Home / UCLA Housing Voice Podcast / Episode 17: Housing Vouchers with Rob Collinson

More

Episode Summary: Every year, more than two million low-income households receive rental assistance through the Housing Choice Voucher program, a federal program that helps renters afford housing on the private market. Currently, only about one-quarter of those eligible for vouchers receive them due to lack of program funding, though Democrats and the Biden administration have proposed expanding it. For our first episode of 2022, Rob Collinson of the University of Notre Dame joins us to talk about how we can get more bang for our buck from housing vouchers, the benefits and drawbacks of the program’s design, and how his research has already helped shape voucher policy reforms in metro areas across the U.S.

  • Collinson, R., & Ganong, P. (2018). How do changes in housing voucher design affect rent and neighborhood quality?. American Economic Journal: Economic Policy, 10(2), 62-89.
    • Abstract: US housing voucher holders pay their landlord a fraction of household income and the government pays the rest, up to a rent ceiling. We study how two types of changes to the rent ceiling affect landlords and tenants. A policy that makes vouchers more generous across a metro area benefits landlords through increased rents, with minimal impact on neighborhood and unit quality. A second policy that indexes rent ceilings to neighborhood rents leads voucher holders to move into higher quality neighborhoods with lower crime, poverty, and unemployment.
  • Bergman, P., Chetty, R., DeLuca, S., Hendren, N., Katz, L. F., & Palmer, C. (2019). Creating moves to opportunity: Experimental evidence on barriers to neighborhood choice (No. w26164). National Bureau of Economic Research.
  • Rosen, E. (2020). The Voucher Promise. Princeton University Press.
  • Collinson, R., Ellen, I. G., & Ludwig, J. (2019). Reforming housing assistance. The Annals of the American Academy of Political and Social Science, 686(1), 250-285.
  • “Housing Choice Vouchers use the private market to provide rental units for 2.3 million low-income households … The tenant pays at least 30 percent of her income in rent and the housing authority pays the difference, up to a rent ceiling. The local housing authority chooses a Payment Standard (which we refer to as the “rent ceiling”) from 90 percent–110 percent of a federally-set “Fair Market Rent” (FMR). HUD typically sets FMRs at the fortieth percentile of area-by-bedroom level gross rent (rent to landlord plus utility costs).”

 

  • Voucher holders renting units below the rent ceiling generally pay nothing when rents rise; the housing authority pays each extra dollar of such a rent increase. This is important because when the rent ceiling rises landlords can increase rents without worrying that this will cause the voucher holder to move.

 

  • “The primary dataset we use in this paper is a HUD internal administrative database called “PIH Information Center” (PIC) that covers the universe of voucher holders. It contains an anonymous household identifier, an anonymous address identifier, building covariates, the rent ceiling, the FMR, and the contract rent received by a landlord on an annual basis, beginning in 2002. The data have two strengths that we exploit in our analysis. First, we can follow a household if they move in response to a policy change.  Second, the address identifier, coded as a nine-digit ZIP code, enables us to follow a single address over time if it has multiple voucher occupants, which is useful for estimating the impact of an increase in the rent ceiling while holding constant many aspects of unit quality … We supplement PIC with four other datasets … the American Community Survey (ACS) … American Housing Survey (Section IIIB) … Multifamily Tenant Characteristics System (MTCS) … HUD Customer Satisfaction Survey (CSS).”

 

  • We estimate the causal effect of uniform rent ceiling changes on neighborhood and unit housing quality and on voucher rents using two natural experiments. In Section IIIA, we study a 2005 change in FMRs due to availability of updated 2000 Decennial census data. … In Section IIIB, we investigate potential within-unit quality improvements. We make use of a detailed HUD survey that asked 26 questions about time-varying unit quality and was administered to voucher holders on a widespread basis from 2000 to 2003 to construct measures of housing quality. Here, we exploit a 2001 change that raised FMRs from the fortieth percentile to the fiftieth percentile of rents in 39 metro areas.”

 

  • The impact of raising the ceiling on observable quality is very small … A $1 increase in the ceiling has no economically significant impact on the neighborhood quality of voucher tenants, as measured by neighborhood rents or poverty rates, and raises composite hedonic quality by a mere $0.05 cents. In contrast to the quality results, average rents rise by $0.46 cents in response to a $1 increase in the rent ceiling … These results imply that only (0.05/0.46 =) 11 percent of the increased government expenditure went to improvements in observable unit or neighborhood quality.

 

  • A concern with the first research design is an inability to measure detailed elements of quality that might vary over time within the unit. In a different dataset, HUD measured unit quality in much more detail from 2000 to 2003. Using this dataset requires a different identification strategy based on a policy change in 2001, when HUD switched from setting FMRs at the fortieth percentile of the local nonvoucher rent distribution to the fiftieth percentile in 39 metro areas. This policy was implemented not in response to recent housing market conditions, but rather with the explicit goal of “deconcentration” of vouchers from the lowest quality neighborhoods … Setting FMRs at the fiftieth percentile of the local nonvoucher rent distribution raised rent ceilings by an average of 11 percent. For every $1 increase in FMRs, rents rose by $0.47, which is very similar to our estimate of $0.46 when using the rebenchmarking research design. In comparison, composite hedonic quality rose by $0.04.

 

  • Our empirical results from two separate natural experiments show that uniform changes in the ceiling do little to improve either neighborhood or observed unit quality for voucher tenants while increasing rents substantially. We interpret our findings as likely reflecting landlords price discriminating by raising rents in response to rent ceiling changes. Our empirical findings are also consistent with landlords improving unmeasured aspects of unit quality and raising rents to cover the cost of these improvements. However, we view unmeasured quality improvements as unlikely to fully explain the estimated rent increases because we have very detailed measures of unit quality, and if a landlord decides to make unit improvements, then at least some of them would show up in the observable dimensions of unit quality.”

 

  • In contrast to the results in the previous section, we find that tilting the rent ceiling has a big impact on neighborhood quality. Following a court settlement, HUD replaced a single metro-wide FMR in Dallas with ZIP code-level FMRs in early 2011. The new ZIP code-level FMRs were set by multiplying the metro-wide FMR in Dallas by the ratio of the median gross rent of rental units in the ZIP code to median gross rent of units in the metro area. The demonstration caused sharp changes in local rent ceilings, ranging from a decrease of 20 percent to an increase of 30 percent.

 

  • “Rents at the ZIP code-level were highly responsive to the policy change … We find substantial rent increases in more expensive areas and rent decreases in cheaper areas; every $1 change in the rent ceiling caused a $0.57 change in rents. This estimate is similar to the estimates in Section III that a $1 change in the rent ceiling raised rents by $0.46–$0.47 … We find that for every dollar change in the rent ceiling, structure quality changed by $0.19.”

 

  • After the policy change, voucher holders exit the lowest quality neighborhoods in the inner city, moving further south and east to better neighborhoods. Figure 9 shows that the improvement in neighborhood quality was broad-based, and not driven by moves to or away from a single neighborhood.” Neighborhood quality improved by 0.23 standard deviations for Dallas voucher holders who moved during the study period, compared to Fort Worth voucher holders where a single FMR was used.

 

  • Averaging across the entire Dallas metro area, average voucher rents are essentially unchanged after tilting the rent ceiling, as shown in Table 4. Given that average neighborhood quality rose, it is somewhat surprising that this policy was budget neutral. The reason for this is that there is heterogeneity in where voucher holders live and they usually live in low-quality neighborhoods. Because they are concentrated in low-quality, inexpensive neighborhoods, the policy would have saved money absent any behavioral response in terms of improved neighborhood quality. Coincidentally, the additional expenditure on improved neighborhoods almost exactly offsets the cost savings from the policy.”

