Episode 36: Rent Control in India with Sahil Gandhi and Richard Green

Shane Phillips 0:04
Hello, this is the UCLA housing voice podcast, and I'm your host, Shane Phillips. This week's guests are Sahil Gandhi of the University of Manchester and Richard Green of University of Southern California, and they joined us to talk about their research on a problem facing many Indian cities - a somewhat unusual combination of high vacancy rates and high rents. Under normal circumstances, high vacancy rates would predict low rents, but that relationship breaks down here due to what Sahil and Richard describe as insecure property rights. Specifically, much of India has very strict rent control regulations, and it has too few judges to process evictions in anything like a timely manner, sometimes taking over a year to evict a non-paying tenant. Most of this conversation is focused on rent control, and specifically first-generation rent control, which we'll define and contrast with the second-generation version that's now more common in much of the world, including in the US. As we also say in the interview, India's experience is instructive but it's also extreme. Because India's rent control regulations are so strict, and the time it takes to process evictions so long, these findings shouldn't be interpreted as a definitive case against rent control, no matter how it's designed, nor an argument for turning courts into eviction factories that turn out warrants the instant a case is filed. At the same time, it does illustrate how important it can be to get the details right, and it can help other places avoid similar mistakes, or keep us from repeating our own mistakes. That learning process is exactly how we ended up with second-generation rent control after our experience with the first generation. The Housing Voice Podcast is a production of the UCLA Lewis Center for Regional Policy Studies, with production support from Claudia Bustamante, Olivia Urena, and Jason Sutedja. This is actually our last episode working with Olivia and I just want to say how much we've appreciated her input and wish her the best in her next role. As always, feedback and show ideas can go to me at Shanephillips@ucla.edu, and you can give the show a five star rating and a review on Apple or Spotify. Now let's talk to Professors Gandhi and Green.

Dr. Sahil Gandhi is lecturer in Real Estate and Urban Economics at the University of Manchester and has a background here in LA as a former USC PostDoc, and Professor Richard Green is Director of the USC Lusk Center for Real Estate, and they are both here to talk with us about rent control, insecure property rights and the high price of housing in India. Sahil and Richard. Welcome to the housing Voice podcast.

Richard Green 2:54
Thanks for having us.

Sahil Gandhi 2:55
Thanks for having us. Shane.

Shane Phillips 2:57
And Mike Manville you are our co host today. welcome!

Richard Green 3:00
Thanks, great to be here.

So as always, we start with a tour from our guests. We've got two today, so we'll we'll try to get through them as quick as possible. Sahil, you were an assistant professor in Mumbai at one time so I wonder if we could start with you. This would be our first urbanist tour of an Indian city on the podcast, where should we go?

Sahil Gandhi 3:18
So I've spent most of my life in Mumbai, a city with a density greater than 20,000 per square kilometer, and a population of around 20 million for the metropolitan region. So it is kind of a bustling city, you will see a lot of people you see a lot of vehicles when you walk around. Now there is a very lively informal economy. You will see this on the streets. There are a lot of street vendors, you'll find a vegetable cart vendor, a coconut vendor on most residential streets. And at every corner of the road, there will be a roadside team and a chaiwalla or waterpower stall. It clearly seems that I'm missing Indian food right now. You will also see a lot of traffic everywhere in Mumbai. Along with traffic you will hear the car and bike honking quite often. Over the last two decades, there has been a lot of real estate activity, but public infrastructure investment has not kept pace. And you can see that through the crumbling infrastructure in the city. Coming to housing, almost 40% of Mumbai lives in slums. They are distributed all over the city but there are certain parts where they are concentrated. Dharavi is one such area which we know that is one of the biggest slums in the world. These places are quite safe though, and crime is not so much of an issue. On the other hand, we will see that there is a lot of luxury buildings that overlook the sea. One such building is the tower where Mukesh Ambani stays, he is one of the top 10 billionaires in the world. So I hope that gives you an idea or a feel of Mumbai.

Shane Phillips 5:02
is that the tower where it's like literally the whole building is just him and his family essentially.

Sahil Gandhi 5:07
No one else lives there yes,I recall reading about that.

Shane Phillips 5:10
Richard, how about you? Where are you going to give us a tour.

Richard Green 5:13
So I'll give you a tour of the town I grew up in, which is Lacrosse, Wisconsin. And the reason I'm going to give you a tour of is it explains why I love living in Los Angeles, California. So Lacrosse is in a beautiful location on the Mississippi River, It's on a plane that surrounded by bluffs. And by the way, for a weekend in the late summer, early fall, I would say it's an absolutely marvelous place to spend a couple of days. It has no traffic, very little crime to speak of, the cost of housing is quite low, people are pretty nice to each other, and I couldn't imagine actually living in a place like that. Because the variety that one gets in a really big city, and Mumbai is clearly an example of a really, really big city is something that is important to me. I like having my various senses stimulated in various ways on a regular basis but the people I grew up with in Lacrosse can't imagine why I like living in Los Angeles because they do enjoy the absence of traffic, and the cheap housing, and so on. And so I think it's just a great example of how when people try to rank places to live, that they're really, it's a fool's game to do it because where one wants to live is so much a function of one's own personal tastes and preferences, and the reasons I couldn't abide living in Lacrosse are reasons that a whole lot of people want well not a whole lot, it's about 100,000 people in the metropolitan area so it's not a whole lot of people but some people want to live there.

