Ep 84: A Review of Rent Control Research with Konstantin Kholodilin

Rent control is one of the most hotly debated housing policies, and also one of the most researched. Konstantin Kholodilin reviewed over 200 rent control studies, dating back decades and spanning six continents, and he joins us to give an overview of their results.

Show notes:

• Kholodilin, K. A. (2024). Rent control effects through the lens of empirical research: An almost complete review of the literature. Journal of Housing Economics, 101983.
• Konstantin’s massive database of rent control policies across the world: Longitudinal database of rental housing market regulations: 100+ countries over 100+ years.
• Kholodilin, K. (2020). Long-term, multicountry perspective on rental market regulations. Housing Policy Debate, 30(6), 994-1015.
• Wikipedia article on ‘kommunalka’ (communal apartment).
• Fogelson, R. M. (2013). The Great Rent Wars: New York, 1917-1929. Yale University Press.
• Episode 36 of UCLA Housing Voice on rent control in India with Sahil Gandhi and Richard Green.
• Willis, J. W. (1948). State rent-control legislation, 1946-1947. The Yale Law Journal, 57(3), 351-376.

Transcript

Shane Phillips  0:05 

Hello, this is the UCLA Housing Voice Podcast and I'm your host, Shane Phillips. This week we're joined by Konstantin Kholodilin to share findings from a wide-ranging review of over 200 rent control studies spanning six continents and many decades. 

Rent control can be a challenging topic to cover because it has so many consequences beyond its intended purpose of lowering rents or slowing their growth. Some of those consequences are good and some are bad, most are unintended, and individual studies tend to focus on just one or two of them at a time. But if your goal is understanding whether rent control is worth it, all things considered, you ideally want to take a wider view. And Konstantin's review of the literature offers exactly that. Rent control also might be the policy I hear the most visceral responses to, both for and against, and most often without much understanding of its history or how it's typically implemented. I don't ever expect conversations like this to completely change anyone's mind. But whether you count yourself for or against rent control, I hope this overview complicates your views on it. Really, that's all an academic institution could ever ask for. 

The Housing Voice podcast is a production of the UCLA Lewis Center for Regional Policy Studies with production support from Claudia Bustamante, Irene Marie Cruise, and Tiffany Liu. 

If you appreciate the show, be sure to tell your friends and colleagues about us and send your questions and feedback to shanephilipsatucla.edu. With that, let's get to our conversation with Konstantin. 

Konstantin Kholodilin obtained his PhD in Barcelona and is a doctor habilitatis and senior researcher at DIW Berlin, and he is here to talk with us about rent control, the many, many empirical studies that have been done on it over the decades, and what those studies have tended to find. Konstantin, thanks for joining us and welcome to the Housing Voice podcast. 

Konstantin Kholodilin  2:15 

Thank you, Shane. I'm happy to talk with you. 

Shane Phillips  2:17 

And Mike is my co-host. Hey, Mike. 

Mike Manville  2:19 

Hey, good to be with you both, you guys. 

Shane Phillips  2:21 

So usually we start with a question about a tour, but Konstantin, I'm actually first going to ask you about the doctor habilitatis. What does that mean? 

Konstantin Kholodilin  2:31 

Well, if I understand it correctly, it seems to be something peculiar for Central Europe, where, starting probably from 19th century, but I'm not sure, there were two degrees. The first one, candidate degree, and the second, doctoral degree. And this doctor habilitatis is actually a doctoral degree that enables you to become a professor. In Germany, they used to have it, and in the past it had major importance, but nowadays it's just a kind of ceremonial title. 

Shane Phillips  3:00 

Well, this is news to me. I've learned something new today. Okay. Well, now we can get to our tour. So I didn't ask you about this beforehand, but where would you like to lead our audience around and tell us a few things about? 

Konstantin Kholodilin  3:10 

Well, I would love to lead you on a virtual tour to my hometown, to St. Petersburg in Russia. And I haven't been, unfortunately, there for almost five years, so that's one reason more to go there, at least virtually. And I'd like to talk about three things. The first thing is the beauty of this city. The second thing is somewhat related to urbanism, namely height restrictions. And the third thing is something very specific to St. Petersburg and probably Moscow, too, is Komenalkas. As a tour guide, I would like to praise the beauty of this city, about natural beauty, because it has dozens of islands, it has a very wide river, it has a sea, and it has wide nights. So starting from May till mid-July, we almost don't have nights there, so you can read books during the night. And it also has architectural beauty, because almost from the very beginning it was conceived as a capital of a huge empire. Also the authorities, the tsars, or emperors, they invested a lot of money in the improvement of this city, and so a kind of ensemble emerged. So it's a kind of combination of natural beauty and architectural beauty that makes it very special. So you have palaces, you have parks, you have bridges, some of the bridges open in the middle of the night to let big ships go down and up the stream. And if it's in the white night period, then it's especially nice. And since we started to talk about architecture, one of the things that was adopted in Saint Petersburg was the height restriction. So the highest building had to be, no surprise, the Winter Palace, which was the residence of the emperors. No other building could exceed that one. And so over the centuries, it shaped the center of the city. So in the center of the city, you don't have high-rise buildings. So probably the highest is 40 or 60. Well, if it's a cathedral, it can be a bit higher. But if it's some multifamily house, then it can't be higher than 40 or 50 meters. While on the periphery you have very high-rise buildings. So in the last 30 years they built houses with 30 or 35 stories. So it's something like an upside-down bell. If you compare it to a typical American city, then it's just the opposite because in a typical American city you have very high-rise buildings in the center and then very low-rise buildings on the periphery. In Saint Petersburg it's just the other way around. So the last thing I'd like to mention, and it's, as I said, very specific to Saint Petersburg is common alcas, singular common alca, or communal apartment. It's very personal in the sense that in the late 1910s, many apartments in the city center were nationalized. And so imagine a five-room apartment, which used to be occupied by one family. From then on, it was occupied by several families. In the extreme case, five families. So one room, one family. They shared bathroom, they shared kitchen and the corridor. And the interesting thing is that there are hundreds of thousands of such dwellings. Many of them exist. In the 90s, most of them were privatized, but they were privatized also in a peculiar way, because the people were allowed to privatize the rooms where lived. So not even apartments. This is a very extreme form of homeownership. It's room ownership. So you could be owner of a room in an apartment shared with other people. And that's one thing I never, never faced, never encountered in other countries. 