 

  • “We also compare the neighborhood quality impacts in Dallas to other randomized voucher interventions in Table 5 … The Chetty and Hendren (2017) results, combined with our assumptions, suggest that tilting the rent ceiling in Dallas with ZIP-level rent ceilings would raise a child’s income rank at age 30 by 3.1 percentile points, from the thirty-ninth percentile to the forty-second percentile. This improvement for Dallas is smaller than the predicted improvement for the MTO Experimental group (17 percentage points), about one-half of the impact of offering unrestricted vouchers to public housing residents in MTO, and larger than offering vouchers to unassisted tenants.” Because tilting the rent ceiling is budget neutral, however, it is a very cost-effective way to improve neighborhood quality relative to other interventions.

Shane Phillips 0:04
Happy New Year. This is the UCLA Housing Voice podcast and I'm your host Shane Phillips. Every two weeks we discuss a different housing research paper, translating it into non-academic language to better understand how to create more affordable and more accessible cities. Mike Lens is my co-host for this episode, and we're talking to Assistant Professor Rob Collinson of the University of Notre Dame. You can probably also hear that I'm a bit under the weather as I record this intro, so I will make it quick. This week, our topic is housing choice vouchers, the federally funded locally administered rent assistance program, formerly known as Section eight, and also the inspiration for our podcast name. Over 2 million households rely on the program. And I think it's fair to say that Rob's work in this field has been really influential. His and his colleagues' research helped inform the expansion of more locally responsive and cost-effective voucher design in metro areas across the country. And that's what we'll be talking about. Specifically, we're discussing different ways that the program can be designed to provide the most benefits to renters and the least capture on the part of landlords, as many of our listeners will know, Democrats and the Biden administration have proposed expanding the program to make it an entitlement. So this is an especially timely conversation. Like a lot of what we talked about. It's a great case study on the importance of wonky policy design details. The Housing Voice podcast is a production of the UCLA Lewis Center for Regional Policy Studies. And you can contact me with questions or research paper ideas at Shanephillips@ucla.edu or on Twitter @ShaneDPhillips. With that, here's Professor Rob Collinson.

Michael Lens 1:54
Welcome everybody to the Housing Voice podcast. I am Mike Lens of the UCLA Lewis Center and UCLA Luskin School of Public Affairs. I'm joined this week with my co-host, Shane Phillips. Hello, Shane.

Hello.

And our special guests this week is Rob Collinson, Assistant Professor of Economics from Notre Dame, an old friend of mine, although I'm older than he is, and Rob is one of my very favorite, not just because we're friends, but one of my very favorite economists, studying housing generally. And most specifically, we're going to get into a paper that he wrote with Peter again on called, How do changes in housing voucher design affect rent and neighborhood quality? This was published in 2018 in the American Economic journal. Welcome, Rob.

Rob Collinson 2:50
Thank you. Thanks so much for having me, Mike. Great to see you. Shane, very nice to meet you; excited to be on the podcast.

Michael Lens 2:56
Awesome. So our first question that we always ask our guests, is, if you were giving us a tour of your city, it can be South Bend or somewhere else that you love, what would you want to show us?

That's a good one

There's something that we have to say yes.

Rob Collinson 3:12
So no, I mean, coming from Indiana, at least where I'm this new home state of mine. You know, we I think it kind of has the reputation as a flyover state of sorts. And so, you know, I think there are maybe not a lot of famous sites that I would point people towards, I think for people that are sort of genuinely interested in cities, I think South Bend itself has a pretty interesting history. It was once the home of Studebaker, which is a big car manufacturer for a long time, which sort of folded much, much earlier than a lot of the American manufacturing decline occurred. And so South Bend's got some interesting sort of old vestiges of the auto manufacturing past, a little south of downtown, so I might take you around there. But for those that sort of may have more conventional preferences for sort of what they like to do, I probably steer people, honestly just towards the beaches of Lake Michigan. So Mike, you're a native midwesterner so you can have a greater appreciation for the Great Lakes than most. But you know, we have a lot of international faculty and stuff like that at Notre Dame that come and have never seen the Great Lakes before. You know, they take a job at Notre Dame and you know, I think personally, having grown up around them that they're just like an amazing natural asset and are beautiful, particularly in the summer. And so I would probably steer people to go up to the sort of south east end of Lake Michigan, not too far from where we are, and enjoy some of the beauty of Lake Michigan for a day.

Michael Lens 4:48
I'm not sure I miss the beaches of the upper Midwest. When I'm not living in London as I am right now. I am of course in in Los Angeles and about four, four miles from the Santa Monica Beach. So I think the beaches are totally fine where I'm at now, but any Californian would I think find it strange how much I miss just like floating on a boat in the middle of a lake somewhere in the Upper Midwest. Like, I am sorry, I do not wish to be like, rustled and bustled on an ocean doing that, like it's too much a little scary. Yeah, you know, I've got a weak stomach, there may be sharks, etc... pretty sure. But like, the tranquility of like sitting in the middle of a lake, in Michigan, Minnesota, Wisconsin, whatever, it's something that most Californians would have no idea about.

Shane Phillips 5:52
We really don't have much lake going on here...

Michael Lens 5:55
No

... in Southern California got like, what are you gonna go to the Salton Sea or something? I'm not sure. So we'll get into these papers. And I'll actually say, first off, kudos on a paper title that just very clearly conveys what we're going to be talking about, 'How do changes in housing voucher design, affect rent and neighborhood quality?', that's exactly what we're looking at. Your paper is looking at two different changes to the design of the voucher program. And their impact on two things, the rents that landlords charge, and the quality of neighborhoods and housing units that voucher holders live in. The first change that you look at is a uniform increase to the maximum rents the vouchers will subsidize and the second change increases the cap in high-cost neighborhoods, and actually lowers it in low-cost neighborhoods, what you describe as tilting the rent ceiling. And you find that the tilting approach is both more impactful in ways that we'll discuss and more cost-effective to the point where the change in your study at least is basically cost-neutral. And that's important, of course, because it means that we're ultimately getting more out of every dollar we spend on the program. So we'll get into those details and what lessons we can take from them. But let's just start with the basics, as we always do. What is the Housing Choice Voucher Program, if you could just give us a quick overview on who it helps, how it works, and you know if you can give us a little bit of context as well, about this process we've been going through for the past 40 years or so where we switched away from, especially the federal level, away from public housing, and more toward housing vouchers and the Low Income Housing Tax Credit, how it all fits into that picture.

Rob Collinson 7:40
Sure. Yeah, absolutely. So the housing voucher program is the largest rental assistance program, it was administered by HUD that we have currently in the US, it was created sort of in the mid-70s. They experimented with a couple of different designs of it. The way it works today is a tenant, a low-income tenants, is issued a voucher from a Housing Authority, it's then incumbent on the tenant to try to find a landlord that will lease to them. And so they search for landlords, primarily on the private market. And if they find a landlord that is willing to lease to them, the voucher tenant, the person with the voucher is going to pay 30% of their adjusted income towards rent. And the government sort of going to pick up the tab for the difference between that 30%, and the market rent on the unit. And so the voucher program serves about 2.5 million or so households now, since sort of moderate growth over the last few years, it serves overwhelmingly extremely low-income households so the vast majority of people receiving vouchers are in less than 30% of area median income. So the sort of typical income for a household with a voucher is going to be sort of on the order of $12,000 a year or something there about. So it's very critical tool for trying to allow low-income households to be affordably housed. But it's also sort of through its history had maybe a dual goal in some sense of creating more choices that are available to these low-income households about where to live right? And so Shane, you referenced earlier that the movement away from something like public housing and towards vouchers, there was sort of two strains of that one was, you know, a lot of this was happening during sort of Nixon era stuff. And there was a movement towards more private market involvement in the provision of housing for low-income households. And so some of that was sort of driving that progression. But there was also sort of an acknowledgment that there were early concerns about potentially concentrating, you know, very low-income households and only a small set of communities. And so, the idea of sort of the long story discussion about slums and sort of the implication of "slums" for people welfare, kids development, all of these different things, that was also sort of a strand in the development of the voucher program, and the sort of aspirations for the program to potentially open up new neighborhood choices for low-income households.