Shane Phillips 7:00
Lots of people live in very similar places

Richard Green 7:02
A lot of people live in very similar places. But for me, I I lived in Washington a little for a while before I lived in LA and even DC is a little too small for me. Not bad, but a little too small so I love how complicated, how messy, how interesting Los Angeles, and I think it's a large part because of where I grew up.

Shane Phillips 7:23
I think that one of my questions at the very end here sort of relates to this, this idea of we're all different people, and we were even different people ourselves at different times. So we will get to that different strokes for sure. So the article that we're discussing is in the Journal of urban economics, and it's titled 'insecure property rights and the housing market explaining India's housing vacancy paradox', and in addition to you two, your third co author is Shaonlee Patranabis. The housing vacancy paradox you're referring to is that India has a high housing vacancy rate and high prices relative to incomes, and that's a fairly unusual combo. There's a lot of data for cities all over the world showing that when vacancy rates are low growth in housing prices, and especially rents tends to be pretty high. And when vacancy rates are high, then rent growth is low. India defies expectations on this, and you attribute that to what you describe as insecure property rights, of which rent control is one very important componentl. The relationship between housing vacancies and the density of judges able to adjudicate cases was also explored as a part of this paper, and so we'll try to touch on that during this conversation as well. I'm going to give a little more introduction here, because this is our first episode specifically about rent control. And so I want to preface it by saying that the rent control we're going to be discussing here today is a pretty extreme version by world standards, or at least by US standards. In some ways, it's kind of a worst case scenario in terms of unintended consequences. As we'll discuss the experience of cities like Mumbai, with first generation rent control, or hard rent control is not the whole story of rent control. In the US, almost all cities that have rent control today have what is known as second generation rent control, which also certainly has downsides and trade offs but it's considered by pretty much every economist and housing scholar on the planet to cause fewer problems compared to that first generation version. Reasonable people, including folks on this exact podcast can and do disagree about whether second generation rent control is worth the costs. I personally do think so but plenty of others don't, and one thing that skeptics and I and supporters generally agree on, is that rent control isn't a substitute for building enough homes to keep up with rising demand. We will do our best to define our terms throughout this conversation and to make it clear when we're talking about the impact rent controls as a general policy, and first generation rent control specifically, and we will definitely plan to do a future episode that focuses on second generation rent controls in particular. But if you understand first generation rent control and its documented impacts, then you're on your way to understanding second generation rent control. And as Mike, our co host here said to me when we were planning this episode, the second generation version can look weird and needlessly complicated, unless you first understand first generation rent control and the problems that the second generation policy was responding to. So, Sahil, or Richard, before we get into rent control in India, could you first just give us a snapshot of India's housing market more generally, when you say that housing prices are high and vacancies are high, what does that actually look like? How much are people paying relative to their incomes? And how many vacant houses are we talking about?

Sahil Gandhi 11:01
So housing is expensive in a lot of Indian cities, the house price when compared to incomes is extremely high in the larger metropolitan cities. It is greater than 20 in Mumbai, and around 15 in Delhi and Kolkata. According to the Reserve Bank of India, which is the Central Bank of India, for households who can borrow from the bank, the median monthly installment that they have to pay off their incomes in 2019 was around 45% for cities in Ahmedabad and Mumbai, and 40% for Pune. So that is quite high.

Shane Phillips 11:41
And just so I'm clear the stat you gave before about 15 to 20 Is that the ratio of house prices to incomes?

Sahil Gandhi 11:48
Yes, the house price when compared to incomes Yes.

Shane Phillips 11:51
Got it, okay.

Sahil Gandhi 11:52
So many households cannot afford formal housing. And you have a majority of like, we have quite a few percent of people living in slums for Indian cities 16% of the urban households live in slums. So, there is a disparity there are few households living in luxury housing and there are like a lot of households living in slums. In 2011, It has been estimated that the housing shortage in India's urban parts is around 19 million, and and the urban household number is 80 million. So when you compare 80 and 20 million, one in four households have some kind of a housing shortage. We have around 11 million vacant units. So while the housing shortage is 19 million, the vacant housing number is 11 million. So there is some kind of a misallocation when it comes to housing in India.