Shane Phillips  6:59 

Yeah. Kind of reminds me of like the tenancy in common or this kind of thing where you sort of have a share in a building, but it's also sort of like a condo where it is, you probably just own the unit itself and everything else is kind of, is sort of shared. That's a very, I don't know how you would. Who owns the hallway, the kitchen, the bathroom? Is that each person it's like owned in common between the people who own the room or each room in the single unit? 

Konstantin Kholodilin  7:25 

Exactly. It's a condominium with a fractalization of condo ownership. Yeah, exactly. So it's ad absurdum. So when you start with house ownership, then you go to condominium, which actually is not very old thing because in many European and American countries, the apartment homeownership originated in the 50s, 60s, 70s. So it's pretty young phenomenon. And here we have an extreme form and another pole of the range of the spectrum.  

Shane Phillips  7:57 

Yeah, that's super interesting. Okay. So our conversation today is about an article written by Konstantin and published in the Journal of Housing Economics earlier this year with the following title, Rent Control Effects Through the Lens of Empirical Research, an Almost Complete Review of the Literature. I think that title says just about all you need to know, but I will say just a little bit more before turning it over to our guest. First is just to note the breadth of this review, which started with identifying 206 different studies on rent control of which 112 were published empirical studies, going back all the way to 1967 and covering every continent except Antarctica. He identified and categorized 26 different kinds of effects investigated in these studies, ranging from the most common topics like the effects on controlled and uncontrolled rents to less commonly investigated things like its effects on eviction, tax revenues, and even marriage rates. The article does a great job, I think, of presenting the theory behind how rent control affects rents and mobility and housing supply and all the rest, including why it might be expected to have a positive or negative effect on each or sometimes positive effects for one reason and negative effects for another that balance out to some extent. And then it presents in a straightforward way what the empirical studies have actually found, including a great figure that summarizes it all very clearly. And although it is probably difficult to walk away from this literature review feeling like rent control is definitely a great policy where the benefits always outweigh the costs, I appreciate that it offers a lot of nuance and arguments on both sides of this debate or conversation and ultimately leaves it to the reader to decide where they land on it. So I guess you could say I am a fan. To get us started, Konstantin, most of our listeners probably have an idea of what rent control is, and some will even have a version of it where they live, including us here in Los Angeles, which has its own version, and the state of California has a kind of lighter touch version that covers the entire state. But for those of us who don't know as much about this or may have some misconceptions, let's just make sure we're all on the same page. What is rent control? What are its intended outcomes, and how does it try to accomplish them? 

Konstantin Kholodilin  10:13 

Well, rent control is a specific case of price control. So you can control housing rents, you can control prices for bread, you can control prices for fuel. In this particular case, the intended goal is to either freeze housing rents, so prohibit them from growing, or to cap the rent increases. And the idea here is to improve the affordability. Typically, we measure the affordability of housing by rental burden, which is the ratio between the rental costs and households' disposable income. So when the share of rental costs in the income is very high, say above 30%, then you are in trouble. Anyway, you have to pay for your housing. You can cut your food costs, you can cut your cloth costs, you can cut probably your transportation costs, although it's not always possible, but you cannot really cut on your housing costs.  

Shane Phillips  11:11 

Just to be clear there, you just mean because you can't rent out one room in your home for a month or something. It's just a single rent, and there's no downsizing in the near term, whereas you can maybe make fewer trips, you can buy a little less food or cheaper food, but you can't say, this month I'm going to consume less housing and only pay 80% of the rent. 

Konstantin Kholodilin  11:33 

Yeah, sure. The rent is fixed in case of rent control, it's fixed from above. In case of no rent control, it's fixed from below. Even if you consume just one room in your apartment, in your dwelling, in your house, you still have to pay for the entire apartment. 

Mike Manville  11:50 

It's hard to consume marginally less housing to save some money the way you can, as Shane said, I'm not going to buy from the bottom shelf of the supermarket for a month or I'm going to, like a college student, not do my laundry for a month. You can't do that with your rent, it's close to an indivisibility. 

Konstantin Kholodilin  12:09 

If you control the rent, if you control the numerate of this rental burden, then you can increase the affordability because if the income doesn't change or increases, then the whole thing is declining. So you say from 40% rental burden, you can go down to 30% or even 20%. Just one example, for instance, in socialist countries like in Soviet Russia or in socialist Germany, German Democratic Republic, they used to control rents all the way from 1930s till 1990. And by the end, this rental burden was just five, sometimes even 2%. So that's the effect of freezing rents for four decades, for example.  

Shane Phillips  12:57 

Okay. So that's actually a great transition to talking a little bit about the history of rent control. And Konstantin, I know you know the history going back a very long way. So maybe we can start with the prehistory as it were, and maybe you bring us up to, in just a short time, the 1940s roughly, and Mike can pick up there for the history in the US since that time. 

Konstantin Kholodilin  13:23 

Yes, sure. It turns out that rent control is older than many of us believe. In fact, one of the first documented cases dates back to 4950 before Christ. It was the times of Julius Caesar. And at that time, well, at that time, he wasn't temporary yet, but at that time, rents were frozen. Actually, the housing market in ancient Rome was very similar to housing market in Los Angeles or New York City because it was composed of big multifamily houses where rooms were rented. And you can imagine that rental burden was pretty high at that time. There are no estimates, but probably it was pretty high. In several cases, the emperors introduced rent control. They froze rents at a very low level for a year or so. We also find examples of rent control in Song dynasty China in 12th century. We find further examples in Europe, like in Madrid in 16th century, and it was related to the move of a capital from Valladolid to Madrid. So at that time, Madrid was pretty insignificant town. And then all of a sudden, its population tripled and quadrupled, leading to excess demand and to rising rents and to the government introducing rent control. We find plenty of examples of rent control related to founding of a university. So in the middle ages, starting from 12th century, many universities were founded in Europe. And in many cases, a committee was created formed by the representatives of the university and of the city that had to set fair rents. So the rents to be paid by tenants and by professors. You have also many other episodes of rent control, for example, protection of Jews in Jewish ghettos. So Jews had no right to be homeowners, so they only could be tenants. And in most cases, they had to live in specific places called ghettos. And in these places, rents were frozen. And actually, they remained frozen for a pretty long time, until more or less 19th century, mid 19th century. By the way, in Madrid, in the case I mentioned, rent control existed from approximately 1565 to 1842. So almost 300 years. You can imagine how long this… Yeah, longer than we've had a country. And then we come actually to the 20th century, because before 20th century, we have specific cities subject to rent control, in some cases, specific neighborhoods, specific population groups. And it's only in 20th century that there was a true explosion of the number of countries that introduced rent control, and it became a mass phenomenon. So it covered entire cities, millions of tenants were protected by rent control. And it became a very large scale phenomenon. If we want to say a few words about the US, I should probably break one of your illusions, if I may say so, because we find the evidence of rent control in the US already in 1919. So in the District of Columbia, rent control was introduced. And then later on, either in 1919 or in 1920, rent control was introduced also in New York City. There's a nice book on it by Professor Fogel. But it was typically confined to specific cities. So Washington or New York City, and the true nationwide rent control was introduced only during World War II. 