Michael Lens 10:15
So, Rob, thank you so much for that context, it's really important. It's something that I think a lot of people don't know about kind of the weeds of federal housing policy over the last few decades, just how profound a shift that's been, and how large the voucher program is, even if I'll, you know, kind of out myself as somebody who doesn't think it's large enough. And I guess, you know, another kind of on the 'is it large enough' front, you know, for economists in particular, and people who really study incentives, you know, incentives are both those who benefit from this program, and also landlords and various other actors in housing provision, you know, there's really something about the efficiency of this program. Can you spell that out a little bit for us? Like, like, why might it be better to...

Rob Collinson 11:13
Why do economists love vouchers?

Michael Lens 11:16
Right, why do economists love vouchers? Like why might we want to do it this way instead of, you know, just building....

Rob Collinson 11:21
Yeah, so I'd say there's kind of two... so generally speaking, if you're at a poll economists, you're exactly right, I think there's going to be a strong preference in general, among economists to use a tool like vouchers. And I would say, there's sort of two main reasons for that. So the first goes back to sort of very basic intro micro days that probably many of your listeners if they're forced, or voluntarily took an economics class will be familiar with, which is that economists tend to prefer more cash-like forms of assistance rather than giving things in kind. And so in kind meetings, or paying for some nice food or paying for somebody's housing, or somebody, you know, health insurance or something like that. Now, a voucher is still in kind because it can only be used towards housing. However, it allows for more flexibility on the part of the recipient to sort of choose how much housing they want the particular features of housing that they care about most, at least in the ideal, right? So in the ideal, the voucher is allowing people to not be sort of forced into particular selection of a public housing unit that becomes available that might be, you know, a certain quality level or in a certain neighborhood. It's saying, "hey, people are actually going to be better off if we give them sort of an equivalent amount of resources, but allow them to instead choose, do I want, you know, the sort of nicer maintained unit? Or do I want the unit that's in the district that has really strong schools? Or do I want, you know, the unit that's close to my job", or what have you. And so part of the reason why economists tend to favor something like vouchers over public housing, is precisely because vouchers enable households in theory to have more choice over the type of housing that they want, and what they their sort of consumption split looks like between housing and other goods. And so that's sort of reason number one. The second reason why economists tend to prefer vouchers is that when we provide a subsidy on the supply side, so when we subsidize producers to make stuff, a general concern in economics is what's known as crowd out, which is the idea that some of the people that were paying to make stuff that they either they would have made stuff in absence of the subsidy or that other producers of The Good question would have also made stuff. And so here are the ideas. You know, if we're in the business of subsidizing developers to make housing, well, they may make some new public housing. But that may take the place of some housing that would have been built otherwise, like some, maybe its market rate, maybe it doesn't necessarily have to be subsidized housing, of course, but that sort of some of the net new units that are added are lost through this process whereby developers build more, it pushes down prices, and then some developers are like less encouraged to go and enter the market and build more. And so economists have sort of like, been skeptical of supply side subsidies in like the housing setting because of these concerns that by subsidizing developers, where maybe each dollar that goes towards the development of these new units, some of it isn't actually really contributing to a new unit, because it's crowding out some other housing production that would have otherwise already taken place. Now, you know, I think we could have a much longer conversation. I think one thing you would hear from economists overwhelmingly these days, we need a lot more housing. We need a lot more housing built writ large, and we need it a lot, especially in particular types of housing markets, where regulation exists and high levels, there's a lot of sort of exclusionary zoning and sort of other regulatory tools that are limiting how much production happens. And so to the extent that, you know, a subsidized public housing program was really able to add new units to a market that otherwise has a lot of constraints that restrict new building, I think you could get potentially more economists to agree that that might, you know, be good all together, because sort of the amount of new production that it would really be displacing is maybe not that large. But historically, sort of the reason that economists have tended to like vouchers is for these two reasons that the vouchers provide sort of potentially more choices to the tenant, obviously, this is all in theory, we can talk about in practice, how much it does that. And then second, sort of that it's not crowding out sort of private market activity that might otherwise have

Shane Phillips 15:51
You sort of anticipated my question about whether this crowd out is really happening, especially since you know, if you're building housing that's targeted at people earning 30% of area median income or below, that's an entire, you know, I won't say it's an entirely different markets have unmasked, and others have shown that, you know, the housing market is connected from the top to the bottom. But, you know, generally speaking, it's not as though by building a home for someone earning 30% of AMI, you've crowded out the construction of a home for someone who can afford 2500 or $3,000 a month.

Rob Collinson 16:24
Yeah, no, I think that's right. I mean, I think where you might work, Craddock concerns, at least strike me as potentially more valid would be maybe Low- Income Housing Tax Credit developments in markets that are already really kind of like low rent, where e the rent and the tax credit is pretty darn similar to prevailing market rents and things like that. So I think, to the extent that crowd out is a concern is sort of more pronounced there than in a setting like, you know, public housing where these are deeply, deeply subsidized. And people, you know, really can't find other housing that has comparable rents.

Shane Phillips 17:00
Yeah, right. We've talked about why vouchers generally. But there's a lot of ways that voucher benefits can be structured. And so what are some of the challenges or the inefficiencies of the program as we have it structured now? Or, to put it another way, why does the design of the Housing Choice Voucher Program matter for the households who rely on them?

Rob Collinson 17:23
Yeah, that's a great question. So I'll say that matters writ large, first, sort of two major reasons. So one reference is kind of this aspirational goal for the program that I highlighted earlier, which was that like vouchers, you know, regardless of the sort of place base versus people mobility sort of debate, in which side do you fall on, vouchers, at least in theory, I've always aspired to be or at least from policymaker genuine sort of policymakers that care deeply about the program, you know, the goal for the program has always been that it could be a tool to potentially unlock sort of new neighborhood options for very low-income households. And so to give them the possibility of, if they want to live in a neighborhood that has low poverty, or has low crime, or has good access to jobs, or has high-quality schools, that the voucher could be a tool that would enable households to make those types of moves. So the sad reality is that sort of when we've looked at the best empirical evidence on the extent to which vouchers do that, it historically hasn't had particularly large impacts on some of the sort of common measures that people have looked at. If you're a high market tenant, you get issued a voucher, you tend to move to neighborhoods that are somewhat similar to the neighborhoods you came from. You know, might provide, and like I, you could probably speak more on things like job Access, but historically, at least, in terms of moving the needle on poverty, or sort of school quality or crime, you know, that the effects haven't always been huge. And so it hasn't really sort of, I would say, lived up to the promise of trying to deliver sort of a truly wider set of neighborhood options for low-income households that have a voucher. And so part of what we're interested with this study, and I can certainly talk about more is like, well, what could we do differently about the design that might allow the program to better meet this sort of aspirational goal. The other element of this is that, and this was something that Mike's or indirectly referenced, is that the voucher program is not an entitlement, meaning there are many more households want a voucher, than there are resources available to serve them. So about one in four, one in five eligible households for vouchers your actually receive it. And so, given that resources are very constrained, we want to be sure that the resources that we're devoting are allowing sort of as many households as possible to take advantage of the voucher program. So one of the things that has been a concern for a bit of a time in the voucher program, is that sort of landlords may in some sense, price discriminate against voucher holders and the idea that they sort of no people get a voucher. So in order to lease up with a landlord, in the voucher program, you have to inform them that you have a voucher and they have to be willing to accept it. And so there's sort of this weird dynamic at play whereby landlords kind of know voucher tenants can get rents of a particular level. And they might use that knowledge to sort of charge them potentially more than they would a non-voucher tenant. We can talk about sort of the evidence for that writ large, as well as what we're finding in the paper and more directly, which is slightly different dynamic. But if we care about, you know, serving as many low-income households with a tool like vouchers as possible, right, ensuring that landlords aren't able to capture a bunch of the benefits of the program is going to be important.

