cause of a lot of our problems that have plagued society. And I'm heartened by the fact that one people, including me, are recognizing it. But it seems that there's more and more sort of bipartisan push that everyone realizes this is a problem that touches a lot of facets of life. And we should do probably lots of other things. But if we don't tackle the housing supply problem, we just aren't going to be able to solve those big problems. All right. I love that as a place to end. Kevin Corinth . Thank you for joining us on the UCLA Housing Voice Podcast. Thanks, Shane. Thanks, Mike. You can read more about Kevin's work on our website lewis.ucla.edu. Show Notes and a transcript of the interview are there too. The UCLA Lewis Center is on Facebook and Twitter. I'm on Twitter at ShaneDPhillips and Mike is at MC_lens. Thanks for listening. We'll see you next time.