Shane Phillips  17:04 

Yeah. And what I'm hearing is, it sounds like a lot of these cases are times when there was a sudden influx of a lot of people, a lot of demand. The capital moves, the university opens, even these new countries forming, there's probably lots of urbanization occurring as a result, or just because of the time of technology shifts and everything happening. So I'm sure there's more to say about that, including during wartime and in the post-war period. But I'll hand it over to Mike in a second. But did you have something else to add there, Konstantin? 

Konstantin Kholodilin  17:36 

Yeah. I just wanted to add that in the case of New York City in 1920, it was not an influx of people. Well, there might be some influx related to workers coming to the defense industry enterprises. But interestingly, it was the lack of construction. So for four years, during whole World War I, there was no construction in New York City, and it was enough to create housing shortage. 

Shane Phillips  18:03 

Yeah. Mike, you want to pick it up from there? 

Mike Manville  18:06 

Yeah. I think, as Konstantin points out, you had some starts and stops before World War II, and then the entry into World War II led to the US's only experiment with just essentially nationwide rent control that was actually run by the federal government. This was a product of a combination of factors similar to 1917 to 1920 in New York during the Great Depression. Very little construction happened in the United States, and then suddenly, with the entry into war, the economy roars to life. There's a huge risk of inflation, and there's a massive internal migration because people who previously had been out of work are now seeing that they can, or in the case of people who've been drafted, they're being forced to move toward these designated war production centers, which start off just being a series of large cities on the coast, but then eventually become spread all over the country as the war effort ramps up. There is, of course, no way at that time to ramp up construction because the government has nationalized most industry, and so wood and nails and labor and so forth are under the control of the national government and severely rationed. As a part of that, I think people forget this. It's a long time ago, and it seems so foreign to us, but the US placed price controls on virtually everything. It had a gigantic office under the executive called the Office of Price Administration. It was run by men who later became a very well-known economist, John Kenneth Galbraith, and as part of that, they put in place rent controls. They went county by county with it, and as a county was selected to have a rent control, the government would come in. It would roll back the rents to a particular pre-war date and then just freeze them. These applied to all rental housing. It applied to any – There wasn't much new construction that was going to happen anyways, but if new construction did happen, it was rent controlled. It happened regardless of turnover. These are what we classically call hard rent controls or perhaps a little bit inaccurately given that this goes back to pre-Jesus times, first-generation rent controls in the US. They were very comprehensive, but they didn't last very long. By 1949, Congress had allowed individual cities to get rid of the rent controls they had if they so chose, and all of them except New York did. Even New York ended up adopting a hybrid version, and then rent control was dormant in most of the country until the late 1960s and early 1970s when you started to see, again, a large amount of inflation, construction slowing down, a little bit more experimentation again on the national side with price controls. Richard Nixon famously froze rents and stuff for a while, and that momentum was picked up by tenant activists in a handful of US cities who passed a series of rent control laws. Because they were local, they all differed from each other a little bit, but most of them became or were at that time what we call the soft rent controls where you had the vacancy decontrol provision, the unit turns over, the rent floats to a market rate. You had allowable increases that were set by a local rent control board, and you had an exemption from new construction. If you live in a rent controlled unit in Los Angeles today, as I do, your rent is set by a combination of what the rent control board decides and how long you've been in the unit. That's the classic American soft rent control, which is where we are at today. 

Shane Phillips  21:34 

Yeah, and in World War II, there's clearly a situation where you have rapidly increasing demand in certain cities and an inability to respond to that increased demand with more supply because all production of any kind is really being directed toward the war effort. It's interesting actually that this emerges again in the 60s and 70s when we're starting to... I don't know, maybe this predates it a little bit, but I'm thinking about the stagflation era where we have high inflation, the job market's kind of a mess, and so I would imagine that was also affecting supply. There was also down zoning and things that were just regulation was maybe pushing downward on supply, but there's also just macroeconomic things going on where maybe even if development was legal and it was relatively easy to get a project approved, perhaps getting the financing and because of things going on in the economy, it might have been difficult to build the housing people needed where they needed it. 

Mike Manville  22:33 

I think that's right, and I think this is in some ways a whole other subject about just the changing political economy of development in the US, but like you said, in the 1940s, it was just literally this emergency set of price controls, rent control happened to be one of them. In the 1970s, you had even a Republican president fiddling with price controls, and so they seemed acceptable, but then you also saw the people who pushed for rent control at the local level, particularly in California, but in Massachusetts as well, they started to sound explicitly against, there were people explicitly against development. The Santa Monica Coalition for Renters' Rights, which was sort of a bellwether and pushed rent control in Santa Monica, they explicitly said they don't like new supply. Your point is that this could have been paired with more supply in some counterfactual world, and I think that's probably true, but there is a fact that the proponents for different reasons became very suspicious of new supply at the same time that they put their faith in rent control. 