Shane Phillips 20:42
Can you can you just really quickly explain what the fair market rent is, and how that factors in, how landlords are able to sort of use that to potentially capture increases in it?

Rob Collinson 20:54
Sure thing. So as I mentioned earlier, the tenants in the voucher program pay 30% of their income very typically, or they pay a minimum of 30% of their adjusted income towards rent. And then the government is sort of footing the difference between that 30%, and the market rents on the unit. But policymakers don't, or at least historically, haven't wanted low income households to live in super plushy luxury units. And so they put a cap on sort of how much the government will pay up to they're not going to pay the difference between 30% of your income and, you know, the $5,000 a month suit, you know, luxury apartment. And so they cap, the amount that they're willing to pay with something called the payment standard. And the payment standard is basically set as a percentage of something called the fair market rent. So we're in the housing program world, which means were intensive on jargon and acronyms, but...

Michael Lens 21:47
So many phrases that have no meaning to anybody,

Shane Phillips 21:50
And percent of percents. I love that that aspect of all of this. It's like....

Rob Collinson 21:55
Exactly, so the key thing is that the payments entered is very closely tied to the fair market rent very often, it's like 100% of the fair market rent, meaning those two things are equal, or it's like 90 or 110% of it tends to be very tightly tied to the fair market rent, well, where's the fair market rent come from? And what is it designed to do? Well, HUD publishes basically estimates in each metro area, and sort of county groups and non-metro parts of the US, where they basically specify what they think sort of a standard quality rental should rent for, by particular bedroom sizes, which is allowed to vary across all these sort of metro areas of the US. So HUD publishes these fair market rents, they're set at the 40th percentile of the rent distribution for the metro area, or more formally, the FMR area. And those are basically providing dollar amounts by bedroom size that that then basically very tightly informed the local cap, rent ceiling that PHA said, by virtue of their payment standard.

Shane Phillips 22:57
And so PHA is the Public housing or...

Rob Collinson 22:59
Public Housing Authorities. Yeah, so maybe it's like a little helpful to ground a numeric example. So like, if I'm a low - income tenant with a voucher, suppose I make $12,000 a year, so I make $1,000 a month, I'm going to pay 30% of my income towards rent. And let's say the local housing authority is set the payment standard to be $1,200 per month. Well, what that means is, if I find a unit that rents exactly for $1,200, I'm going to pay that $300, and the Housing Authority is going to pay that $900 difference between my $300 contribution and the $1,200 rent, that happens to be the same as the payment standard. If I were to rent a unit that's less than the payment standard, like say I rent a unit, that's only 1000 bucks, then the government will pay that $700 difference, right? If I rent a unit that's more than that $1,200. So suppose I ran a unit for $1,300 a month, I can actually rent that unit under the rules of the voucher program. But now I'm going to pay $300 so my 30% of my income, plus the difference between that payment standard amount, which we said was $1,200, and the rent on the unit, which is $1,300. So I pay that additional $100, giving me a grand total of $400 in rent contribution. And so for rental units that are priced above this payment standard, this rent ceiling, as we call it in the paper, those additional costs above and beyond the payment standard come out of the pocket of the tenant. And that's important because what it means is that this this payment standard is going to have implications for the types of units that voucher tenants are going to elect to rent out, because they become the sort of residual payer, they have to foot the bill for any amount more than this payment standard. And so this is really closely linked to what I would call like the generosity of the voucher program, basically, are you going to allow people with vouchers to rent sort of relatively more expensive units or are you going to curtail it so they can only rent sort of very low-cost units. And so what we end up doing in our paper is studying some different types of changes in that amount that rent ceiling. And look at the implications for those changes both for the types of neighborhoods that the tenants get, and also for the rents that landlords charge. Right. And so, you know, we look at really two major types of changes. So one is what we call this sort of uniform increase or the across the board. And so in my little numeric example, where they initially are setting the payment standard at $1,200. And that's obviously a rent that's way too low for LA. So let's say that's the fair market rent for, for a three bedroom in South Bend that's about I think, a three bedroom in South Bend to give you an idea of the low-cost housing we have here. But so let's say the payment...

Michael Lens 25:45
Are you hiring at Notre Dame?

Rob Collinson 25:49
So we have their $1,200 payment standard for a three-bedroom in South Bend. One of the sort of changes we're going to look at is suppose HUD just decides to increase the FMR, and by virtue of that increase in FMR, it's going to trigger local housing authorities to adjust up their payment standards. And so we say what's basically what's the impact of moving from basically that $1,200 effort mark to maybe, let's say, 1350, or something. So they increase basically that ceiling, from 1200 to 1350. Well, in theory, that's more potential income, in essence, for the voucher tenant to potentially use to go out and try to find a new unit, right? Like this is raising the cap, its meaning that there's potentially nicer units out there that the voucher holder could go out and lease with. And we have some sort of, you know, clever research designs that, you know, we call quasi-experiments. And I'm happy to dig into the details or not of those. But we're basically going to look at what happens to voucher tenants in areas that get somewhat random-ish increases in the ceiling, and compare them to the voucher tenants and places that either don't get those increases, or even potentially see decreases. And one thing that we see is sort of when we increase in an across-the-board way, meaning that that move from $1,200 to 1350. That's true, irrespective of the neighborhood that you locate in within, let's say, South Bend. So we raise it everywhere, potentially, what we find is that voucher holders don't seem to necessarily respond by going out in bed getting better quality units, at least as measured by both sort of like a variety of survey data on sort of what they what they report about the quality internally of the structure that they're in, but also by sort of measures of neighborhood, we don't see that they're sort of taking that additional money and going out and finding better units. What we do find, however, is that landlords seem to increase rents potentially little more than they otherwise would, in response to the availability of basically more subsidy for which the government is the residual payer. So they're going to respond to this increase in the cap by saying, hey, actually, you know, maybe we can raise the rent a little bit more on our voucher voucher tenants, because now we know that the local FMR is going up from $1,200 to 1350.

Yeah, it doesn't change how much the tenants themselves have to pay. Exactly. So the because the tenant is price insensitive, as, as an economist would describe it, you know, they're maybe not taking advantage of this in the same way, and it's enabling landlords to maybe get a little bit claw some of that benefit towards themselves.

Shane Phillips 28:27
Mm hmm. And, Mike, do you want to ask about the the research design, I do think this is worth we're talking about?

Michael Lens 28:33
Well, I mean, I was first going to make a couple of very important insights. So one is that it is pretty crucial to have economists on the case, or at least people who are very good at math, because there was a lot of math and there, there are a lot of dollar values that you have to keep track of, not just the jargon. And then as somebody who has spent, you know, a decade teaching this stuff, and more than a decade, researching this stuff, it is so complicated, this policy- making contraption that we have created and kind of layered on over the years, that I still learn things when other people explain it. Or I'm reminded of things that I forgot about the complexity of the voucher program. So thank you for for walking us through that. And then getting into, on some of those policy changes that you're able to take advantage of. So I guess you talked about the you know, being able to study this, you know, random or quasi-random variation, can you walk us through like, what makes the kind of effects of these changes plausible or more plausible because of the nature these programs, and how you were able to study?