Shane Phillips  23:35 

Yeah, yeah. Well, I think that's probably enough history. We will likely touch on a few other elements of the history in our conversation, especially explaining a lot of the effects, but I do want to get into these effects because that is the heart of Constantine's paper. Constantine, I mentioned that you identified 26 different effects of rent control that had been studied, and I should note that most of these effects are unintended consequences, not really core to the purpose or purposes of rent control. Most of the categories did not have that many studies behind them, and you limited your review to effects that were investigated in at least half a dozen published studies, which left you with eight topics. We're going to do our best to go through each, but let me just first list them and then we'll get moving. In order from most studies to least, they are rents in the controlled housing stock, mobility, home ownership, construction, housing quality, rents in the uncontrolled housing stock, housing supply, and misallocation of housing. First up, let's talk about controlled and uncontrolled rents together. On this, the empirical research is conclusive. Rent control lowers rents in the controlled stock and raises rents in the uncontrolled stock. I don't think we need to do much to explain how it lowers rents in the controlled sector, but how does it increase rents for uncontrolled units? 

Konstantin Kholodilin  24:56 

Well, I probably provide an example, a recent example from Germany, from Berlin, because in 2020 in March, actually a few weeks after the pandemic started, a pretty strict form of rent control was introduced in Berlin. Typically, in Europe, rent control is the matter of the national government, but sometimes, and we've seen it twice in the last couple of years, subnational governments tried to get in this area. So Berlin decided, although it wasn't really clear whether it was compatible with the constitution, and one year after the introduction of rent control, actually this law was overthrown by the constitutional court. But from the very beginning, it was this uncertainty. Berlin's government wasn't really sure that it's constitutional or not, and still they introduced this rent freeze. In Germany, it's called Mittendeckle. Mitte means rent, dekel means something like cap, so rental cap. And it was a pretty extreme form of rent control, so something which I could denote as first generation rent control. So the rents for the first two years, the rents were frozen. They were fixed depending on the construction year and equipment at three to 10 euros per square meter per month. Imagine if it's 100 square meters. I'm not very good with feet, so it's about 1000 square feet. So if you have 1000 square feet apartment, and if the rent is fixed at three euros per square meter, then you will have to pay 300 euros a month, which is ridiculous to go with. It's not bad. Sign me up. And okay, it's very nice for a tenant, but not that nice for a landlord. In some cases, the actual rent was much higher than this legal rent. So at some point, landlords had even to reduce their rent because you could hardly find an apartment costing 300 euros a month. So typically it was the difference between legal price, this price set in the law, and the actual market price was at least 20%, 30%, 50%. So starting from 2022, the idea was that the rents would increase maximally at 1.3% a year. So imagine that if they kept this law, so in this counterfactual world, the law was not overthrown by the constitutional court, and if it were kept for a longer time, imagine inflation rates in 2022 and the maximum increase of rents of 1.3% a year. So it was really a rent freeze. So we did a study on the effects of rent control, and we found these two effects. So rents for controlled dwellings effectively decreased, but there were two exemptions in the law. So on the one hand, the newly built housing was exempted from rent control, and the newly built housing is the one that was completed starting from 2014. Plus this regulation was confined to Berlin's territory only. So all the settlements outside of Berlin, they were not subject to this strict form of rent control. And that was the variation that we took advantage of. So we looked at the dynamics of rents for newly built housing and for the housing outside of Berlin, but not far away from Berlin. So in Berlin, you have a pretty well developed network of railroads that connect city center to the outside areas. So we took those areas that are well connected to the city center. 

Shane Phillips  28:47 

So you can compare new housing to old housing in Berlin and old housing to old housing that is similarly connected outside of Berlin. 

Konstantin Kholodilin  28:56 

Yeah, exactly. So we found that these uncontrolled dwellings, so newly built dwellings in Berlin, or all types of dwellings outside of Berlin, they experienced a rent increase, which is far above the increase that we could have observed in the absence of rent control. And our explanation was that, well, there are two things probably that might be interesting to you. The first thing is that in Berlin, 95% of housing were subject to rent control. So only 5% of dwellings were not subject. These were dwellings built since 2014. So the vast majority of housing was under rent control. And it implies that if rent decrease for controlled dwellings was say minus 20%, then 95% of tenants would enjoy this decrease in rents. And in Berlin, by the way, the majority of households are tenant households, like 87% of all Berlin's households are renting apartments. So homeowners are just 13%. So the vast majority of Berlin's population was enjoying the decline in rents, while just 5% of all dwellings were experiencing rent increases. So if you compare the negative effect on controlled rents to the positive effect on uncontrolled rents, then of course, the first effect is by much outweighing the second effect. So the net effect would be rent decrease. But over the time you could expect that first new dwellings are constructed that are not subject to rent control. 

Shane Phillips  31:10 

Old dwellings can be demolished because over the time they get into bad shape. And landlords could convert the rental dwellings into condominiums. And actually we observed it in the real data because in Berlin, if you want to convert your rental house into condominium, you have to report it, you have to obtain a permission. So there's statistics on that. And we observed that immediately after introduction of rent control, the number of applications for these permits increased dramatically. So this would also lead to a reduction of the segment which is subject to rent control. Yeah, I guess there's sort of a distinction here where in a way, the stronger the price control, the stronger the rent control, the more it does immediately. And Berlin's seem to be very, very strong right away. Whereas a lot of rent controls are just, you know, we're going to limit how much you can raise rents each year, maybe less than would normally happen. And so over time, rents will go down and sometimes that will lead landlords to reduce production. Sometimes that will lead people to hold on to a unit even though their household has shrunk and they're kind of over consuming and it's leaving less for other people. Maybe less construction is happening for one reason or another. So there's just less housing going on the market as demand keeps growing. All of those things can have an effect over time. But it sounds like in this case, in part, just because it was such a strict form of rent control, you saw more of a response in the uncontrolled sector more quickly. Was that the rhetorical question? No, no. It's a real question. Yes. Yes, I think. But anyway, I mean, the astonishing thing that I discovered in my literature view was that, for example, the effect on rent, on controlled rents and on uncontrolled rents is pretty the same regardless of whether you take the first generation or second generation, whether you take the USA or Germany, whether you take 1920s or 2020s. So that's the amazing thing that you pull together different episodes and the effect is the same. There might be differences in terms of the quantity, but the qualitative effect is pretty much the same, at least on these two effects of rent control. So I'm going to intentionally ask a naive question. And part of the reason I think it's a good question to ask right now is, at the time we're recording this, it's actually the day before the election. It's November 4th, and we're going to be voting for president and other things, but we're also going to be voting here in California on something called Proposition 33, which would repeal a state law that restricts the kinds of rent control that cities can adopt. And that would allow cities to apply rent control to new housing, among other things. Right now, they're limited to units built before 1995. They're limited to multifamily housing. There's various restrictions. You can't have vacancy control, which we talked about. But just thinking about being able to apply this to newer units and different kinds of units, if rent control only increases rents in the uncontrolled stock, why don't we just apply it to all rental units? Would that not solve the problem? 