Rob Collinson 30:02
Yeah, so I think the key thing to think about was, you know, we'd liked it, the sort of thought experiment that we'd like to run is sort of, we've got to set up housing markets, and we want to randomly give some housing markets, voucher values that are higher, basically FMRs, and consequently, payment standards that are higher in some markets and leave others untouched. And then we would sort of see how do tenants respond and change behavior? How do landlords respond to these particularly randomly allocated, increases in FMRs, and consequently, payment standards? Now, the challenge was just saying looking at, say, in what places are fMRI is going up and in what places are they not and, and comparing the activity in places where FMRs are naturally going up, versus places where they're not, is that the places where FMRs is are naturally going up, presumably, are places that rents are going up writ large, that's why the FMRs is going up. Maybe the housing market is booming, you know, there's maybe a lot of forces that might cause a voucher holder, to have maybe more trouble getting into a better neighborhood or a better unit, or, you know, might cause landlords to be particularly aggressive at jacking up rents or what have you. And so, you know, just...

Shane Phillips 31:12
The idea, just to spell this out the idea is, basically that in those places, the FMR might be going up, but your ability to afford a place at the higher FMR is basically staying the same, no higher quality neighborhoods or units are opening up to you because the rents in those are also going up by a similar...

Rob Collinson 31:30
Yeah, yeah, exactly. And so what we're trying to do is separate out a little bit the what's the role of, you know, increasing basically how much subsidy is available to the voucher holder, while sort of holding constant the dynamics of the housing market, if you will. And so the way we try to get at this, we do two different ways. Let me describe the first, because I think it sort of relates intuitively to that random experiment that I just described, which is, we look at these years in which HUD updates their FMR values according to new census data. And so in a really sort of arcane fact of how housing, housing FMRs are set, HUD basically, for a long time would get some data from census, and that census data would only come in once a decade, because it was census data, and they would have to do various updates to it and basically, and then basically, as time passes, a new census would come out. And they would have to update the values. And because of the nature of the corrections they were doing, and because they're sort of this large change to the amount of information they have to set FMRs, you would get these really wild swings and FMRs that are basically in a sense correcting for accumulated measurement error.

Shane Phillips 32:46
And because they're changing them year by year based on limited data, but then every 10 years, they get this really exhaustive data that they can...

Rob Collinson 32:55
Yeah, exactly. So they insert this new injection of data. And yes, there's some signal in it, but there's a lot of noise in the correction, right? Because they're they're the HUD sort of guesses 10 years after the data came out about what the rent should have been, in particular places tends to be a pretty noisy measure. And so basically, what we see is that there's this sort of large correction, in some places, the correction can go up or down. And that correction is uncorrelated with sort of, like measures of what's going on in the housing market. So it sort of passes the smell test of this sort of randomized thought experiment that we had in our minds of some places get random increases, other places don't. And we use that to basically then say, well, what happens? What's the experience about your tenants in those places? Do they, in fact, move to better-quality units neighborhoods? You know, do landlords change sort of the rents that they're charging voucher holders in the different regions?

Shane Phillips 33:53
And that's how, or at least, one of the ways that you looked at the uniform rent increase and what the impact of that was? Now, can you tell us what the tilted ceiling scenario is and how you studied that one?

Rob Collinson 34:06
Yeah, absolutely. So as I mentioned, you know, with FMR is they're set over the rent distribution of entire metro area. So when HUD goes and gets the data from census that informs that final FMR calculation, they're getting a statistic that you know, for the Los Angeles area is basically the Los Angeles Metro. So it's going to use units in Silverlake as well as units in Englewood and South Central and it's going to use Hollywood and Melrose you know, all of these different neighborhoods that themselves have a ton of variation in underlying rents, right. So, as we know, in cities, there's wide variation across cities or metro areas in the rent levels within particular neighborhoods. And so starting in 2010, as a result of a lawsuit that actually took place in Dallas, there was a measure by HUD to experiment with a new way of setting these fair market rents. So instead of setting them uniformly across a metro area, they would basically calculate within zip codes, the FMR within each of these different zip codes, and apply those instead of a single number by bedroom size across. So they do sort of zip-by-bedroom rent figures. And what that does is it has the effect of, you know, as we would expect, increasing the fair market rents in high-rent, zip codes, and lowering the fair market rent in low-rent zip codes. And so in essence, if you sort of imagine on your Y axis, sort of the rent ceiling, basically what the government will pay up to you and you imagine on your x axis, some notion of sort of neighborhood quality that's somewhat related to rents, what's happening is that, and my camera's reverse, the rents are going to go up in the neighborhoods with higher quality, higher rent values, and presumably higher quality, and the Payment Center is going to fall in neighborhoods that have sort of, "lower-quality" or lower rents. Now, I want to be careful about what neighbor quality means because I think it can mean a lot of different things. Here, I'm just sort of talking about it as being very closely connected to the market rents in those neighborhoods. But basically, in response to this lawsuit in Dallas, starting in 2010, they began to introduce these zip code based fair market rents, rather than setting it uniformly across the metro area. And so we take that particular sort of natural experiment in some sense. And we compare basically what happens to voucher holders in Dallas who are affected or subjected to this new policy, while the voucher holders in Fort Worth just next door, featuring a similar housing market, they are not able to take advantage of this policy. And so what we do is basically compare what's happening to the amount that voucher holders are getting in terms of neighborhood quality, what's happening to the types of communities that they're moving into, in Dallas relative to in Fort Worth, sort of before and after this change?

Shane Phillips 37:05
This is a bit of an aside from the previous question, but I did want to ask it, the update that happens based on the census data takes place, in your case in 2005, for the 2000 census. Yeah, why does it take five years to update that? I mean, I know it takes time to tabulate everything, but like five years seems like a long time.

Rob Collinson 37:28
Yeah, so they've had periods where it's happened a little faster. You know, I don't want to certainly disparage my colleagues at Hyde. I think it does, legitimately it takes a long time for the estimates to come from Census Bureau, it's what's known as a Special Tabulation. And there's just kind of a lot of work that goes into incorporating basically, this new census data, making sure it's delivering sensible estimates. This is a figure that's going to impact, you know, millions of voucher households. And so, you know, I think, historically, there's been some longer lags and when those things get updated, you know, the silver lining of all this is that we now have the American Community Survey, which is replacing the census long form, which was the previous tool used by HUD to construct the FMRs. And so now we get fuzzier, but more regular sort of pictures of what's happening to rent in different communities.

Shane Phillips 38:26
So it's still pretty good quality.

Rob Collinson 38:28
Yeah, yeah. Yeah. So this corrections are happening a bit quicker now.

Shane Phillips 38:31
Okay, so we've described what the uniform versus the tilted rent ceiling changes look like? What was the result of these two different changes, then? What did you find?

Rob Collinson 38:41
Yeah, yeah, so with the with the uniform increase, basically, where we don't find much evidence that tenants go out and take advantage of essentially this extra money available in, that their voucher is now worth more, they don't seem to to rent units that are sort of higher quality as measured by sort of the particulars of the unit when it was built the size, whether or not the tenant reported, you know, physical issues with the unit, etc. They don't seem to lease better quality units. And they also don't necessarily seem to go to neighborhoods that look very different from the neighborhoods they were coming from. So they tend to sort of stay in the types of neighborhoods that they were leasing in before this increase. We do find that landlords increase rents somewhat in response, you know, it's very far from capturing the full dollar of that, you know, dollar increment increase in the rent ceiling. Nevertheless, like it does accumulate over time. And so, you know, landlords are benefiting from these increases. That's what the uniform increase, when we look instead at this policy that then changes it and basically allows the rents that vouchers can pay to go up in sort of higher rent neighborhoods that we presume are somewhat higher quality and cuts the value in lower rent neighborhoods. There we actually find evidence of fairly sizable behavioral response on the part of the tenant. So tenants in Dallas that were able to take advantage of this increase in FMR in the sort of higher quality neighborhoods, they appear to respond to this increase in like, and also the cut in lower-end neighborhoods by moving out of a lot of the poorest neighborhoods and toward new neighborhoods that have both lower poverty, but also quite a bit lower crime and lower levels of unemployment. So there, we see a fairly sharp behavioral change on the part of tenants where they're now migrating to less impoverished sort of more somewhat safer neighborhoods as a result of this policy change, in contrast to sort of the across the board increase, so as economists we like to say like, you know, incentives matter in a sense that, you know, by changing a bit the incentives that voucher holders face to go lease up in higher quality or sort of less impoverished lower crime neighborhoods by virtue of having the voucher worth more in those neighborhoods, but also worth less in the neighborhoods that maybe they would have chosen otherwise, people seem to be spurred a bit more into action and seem to be migrating more to neighborhoods that we would presume, potentially, maybe offer sort of better opportunities for children and things like that.

Michael Lens 40:55
What role do you think information plays in that. I mean, there's been a lot of discussion at various points in kind of thinking about how to reform the voucher program that really got to put some faith in like, how we can, you know, maybe counsel people to different neighborhoods, right? And like, so what do we know about like, when these changes are made, obviously, like, a beneficiary of this program is not going to be versed in, in things like, uniform and tilted changes, and you know, but I mean, even more simplistically, like, right, to what extent do they know like that, "okay, I can use a higher dollar voucher in particular places"?

Rob Collinson 42:03
Yeah, so I think it's fairly saline in the sense that the housing authorities in response to getting this, you know, zip code list, essentially, from HUD, they then have to publish sort of what the payment standards are in the different neighborhoods by bedroom size. And so if you're searching around for a unit, there's a good chance, you're going to look up your zip code, potentially, and say, like, "would a voucher work here?" Now, you know, I don't know that a lot of tenants would be like, "Oh, I'm getting the zip code payment standard, rather than, you know, this uniform increase", and necessarily know the reasons behind all of this. But I do think the change in in the sense of providing potentially a very different dollar threshold that becomes, you know, the focus of finding units often is relatively salient. I will say, like, and this is a little bit what you're asking, but sort of somewhat tangential, which is that I think, like the role of information and counseling in neighborhood choices is tremendously important. And so I think this is one part, there's sort of like, I think the bigger question might be, how well do families know about sort of the level of opportunity that different communities might offer for them? And sort of how accurate does that map to, to reality right? So I think some of this work that Raj Chetty and others are doing with creating moves opportunities, CMTO in Seattle that's providing sort of custom counseling and very concrete information on like, these are the neighborhoods where, you know, atleast the data suggests your child is most likely to thrive. I think, in a lot of ways that the information is quite important there. There's the separate piece of how much do they know the information about the degree to which the program works and things like that. But I think, you know, concretely, if you're a family, and you're trying to lease in a low-income neighborhood, and you know, you could previously use your voucher to rent a unit for $1,200. And now, the Housing Authority saying it only pays up to $800 in that neighborhood. That message is, you know, I think comes through somewhat clearly when you go and, you know, try to release that $1,200 unit in a low rent neighborhood. So, but I do think information in the housing market writ large is, you know, a really potentially important channel that we could make progress on in terms of getting potentially the voucher program to live up to some of its promise in terms of delivering higher opportunities for families.

Shane Phillips 44:35
Yeah, and to talk about the cost here a little bit because I mentioned it at the beginning of our conversation when I was giving the summary. I think there are sort of two extremes that you could see. One is and this is specifically with the tilted ceiling, the small area fair market rent option. One thing that could happen is rents in lower income neighborhoods fall the FMR in those zip codes goes down, it goes down in some places up and others, but no one changes their living situation. And so just the amount that Todd is paying falls. Another option is you do this tilted ceiling, everyone moves to a higher quality neighborhood or higher quality housing. And the costs go way up. And there's a whole spectrum of options in between that. And what you found, basically, is that just enough people moved, and you know, the cost fell in the lower places and right rows in the higher places that it was essentially cost neutral, right?

Rob Collinson 45:33
Yeah, that's exactly right. So if you like long term, if you thought this tool was, you know, really going to encourage a huge number of voucher holders to go out and move to, you know, neighborhoods that are higher rent, overall, this will cause voucher costs to increase over time, that's definitely the case. Yeah, in our particular setting, the amount of migration, which is pretty substantial, is not enough to exceed basically, or they exceed by very much the amount that saved essentially, by cutting back on how much rents are paid in low rent zip codes. So it ends up sort of cost neutral, but you could see it going either direction, potentially costs a third, no migration, you know, it'd be a cost savings thing. We might worry, obviously, the voucher holders in low-rent neighborhoods are paying more out of pocket now, or they're getting lower quality units or what have you. And, you know, in the other extreme with tons of migration or response, it would tend to increase the cost of the voucher program on a per

Shane Phillips 46:36
Yeah. And to be clear that's not necessarily a bad outcome. Because in theory, people well, in practice, people will be living in higher quality neighborhoods, we know there are positive things associated with that. That's right. So it's not to say that if the cost went up, because of this, it would be a bad thing. But the fact that people moved and still cost didn't go up is a pretty impressive outcome. Maybe we can get into that topic a little further by talking about the other programs that try to achieve similar things. So at the end of the article, you compare the estimated impact of the tilted rent ceiling or small area fair market rent programs that design change, to other programs that share the goal of improving housing and neighborhood quality for low-income renters so could you just quickly summarize what those other programs do? And how this design change compared in terms of effectiveness?

Rob Collinson 47:29
Yeah, so so, you know, we obviously compare it to the uniform increase and show that it's more effective than that, you know, we also thought about well, like, how does this compare to the magnitude of changes that we observe in a setting like the Moving to Opportunity experiment that directly required voucher holders to lease a neighborhoods with census tracts below 10% poverty. And so what we see is like we don't generate nearly as large an increase in sort of neighborhood improvements, if you will, as MTO. But MTOs sort of mandate came at a pretty big cost in the sense that only about 47% of the people that went in the experimental group in MTO, that had this restriction leased up with a voucher compared to basically 60% in the group that had sort of an unrestricted, typical voucher. And so one thing that we want to like point out is like, there's this like, yes, you can do better. But can you do better in sort of a cost neutral way that also seems to not have other sort of really significant downsides, although we can talk more about potential limitations in this.

Shane Phillips 48:39
The small area fair market rents, the fair market rents that are calibrated to zip code, those are now being used in many more metro areas around the country. Is that correct?

Rob Collinson 48:50
Yeah, that's right. So after the initial Dallas, HUD did a pilot and I think four or five Forget it was originally going to be, I think, five additional communities. I can't remember exactly if all five follow through, there's been some subsequent evaluation of those that found results that seem really consistent with what we find, which was encouraging. And then, more recently than that HUD actually decided to change what was a previous policy. So the previous policy, which we also study as a part of our sort of exploration of the across-the-board increase, the way it worked was basically housing authorities that were not doing a good job getting their voucher holders into less concentrated poverty neighborhoods of places that were sort of doing a bad job on deconcentration if you will. They would be eligible after some number of years of poor performance for this across-the-board increase that went sort of above the usual amount of payment standard or FMR increase that housing authorities will be eligible for. So we also look at that policy and find super similar results to our other sort of way of investigating the across-the-board increase. And so what happened was decided that to basically take that policy which we showed was not very effective, and instead, replace it with a policy that moves those communities to small area Fair Market Rents or the zipcode, fair market rents as an alternative to that across the board increase. So that was like a nice, very natural policy change that was suggested by our results. And the wonderful people at HUD, including some of Mike and my former colleagues at NYU Furman and other places helped to push through. And so there's been some concrete policy changes. Now I always misquote is either like 27 or 28 metro areas that I think are now subject to the fair market rents that are set at the zip code level that are getting that policy instead of the one that we we found to be less effective.