Konstantin Kholodilin  33:54 

I think it would even exacerbate the problem because the only reason why this exception is made is because the lawmakers do not want to incentivize construction. So imagine you introduce rent control for all types of dwellings, new and existing ones. Okay. If you already possess existing units, there are not so many things to do with them. So you can probably give up on maintenance. You can try to convert your housing into condominiums or in some cases into non-residential purposes. Like you can convert a dwelling into a doctor's cabinet or a shop if it's allowed because not always, but you often have legislation in place that prohibits you from doing that. You can try to demolish the dwelling again if that's not prohibited by the law. But if we are talking about housing, which has not been yet constructed, so what can you do? You can say, okay, I'm not going to construct because from the very beginning, my rental revenue is limited. So I cannot adjust my rent to the increasing costs, to the increasing prices. So why should I do this in the first place? Why should I bother building something? So I would better invest in something else than letting my money disappear in the new housing construction. 

Mike Manville  35:15 

Yeah. I mean, I think, like you said, Shane, it's sort of an intentionally naive question. It seems to make sense. Well, let's just make all the housing, have the law that makes it cheap. But an additional thing happens when you do something close to that, which is that you get a lot of black markets, which is to say that yes, the legal price of this housing is only $500 a month, and if you look at what people are actually paying, because it doesn't do anything to change the demand for housing, and it's very hard. You can regulate some things that are easy to see, like, oh, this person is taking this building down and demolishing it. You can't regulate the fact that the landlord just expects some sort of bribe that brings it much closer to a market rate. And so you have this difference between the statutory rent and the rent people are actually paying, and that sort of undermines the affordability goal. But I think a variation of your question was addressed a long time ago in the rent control literature. Walter Block, who's a notorious opponent of rent control, and so he's not the most unbiased person in the world, but he had this nice quip where he said, if you really want to use a price control to lower rents, what you have to do is apply a price control to everything except rental housing. Because then everything except rental housing will have a lousy rate of return, and all the capital will flow into building apartments, and you'll have an abundance of supply. But it's the opposite of saying that you want to just put a price control on rental housing. 

Shane Phillips  36:36 

Yeah, that makes sense. 

Konstantin Kholodilin  36:37 

If I may add, so we know from the history, just the opposite example. So in Brazil, in 1920s, 1930s, the government was very concerned because most rich people prefer to invest in real estate because it was a secure investment. And so they didn't really invest into industry while the government wanted to industrialize the country. So what they did was to introduce rent control. So immediately all this capital flew into other industries. So they had industrialization. When people have a choice about where they put their money and you just tell them, you can know for sure you're going to get a lower rate of return in this sector, they're just not going to invest there. If we had a problem of an oversupply of housing, that might be a very rational response. But yeah, that is not the position we find ourselves in across really anywhere in the US at this point. And maybe another historical example, it's from 1920s from Germany. So they introduced rent control after World War I. And then you could see it in the numbers, the amount of housing completions decreased almost to zero. What the government had to do was to introduce subsidies to incentivize housing construction. So instead of trying to cooperate with private investors, probably helping them a little bit to construct more, the government had to take over the whole exercise, the whole task of new housing construction. And that can be a vicious circle. So you stayed regulating at one point and then you cannot stop because each time you see that your regulation leads to some problem, then you have to solve this problem. Therefore you introduce regulation at another point and it's a never-ending story. So you will always have to regulate at a new point in order to solve the problems created by the older regulation. 

Shane Phillips  38:37 

Well, we've talked about housing supply and construction a good amount so far, but let's actually talk about the effects that you found. Eleven studies found a negative effect on housing construction, four found no effect, and only one a positive effect. That's I think pretty conclusive, but not as one-sided as the findings for controlled and uncontrolled rents or the findings on housing supply in which nine out of 10 published studies find that rent control has a negative effect and the 10th finds no effect. First, just what is the distinction that's being made here between construction and supply? And could you also explain how rent control is understood to reduce housing supply and rental housing in particular? 

Konstantin Kholodilin  39:21 

One of the problems of such literature review is that different authors define the dependent variable in different ways. They give different names and sometimes it's difficult to find a common denominator. So I prefer to consider separately studies on new housing construction, but even in this case there are different measures. There are housing starts, there are building permits, there are housing completions, which of course it's not the same. It's something similar, but it's not always the same. In case of supply, I'm looking at the housing stock, but again here you can find differences between total housing stock, stock of rental dwellings, stock of unoccupied dwellings. So it's not very homogeneous as a group of indicators. 

Shane Phillips  40:09 

And how does rent control or how is it understood to reduce the housing supply? 

Mike Manville  40:14 

How could it reduce supply in a situation where it didn't reduce construction? 

Konstantin Kholodilin  40:18 

Okay, yeah. So the most evident case, of course, is when the construction is reduced and supply over the time will go down too. But still, even if a new construction is not reduced, you can have a reduction in housing supply. Because, for example, you can convert your housing to non-residential purposes. If you are focusing only on rental housing, then again conversions to condominiums can reduce this rental part. And another possibility is to demolish dwellings. So even if they didn't serve the whole time that they were supposed to serve, they can be demolished to free space for something new, which is not subject to any form of control and which can bring you more money. 

Shane Phillips  41:02 

Mm-hmm. 

Mike Manville  41:03 

And there's many examples of this throughout the US history of rent control and also internationally. In an earlier episode, we had Richard Green and his co-author on talking about the relationship between rent control laws in India and the higher levels of vacancy that occur as a result of it. And closer to home, when Cambridge, Massachusetts, their rent control law ended in the early 1990s, by the time that happened, landlords there had held thousands of units off the market. They had a hard rent control law simply because it was less expensive to keep them vacant than to have someone in them and have to pay the operating costs and so forth. 