Shane Phillips 50:48
So that'll be a future paper I imagine. So we've been talking about the positive impacts of these changes, are there any drawbacks that you've observed? Or you know, you speculate might be happening, you did talk about, for example, you know, if you're lowering the cap in these, you know, "lower quality" neighborhoods, or just less expensive neighborhoods, maybe now you're limiting people to the lower quality homes within those lower quality neighborhoods, for example, is there anything else like that, that, you know, there are concerns with this approach?

Rob Collinson 51:18
Yeah, I mean, I think the one reality is it's hard to find a housing with a voucher. And, you know, in general, we don't have good national statistics on this. This is like a big point of frustration for many housing researchers like us. But basically, we think about 70% of families issued a voucher successfully, at least up so about 30% are missing out. And one possibility is that if historically, those people that were sort of on the margin of missing out were very concentrated in the poorest neighborhoods, in terms of where they ended up successfully leasing up. If we've made it somewhat harder to lease up in those neighborhoods, even if we made it easier, presumably, in these higher quality neighborhoods, you know, there could still be the potential that people are missing out on successfully leasing up in the first place with a voucher. And we're mostly limited by a lack of high-quality data, to be able to study that particular element more in our paper, but it's certainly a concern.

Michael Lens 52:22
Yeah. And I mean, generally, like, that's a big conundrum that we still haven't quite solved right, where as like, you know, at some juncture, you rightfully noted that landlords, some landlords, when they see that you're paying with a voucher, they're like, "Okay, I may know, or intuit or whatever, or stereotype that this is a potentially low-income individual with perhaps inconsistent earned income. But I know that the US federal government, or the local housing authority, more specifically, is going to cut me a check and that thing is not bouncing, right?" And so like, on the one hand, you have these you have, you know, a group, a class of landlords, I mean, there's, you know, a lot written on this, it's more complicated thanI'm portraying it, but you have groups of landlords who are absolutely fine with and if not seeking out voucher holders, depending on, you know, in part because of the market that they typically, you know, rent in or rent out in. And then on the other hand, we have like more discriminatory landlords that specifically don't allow it, or just generally, for a host of reasons that we don't fully understand very well I don't think, because Rob notes data, like, we don't know why, or there's a group of people or a group of voucher holders that just don't successfully use the voucher to lease a unit right? And so, on the one hand, it's like, "oh, it should be easier", or at least easier than like, similarly situated people, and then on the other hand, we have this hole of people that are not able to use it.

Rob Collinson 54:16
Yeah, and I think like, you know, if the policy was just getting a haircut to the landlord that you described, which I, you know, totally.... and Eva Rosen, I think has done some great like qualitative work on that (and) this group, you know, I

She was definitely in mind. Yeah.

And so, like, if we were just forcing those people to take a haircut, you know, this policy would seem like a completely unambiguous whim, probably in the minds of a lot of people. You know, obviously, I still think it's an important and a net beneficial policy change. But, you know, you might worry certainly that some households have a harder time finding housing under it.

Michael Lens 54:54
Right, right.

Shane Phillips 54:55
Just a mechanistically, how does the use of the voucher work? I'm trying to think about how landlords are actually able to increase their rents in response to these changes. So it could be you know, that they know the rents are set at this level, the fair market rents are set at a specific level. And so they set their rents when they have a vacancy at that rate, and just hope that a voucher holder takes it, even if other people maybe don't value it at that same level, that seems like a risky approach, because there are not that many voucher holders, and most units are not rented by them. Another could be they like, advertise it at a certain price. But when they get a voucher applicant, they increase it because the applicant doesn't care. Like that doesn't sound like you can do that, probably legally, at least. I'm just curious how this actually works in practice, like how are they able to discriminate in this way?

Rob Collinson 55:50
Yeah, and just to be clear, so the type we are looking at, in particular, because largely those challenges that you described, it's a little hard to figure out exactly how the mechanism would work. Sort of on average, if we were to just like, compare certain landlords rents, you know, after they get a voucher holder to other neighborhoods around it, or something that might be a somewhat fraught exercise, so what we do is we look at once you're in a program, once you basically already have a voucher tenant, and all of a sudden your area gets hit with, again, this kind of random shock to the payment standard that allows it to go up, right, the market fundamentals for that person, because, you know, we show that evidence that this increase in the payment standard isn't related to what's going on in the market, there's not a market force, sort of enabling the landlord to just increase the rents for the tenant. These are sort of things coming driven by these strange revisions, essentially, that happened to the FMR. And so what that allows us to do is say, like, from year to year, how much does the incremental rent go up, when they get this sort of big surprise increase possible in the payment standard, and, you know, mechanistically, it's important to realize like payment standards are published by the housing authority. So they, you know, if you're a landlord, and you're working with the voucher program, you can go your PHA website, and you can pull up while you're eligible for. And so I think these comparisons of how sort of year to year, voucher rents change in response to these big sort of shocks in the availability of basically more subsidy to charge more to the tenant is what we found to be sort of the the best setting that we could persuasively study it. I mean, I will say, like in the qualitative work, as I mentioned, I think like in this book, you know, she has instances where landlords, you know, have vans near a Housing Authority Office, and they might take them on a tour to their units. And so there, you're getting some potential mechanisms for sort of price discriminating. They tell them this is the advertised rent, you know, in person or by taking them there. And maybe the tenant hasn't sort of looked up to verify if that's in fact the amount charged to potential other tenants.

Michael Lens 58:05
Right.

Shane Phillips 58:06
This might be a question to both of you to an extent. But we've brought up the funding issue several times and how housing vouchers are not an entitlement, I think the figure is roughly a quarter of households who are eligible based on their income actually receive one because there's just not enough money spent on them. There's not enough vouchers to go around. How does your research on this fit into this idea of making it an entitlement, this has been proposed by the Biden administration, I think it's become more of a priority for Democrats and people more on the left, this is something where more people would have it. And so in theory, that should also affect the market writ large, and even potentially people who are, you know, just over the threshold of eligibility. What are your thoughts on how that fits into this?

Rob Collinson 58:54
Yeah, that's a great question. And just a plug a little bit of something else that I wrote not too long ago with Ingrid, Mike, and I's older advisor, and with Jens Ludwig in the Journals of Political and Social Science. We take a little bit of a sort of theoretical look at what are some of the lessons sort of that we might be able to apply, historically, and from economics in thinking about some of these proposed expansions to something like the voucher or to instead do something, maybe through the tax code that provides something, you know, closing spirit to a voucher. So I think one thing that you were hinting at which I which I think is worth talking a little bit about, which is that if we flood the market essentially with a bunch more housing subsidies, you know, if we're in markets where housing supply is pretty inelastic, meaning that it's not super responsive to changes in price, you know, places where housing supply is so much fixed, we all of a sudden give everyone a bunch more money for housing. You know, landlords might recognize this and start to jack up rents as a result, because people have sort of more demand for housing. And so one thing that I think we need to take seriously, as you know, people that are interested in the expansion of the voucher program, and you know, think that the significant amount of oversubscription and sort of the lack of resources is a problem, I think we do want to think carefully about like, how can we design the program in such a way that, you know, if there are these larger expansions and resources, that we could be doing it in a way that's thoughtful of those potential ramifications. So one thing I'll highlight that we mentioned a little bit in the paper, it actually ties in a bit too, to my own geography here. So fun fact for you both in the 70s, HUD, when they were first designing the voucher program, was concerned with precisely this issue, the idea that if you created a new housing subsidy, and you made it quite widely available to households, that the housing market might respond, or that the housing market might in some sense, fail to respond, and landlords would increase rents quite a bit without necessarily expanding supply of new housing. And a lot of the benefits of that subsidy would get captured by the landlord. And so they wanted, of course, like, you know, an economist or program evaluator would want to do, the thought experiment they would like to do is like, "let's dramatically expand housing subsidies in some markets and not others, and then study what happens to rents and all these different things". That's what they want to run, it turns out, that'd be incredibly expensive experiment, right? Because you're creating entitlements in some places. And so, they couldn't do that - this was the 70s, they couldn't do it, politically infeasible. So what they decided to do was, "let's try doing this in two markets that are pretty small ish, that are fairly different housing market conditions, and let's kind of compare what happens in those places to the sort of other markets that look similar", right? So this is kind of like the 70s version of the difference in difference or synthetic control or something like that. And so they do this, and they do it in two places. And one of them is St. Joseph County, Indiana, which is where South Bend is and where I'm sitting currently. So fun fact of history, they do this in South Bend and in Green Bay, incidentally, Green Bay as the tight housing market in their experiments. So it gives you an idea.