Shane Phillips  41:41 

One other case I think is important is the case I find myself in, which is I live in a rent controlled multifamily building. It's a duplex, but I bought it seven years ago from someone who owned the building and lived in one unit and rented out the other, and that is what I continued to do. So at some point, both units in this building were probably rented, but over time it has just become more valuable, a lot of these units are more valuable to someone who is willing to live in it as an owner than it is to someone who would purchase it to rent it out. Like I, as a potential owner, am willing to pay more than someone would be willing to rent it out in part because of the restrictions they have on who they can rent it to and how much they can rent it for. And in Los Angeles, there's something like seven or 800,000 units that are at least potentially eligible for rent control, I believe, multifamily, built before 1979, et cetera. I think at least 100,000 of those today are owner occupied, and these were not condo conversions, they're just rental units that are occupied by the owner. Okay, so we have controlled rents, uncontrolled rents, housing construction, housing supply. Next is home ownership. Most studies on home ownership have found that rent control increases the ownership rate, but a sizable minority finds a decrease. What's going on there? What do you think explains these results? Why would rent control increase the ownership rate? I guess I kind of just answered it, didn't I? 

Konstantin Kholodilin  43:12 

Yeah. In fact, all our previous discussion contributed to answering this question, but to summarize, it's much easier to explain why rent control can increase home ownership. And the reason is that, well, there are several reasons. One reason is that landlords are not satisfied with the reduced rental revenue, so they try to get rid of this housing. They convert it, for example, to condominiums, they sell it to homeowners, and this reduces the proportion of tenants. On the other hand, imagine that you are a newcomer, you're coming to a new city, so you got a job there, but you cannot get access to the control dwelling because the mobility there is very low, because people are not eager to leave these apartments because the conditions there, at least financial conditions there, are very attractive. So you either have to rent an uncontrolled dwelling, which costs much more expensive than a controlled dwelling, or you have to buy an owner-occupied house or dwelling, even if it implies that your financial burden would be too high. So under other conditions, you wouldn't think about buying housing. So all these leads to homeownership rate increases in the wake of introduction of rent control. It's much more difficult to explain why it should lead to decreased homeownership. So today I looked once again through the literature that finds negative effects of rent control on homeownership rates, and I didn't really find, except one, two convincing explanations. One explanation was that the tenants who live in controlled dwellings find them so attractive that they don't want to bother buying homeownership, because they have everything. If you are a homeowner, you have to pay interest, you have to take care of the housing, and not everybody wants it. Plus it will be probably outside of the city center, outside of the CBD, while if you are living in the controlled unit, it's somewhere close to the city center, the price is ridiculously low. So why care?  

Mike Manville  45:26 

Yeah, I think that's the explanation that I've heard the most too, is that it's sort of for a select group of tenants that can interrupt the housing lifecycle as it normally happens. You don't graduate to homeownership, and Richard Arnott phrased this as, you know, one thing rent control does is it transfers more property rights to the tenants, and they can actually behave like homeowners and start dramatically improving the unit if they want, comfortable in the knowledge that they're not going to have to leave. And this all assumes, as Constantine points out, that they also have a taste for the type of neighborhood their unit's in. And so we might expect this to happen in Berkeley and Santa Monica and in parts of Manhattan and maybe not happen in LA's rent controlled stock because, you know, some of the nicer neighborhoods with better schools and so forth in Los Angeles tend to be owner occupied. But yeah, I think it's a very particular set of circumstances where it would actually do that. 

Shane Phillips  46:19 

Yeah. Next up, we have housing quality on which most studies find a negative effect. Rent control tends to be associated with lower quality housing. You discussed this a bit in your article, but there's a question of whether lower quality housing is really such a problem if the lower quality units are also more affordable. People are free to leave rent controlled units when they like, and so the ones who stay seem to be indicating a preference for lower quality, lower rent housing if the alternative is higher quality but also higher rent housing. There are of course different degrees of low quality and a shabby unit is qualitatively different from a dangerous or uninhabitable unit. But I think when critics of rent control talk about housing quality, they can tend to gloss over how the quality of many non rent controlled units is higher because the rents are also allowed to be higher. If you have to choose between a higher quality unit you can't afford and a lower quality unit you can, then that's going to be a pretty easy choice for most folks or not even really a choice at all. People in that position are just one segment of the renter population though, and another is people who might like somewhat higher quality housing and cannot find it. But I'm curious how you think about this one because it seems a little more ambiguous about what lower quality means, whether it sounds at first plus just like a bad thing, but that doesn't seem to necessarily be the case. 

Konstantin Kholodilin  47:51 

Well, three things. First of all, I agree that your point of view is completely justifiable. So if people prefer to live in poor quality housing because it's less expensive, why not? But one thing that I find important is that it's a kind of, for me, it's a kind of inefficiency because you build the housing of a higher quality in order for it to decay over time and become lower quality. So why didn't you build from the very beginning a lower quality housing? And the third thing, the main message that I'm interested in here is that it's yet another proof of the fact that nothing comes at a zero price. So when rent control is introduced, the idea is that people pay for higher quality, lower price. But over the time, this difference disappears because due to lack of maintenance of improvements, the real rent based on the quality characteristics of the housing converges to the real rent that is being paid. So you cannot overcome this.  

Mike Manville  49:02 

I think that's a really important point. I think in part, the emphasis that is sometimes placed on quality is advanced as an important qualifier to the observed differences in rent. Because it's very hard in your standard comparison to fully control for the quality of the housing, especially since it can vary so much unit to unit within a rent-controlled building. And so it's important to understand that even if we see an 8% reduction in rent or something between controlled and uncontrolled housing, some of that is probably actually just lower quality. And to your point, Shane, some people may be like, I'm fine with that. But I think in the kind of rent control we have in the United States with the vacancy decontrol, it's also worth remembering that, as Konstantin points out, the quality is going to decline on average as income goes up. The longer you stay in the unit, that's when it becomes less and less affordable for the landlord or less and less desirable, however you want to put it, for the landlord to maintain that unit. Because in real terms, they're earning less and less money from you. But for most people, there's going to be some people it's not the case, some elderly people, things like that. For most people, their earning power is going to be rising at that time. And so that suggests that maybe this idea that they just want lower and lower quality as time goes on doesn't make as much sense for most of the renting population. 

Shane Phillips  50:22 

Yeah, that's interesting that your income and your housing quality are trending in different directions. 

Mike Manville  50:26 

Well, as Konstantin points out, low quality housing can be incredibly important for a city that has a lot of low income people. But ideally, you would meet that need from the beginning rather than say, well, hang around for a while and your housing quality will get worse and then the price will go down. But of course, by that time, if things are working well, they'll have more money.  

Shane Phillips  50:45 

Yeah, yeah. I think it's also important maybe to think about this beyond the boundaries of a single city with rent control. And if we compare Los Angeles to Phoenix or Houston or something, the amount we are paying for the quality of housing we're getting, rent control or not, is quite poor. We're paying a lot of money for quite old housing that even when it's newly renovated, it's still 50 years old, it's structurally unsound in some cases, just lots of problems that is partly just because we've built very little over the ensuing 50 years. Right. 

Mike Manville  51:22 

I mean, there's only so much you can do with a 1940s building in terms of how well does the disposal work, how well can it accommodate laundry and unit and so forth, things that people really like and it's just, you know, they weren't made for that sort of thing. 

Shane Phillips  51:37 

Next up is mobility and misallocation, which I think makes sense being discussed together as well. Twenty published studies find rent control has a negative effect on household mobility, just one finds no effect. This is another one that has both positive and negative explanations for why mobility is lower. Could you summarize those different perspectives on reduced mobility and do you know of any studies that have tried to distinguish, quote unquote, good e-mobility from bad? 

Konstantin Kholodilin  52:04 

Well, typically people either are looking at the good e-mobility or bad e-mobility. So maybe I'm mistaken, but I can't think now of a study that discusses both of them. But anyway, let's start from the bad e-mobility. So bad e-mobility means that you are staying longer time in the same place and you are not reacting to, for example, to the labor market shock. So if this wage increase somewhere outside of your city, you don't want to go there because you are afraid that you will have problems with affordability of housing, you won't find a nice rent control dwelling. The place you have is already good for you, so you simply don't react to these wage shocks. Now the good e-mobility, at least from my point of view, is that you stay longer time in the same place. You can reformulate it as residential stability. It sounds a little bit nicer than lack of mobility.  

Shane Phillips  53:03 

One man's e-mobility is another man's residential stability or housing stability. 

Konstantin Kholodilin  53:07 

So when you leave for a long time in the same neighborhood, you accumulate social capital there. So you know your neighbors, you can help each other, you know the vendors in all the stores and shops around the corner. This is a social capital because you interact with the people and it can be helpful for your psychological health and it can be even helpful from the point of view of criminality. If you know the person, then the person can either help you in a difficult situation or can tell you that something bad is happening, so it can protect you from being a victim of a crime. And therefore, in this sense, the residential stability is a good thing because you remain longer time in the same place and you don't have to give up on this social capital. You won't lose it. You have all your social networks in place and that's definitely an advantage.  

Mike Manville  54:06 

The only thing I would say about this is just that, like you said, Shane, one man's stability is another person's e-mobility. I think the classic example of the bad mobility or the bad e-mobility that you sometimes hear is just – and this maybe leads us into misallocation – people who let their house outgrow them. You have this wonderful price on a four-bedroom and you raised your kids there and so forth, but now further on in your life, it's just you and your spouse or maybe just you, but you actually would face a substantial penalty for downsizing because the price you pay for that controlled unit is so low and so you rationally hold on to it. But there's a family of four coming up somewhere that would value that, would value that more than sort of like what you would, but you don't have to confront that trade-off. The same opportunities that are given to one generation, the next one is sort of inadvertently deprived of them. 

Shane Phillips  55:06 

Yeah, yeah. I mean, it sounds like the reduced mobility is, to the extent it's bad for some people, is something that is harming you. You get a job offer or have opportunities elsewhere, but because you have this low rent, you decide it's not worth it to take advantage of those opportunities. I might like to move into a bigger home to accommodate a growing family and it's just the gap is so large, I feel kind of stuck. The misallocation is maybe sometimes people are in a position where they're getting a very good deal. They're one or two people living in a four-bedroom home. It's not really a problem for them per se, but it is a misallocation that is causing problems for other people who could really use that unit much more efficiently, effectively. It would suit their needs better and actually a smaller unit might suit the other people's needs better, but it's not harm in the same way. I feel like it's annoying maybe to have too many bedrooms, but some people would also be like, well, I have a space for an office, I can, you know, there are upsides too. 

Mike Manville  56:08 

But you might also keep it empty, right? Yeah. I mean, there are people who would, you say, use, but I think they would just value it more. But they'd actually be willing to pay more than you pay. And that mechanism, you know, as imperfect and cold and whatever it might be, it solves a lot of problems in society. And when it is interrupted, you, like Konstantin said, there's no world without trade-offs. 

Konstantin Kholodilin  56:32 

If I may, one example of bad immobility, so there might be some macroeconomic consequences. I heard about some firms complaining that they cannot attract new employees to Berlin because there's a housing problem. So there's a shortage of housing. It's very difficult to find housing there. And it becomes a really big problem for firms to attract people. So they have job vacancies, but they don't have people to fill in. And that's, well, it's not necessarily related to rent control. It's related to any case of housing shortage in the market. 

Shane Phillips  57:07 

Of which rent control can be a contributor for sure, but to varying degrees. 

Mike Manville  57:12 

You hear a similar story at college towns in the US that people love a lot, right? Like the joke, which I think was in a New York Times story about Berkeley, is new students would arrive and go to landlords and be like, why can't I find any housing? Everything's full. And the landlords would say, well, we're still housing the class of 1980. When they finally move on, you'll have some housing. But no, they moved in, they loved it, the rent's cheap, and it can stay cheap as long as they stay. And why should they move out? 