Michael Lens 1:02:11
Just to interject with some very important and topical stuff. These are not my favorite football teams you're talking about at all. Okay, I will shut up now, coach.We didn't even get to Brian Kelly at all.

Rob Collinson 1:02:30
So anyway, so I suspect I know you that you have some purple garment underneath somewhere. And so as a local Chicago, and I'm not, you know, partial to Green Bay either. So the housing markets, and basically what they find is a few really interesting facts. So one, the big headline is that they don't see their rents increase a lot. Now, there's really two main reasons why that's the case. And then there's a bunch more reasons they contribute to those reasons. So the two reasons that that's the case. One is that the housing markets respond, landlords sort of start upgrading units that were marginal quality to say, like, Oh, we got we there's more money out there for housing. Now, we're going to sort of take some vacant units out then and rehab it. So now we're adding to the supply of housing. So supply is somewhat elastic in this sense, which is meaning that some of that rent increase is being blunted by that. The other fact is that there's not a huge demand surge as a result of this increase in resources for housing. And that's really interesting. And it arises for a few different reasons related to the design of the subsidy. So unlike the current version of the voucher, where I pay 30% of my income, and even if I rent a unit that's $300 below the payment standard, I still pay 30% of my income for housing. So one interesting thing they did with a housing subsidy in that in this experiment, is they allow the tenant to basically keep any amount that was between essentially, what was the fair market rent at the time and the rent on the unit if that amount was less. So essentially, a tenant that elected to economize on rent would essentially get rebated some of the subsidy.

Shane Phillips 1:04:14
Wow

Rob Collinson 1:04:14
And so what that does is it gives the tenant sort of more potential interest in negotiating, being selective about particular units.

Yeah. And it gives them another channel through which like, if rents are going up, like maybe I'll just find, you know, I'll rent this one and I'll use less than the payment standard, and I'll use that money towards something else. And so I do think when we consider sort of the expansion of housing subsidies, and try to think about what are they going to be the market-wide implications of doing so, being careful about how they're designed and thinking, "are there creative ways that we can shape the subsidy that might give tenants more incentive to potentially, you know, economize if they decide that that's what they want to do", which would reduce some of the pressure on rents and reduce some of the bargaining power that landlords might have.

Michael Lens 1:05:03
Right? And then of course, like, you know, you think you could get in kind of an endless or unresolvable debate about, like, kind of paternalism versus cash....just cash

Rob Collinson 1:05:12
Exactly, yeah.

Michael Lens 1:05:15
Like some people would be very adamant that like, "well, you don't want people to, like, skip on their own housing". And yeah, I think like your utility maximizing-believing, you know, economists is like, "hey, let's let them decide like what the best, you know, housing bundle is for them", right?

Rob Collinson 1:05:38
Yeah, exactly. Yeah, I think that's right. Obviously, the reason why we don't have that current design is precisely because of, you know, a paternalistic sort of view or one that like, you know, people are not maybe making the right decisions, we need to make sure they have a minimum quality, and then the landlord has to pass HQS, and they have to do all of this stuff. And so this is providing sort of a different paradigm, but I think, is useful. It's always late, I think, beneficial to try to learn from history, sometimes history is a guide, but I think in this case, it was...

Shane Phillips 1:06:09
You can learn from it that way, too.

Rob Collinson 1:06:11
Yeah, exactly. It's important to learn from both but um,

Shane Phillips 1:06:15
To add one more, you know, beyond the sort of setting a cap, but letting people keep the difference in what they actually pay. I think this issue of having a sharp cut-off threshold is also maybe something that needs to be thought about in the future, where you don't want... you know, if you earn $30,000 a year, you get up to say, $1,200 a month to help you with rent. If you make $30,001 a year, you get nothing. That has, you know, there are political challenges with that. It's just like practical, the person who earns one more dollar is not, you know, so much better off than they don't need any assistance compared to the person who earns $1 less, there's a lot of things to consider. But that is all dependent on having enough money in the first place to make these kinds of decisions.

Rob Collinson 1:06:59
Yeah, exactly. And one of the kind of nice things I will say, so there's always a little bit of tension here for economists about sort of program design, but like the current voucher scheme is very tapered, such that as your income goes up, like your rent contribution has gone up, which, you know, economists might worry discourages work and things like that, but it doesn't have the beneficial property of it means that like, there's not these weird threshold effects whereby, you know, earn a bit more and this year, and all of a sudden, you're no longer eligible for any subsidy. Instead, you know, your subsidy is just going down a little bit because you're required to pay more rent. So the voucher design, you know, doesn't have these sharp cliffs, but some of the proposed I think tax changes and things that have been under consideration for others do maybe have some of these same features.

Michael Lens 1:07:44
Well, I am mindful of the time because, you know, our sabbaticals are ticking away here... not that I'm counting the days or anything. But I do want to make sure you get back to your life, and even if our listeners are thirsty for more. Before we go Rob, won't you tell us a little bit about what you're working on, what's coming up next?

Rob Collinson 1:08:08
Yeah, I mean, like any Assistant Professor, I got a lot of projects going right now. So I've done a lot of recent work on evictions and some of their consequences. But so I encourage people to go go look for those. But I right now I've got some work - trying to understand a bit more about this question that we talked a little bit about, public housing versus vouchers and kind of how to think about that, how to think about it from the tenants' perspective, sort of what types of choices do tenants make if they're offered one, but not the other? If they're offered both, what can we use about those decisions to learn about sort of people's valuation of these different tools? And sort of what are some of the implications of those decisions that tenants make about whether or not like should I go apply for public housing? Should I apply for vouchers? What happens if I get one offer? And for public housing and not a vouchers, what happens if I get a voucher offer not public housing offer? Sort of how does that impact who leases up in the program and sort of like does it influence the mix of who lives in public housing and who's using vouchers, and so hopefully, lots of work on the frontier on those topics.

Michael Lens 1:09:16
Very cool. I look forward to that, and more. Shane, any, any last words, anything we got?

Shane Phillips 1:09:23
I got nothing. Thanks so much, Rob.

Rob Collinson 1:09:25
Yeah, this is awesome. Thank you guys for having me.

Shane Phillips 1:09:32
That is it for our first episode of the new year. You can read more about Professor Collins and research and find our show notes and a transcript of the interview at our website lewis.ucla.edu. The UCLA Lewis Center is on Facebook and Twitter. I'm on Twitter @ShaneDPhillips, and Mike is there @MC_lens. Thanks for listening. See you in a few weeks.

Transcribed by https://otter.ai

About the Guest Speaker(s)

Robert Collinson

Rob Collinson is the Wilson Family LEO Assistant Professor in the Economics Department at the University of Notre Dame and the Wilson Sheehan Lab for Economic Opportunities (LEO). He is an applied microeconomist with research interests in housing policy, urban policy, and the design of anti-poverty programs.