Shane Phillips  57:40 

All right. A lot of this has been focused on the negative, and that's partly because we glossed over the effect on controlled rents, which is real, and people benefit a lot from it. Even if there is a countervailing effect on uncontrolled rents and mobility and all these other things. Another positive thing I want to talk about that I did not see in the list of outcome categories was psychological or health effects of rent control. But I did want to see if anything like that came up in your research. I ask this because I've come to see the primary benefit of rent control not as affordability, but as predictability or security or stability or whatever you want to call it. The effect on rents seems to often be a wash, give or take, with the savings to tenants in controlled units being offset by additional costs for people who have moved more recently or who live in uncontrolled units. But looking beyond that, I have to think there's a real benefit to, let's just call it mental well-being, when you know that your rent can only go up by so much each year and you're protected against eviction so long as you uphold your end of the lease contract. I'm certain I've mentioned this a few times on the podcast before, but the analogy I like to share is of this randomized controlled trial of Medicaid expansion in Oregon, a health program for poor people, which in the first two years found, quote, increased healthcare utilization, reduced financial strain, and reduced depression, but produced no statistically significant effects on physical health or labor market outcomes. On the one hand, you could look at that and say that Medicaid expansion failed because it didn't improve health outcomes, and improved physical health is arguably what Medicaid is most about. I think it's a fair criticism, though maybe a longer study would have found health benefits too. That aside, on the other hand, reduced financial strain and reduced depression are also benefits that have real value to people, even if it is harder to assign a dollar value to them. It's entirely possible that in the case of this Medicaid expansion, the costs were too high relative to those benefits, or that you could get the same mental health benefits with other less costly programs, just as rent control may not be the best or most direct means of improving housing stability. I'm curious, though, how this idea strikes you and whether you came across anything like this, any kind of investigation of this idea in your reading of the literature. 

Konstantin Kholodilin  1:00:02 

Yeah, the idea seems to be quite fruitful to me. However, I didn't find any single paper on this, which does not imply that they do not exist in the first place. The absence of studies also doesn't mean that this issue is not interesting at all. Simply because people still didn't come across, maybe due to the fact that the first idea that comes upon your mind when you talk about rent control is the immediate effect on rents, and all other effects are not that evident, so you have to think about them. But what I found, because now I'm conducting a mega, mega study, which encompasses not only rent control, but also all different types of housing-relevant policies. So like 20 different policies, including even labor policy, monetary policy, fiscal policy. What I found in that literature review is that, for example, housing allowances or housing vouchers and social housing, there are studies that consider the effects of these two policies on physical and mental health. For example, for housing benefits, for housing allowances, there are studies that found positive effects on the health of both adults and children. So at least for these two types of support, you can really prove the beneficial impact of them on the health of people. So for me, it's not something unimaginable that rent control could also have such a positive effect. The only thing is that it hasn't been researched yet. But thanks to this literature review, we see that there's a niche that can be filled in that can be a topic of future research, and I think it's a very nice topic. 

Shane Phillips  1:01:50 

All right, so we've gone long, we're running out of time here. I want to ask just one more question, and it's another one about something that didn't really appear much in the review, which is homelessness and eviction, which are very important topics, especially nowadays, I feel, but they seem surprisingly understudied. Out of all the studies you categorized, there were fewer than 10 on homelessness and just one on eviction. To the extent we have studies on these, what do they find, and do you have a sense for why there aren't more empirical studies on this topic? 

Konstantin Kholodilin  1:02:22 

Well, for homelessness, I'm not really sure if there were a consensus on that or not. So I should admit, I don't remember all the facts. That wasn't the most important in terms of the number of studies, as you mentioned. So there were studies that found positive effect, no effect, and negative effect. So you can certainly find explanations why rent control could reduce homelessness. For the marginal households who are at the margin of losing their dwelling, reducing rent can help them from becoming homelessness. That's probably the most logical explanation for the rent control helping people not becoming homeless. One thing that probably is worth mentioning is that the lack of studies can have to do also with the lack of data. In our case, expression, I think archaeologists are using it very often. So you have to look not everywhere, but only where the light post is, where you have lots of light, in our case, lots of data, you will find something. And that's what most people are doing. Rents are easy to observe, therefore, more studies focus on rent dynamics. Homelessness is something which is pretty difficult to measure. And therefore, there are also less studies that concentrate on homelessness effects or eviction effects because, well, you have some statistics, but you have to, in the US you have Eviction Lab, a nice source of data on evictions. So maybe with the time more and more studies using this data will appear. But in most other countries, it's very difficult to find eviction data. 

Mike Manville  1:04:03 

Yeah, this, you know, this is a topic that this positive relationship has kind of come and gone in discourse about rent control in the 1980s. There was a flurry of it. And of course, homelessness in the 1980s looked a little bit different qualitatively than it does today. But the consensus that I think came was, it just wasn't that much of a relationship, you know, that you could imagine a rent control policy so strong that it really constrained the supply of housing. Well, then, of course, you know, that's going to contribute to homelessness insofar as not enough housing does. That's going to be counterbalanced as Konstantin points out by like, well, if you can hold rent down for people, you know, if those are folks who are sort of at risk of being one step away from being homeless, then that's going to have a mitigating effect. And I think a complicating factor here, in addition to how hard the data can be, is just that one thing we know from recent examinations of homelessness in California and some other places that the typical person on the verge of being homeless just isn't even on a lease. You know, they've reached a point where they're crashing in someone's room, they're couch surfing and so forth. And so the statutory rent has really an only indirect effect on how housed they are. You know, by the time they've reached a point where they're just one step away from homelessness, they're kind of already just chipping in some cash and relying on the goodwill of family and friends and the precipitating event that sends them out of that unit may not actually be a rent increase. It may just be an interpersonal conflict or they can't contribute even the small amount that they were able to contribute to that point. All of which is just to say that I think, you know, it's really hard to tell and homelessness has many causes and I would not put rent control in the top five. You know, I mean, it's sort of, but it deserves more research, sure. 

Shane Phillips  1:05:00 

Well, it deserves more research, I think is a good place to end an episode on a literature review. So, Konstantin Kholodilin thank you so much for joining us on the Housing Voice Podcast. 

 Mike Manville  1:05:55 

Yeah, thanks Konstantin. 

Konstantin Kholodilin  1:05:56 

Thanks a lot. 

Shane Phillips  1:06:02 

You can read more about Konstantin'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 the socials, I'm on Blue Sky at Shane D. Phillips and Mike is on Twitter at MichaelManville6. Thanks for listening. We'll see you next time.