Shane Phillips 12:49
Yeah, and I wanted to add a stat that stuck out to me was that about only about 28% of houses in India were rented in 2011, which was actually down from 53% in 1971. But at the same time in 2011, roughly 12% of houses were vacant, and so you have a 28% of of people households renting, and 12% of housing units that are vacant, so you have almost half as many vacant homes as you do rented homes. It's just astounding statistic to me. So you have a hypothesis about why prices are so high despite vacancies also being high, which rent control and the availability of judges both factor into. Normally, if vacancies were high, you would expect landlords to lower their prices, since it's better to bring in a little less rent than to bring in none at all. But it seems that there are a lot of landlords not making that choice. We can talk about the judge problem later but starting with rent control, can you first just define and contrast the first generation and second generation rent control for us and walk us through what it is about this policy that might cause landlords to leave their units vacant?

Richard Green 14:05
So first generation rent control are sometimes called Hard rent control usually means that once a property is under rent control, it is forever rent controlled. It doesn't matter who lives in the property. If the property turns over, the landlord is still limited in the rent that he or she is allowed to charge. And the rent control under a first generation or hard rent control tends to not take into account at least fully the general increase in consumer prices in an economy. And there's there's some history behind this, which is that rent control has come about both in India and in the United States, in periods of say in the US war induced shortages, and so the idea was that rent itself was an important part of the price index, and so if you allowed rent to move with prices, then you would sort of be undermining the whole idea of rent control. And so rents not adjusting even to the price level, and people in the rent control remaining, regardless of who lived in the unit, I think are the two most important features. Second generation rent control tends to undo those features. So there's something called vacancy decontrol, which basically means that if somebody moves out of the unit, then the rent is permitted to reset to market value. Whereas as I said, the other case, that wasn't true. And you know, on the one hand, what that means is that if you're a landlord with certain expectations about turnover, and for example, near our campuses at UCLA, and USC, landlords are not affected all that much by the soft rent control, because people here are as students or as medical residents, they stay here for three or four years, and then they move out. So even though landlords may not be able to raise the rent as much as they liked during the three or four years, they're going to be able to reset it in a reasonably short period of time. And again, is the other thing is site generation rent control, it usually permits, increases in rent, that's some function of increases in the consumer price index. So for example, here in California, our statewide rent control law says that you may increase rents by CPI plus and I'm blanking now on whether it's 5 or 6% but basically,

Shane Phillips 16:39
By 5 % I think

Richard Green 16:40
Yeah, there's a lot of room. Now as it happens, that's turned out to be binding here in Southern California in the last couple of years, which surprised me. So what I would say is the movement is away from something that just says you can't raise the price of what you have ever to one that say we're going to help people stay in their houses, and we're going to prevent gouging from happening. Now, Shane, as you noted, some of us are still even skeptical about the second generation, but clearly, it is nowhere near as distortionary as the first generation version of this.

Shane Phillips 17:16
And what about the distortions? Why is there this suspicion that the first generation version would cause some landlords to just leave their units vacant, rather than rent them, you know, more cheaply, for example?

Richard Green 17:30
Well, it's because people don't run businesses to lose money. And if you have rents controlled at a very low level, and you can't cover your expenses, then you're not going to do that business. And at the same time, there's a view in India, that real estate will always go up in value. There's a whole host of reasons behind that, and maybe sale can talk about that. But so, if a landlord, I'm gonna say, "You know what, I would rather just keep the place empty, and then sell it someday at a profit, then have the drip, drip drip of actually losing money each year". I think the other thing that's particularly important in the Indian context, and again, Sahil can weigh in on this is when it's difficult to evict people for not paying anything at all, then it really, a landlord might very rationally say, you know, what, I'm just gonna keep this place, and it's a store, it's like gold, it's just a physical asset with a stored value, but I'm not going to actually have it be part of the housing market.

Shane Phillips 18:38
Just so we understand what are the limitations on rent increases actually look like say, say, in Mumbai? I asked that, because if you can raise rents by you know, equal to the inflation rate here in Los Angeles or inflation plus 5% in California overall, is it just 70 years ago or 50 years ago, they they established a rent control law, and they said, whatever the rents are now they're just never gonna go up or is it somewhere between what we have and that like, most extreme version?

Sahil Gandhi 19:09
No, it is very, very extreme and several parts of India. In some states, you're just allowed to raise rents equivalent to like if a landlord does some upkeep of their property or does a major renovation, then some parts of the renovation costs can be added on to the rent, and that can only be added on if the tenant agrees to that renovation, or else it does not get added on. In certain parts, you are allowed only 3% of an increase in rents after five years, and this 3% is annually, but however, the inflation rate when it is 8%, 9%, that 3% is very, very binding, and you are allowed that only after five years, so you've lost all the all the inflation, like your rents have been devalued by four or five years, and then it increases by 3%. It is quite stringent, and when it comes to eviction, it's really, really hard to evict tenants who have not paid rent. You have to go to several courts, and it takes several years to get a non paying tenant evicted. So it is quite stuck in India as compared to the US.

Shane Phillips 20:30
Got it. Yeah, and you can imagine, I mean, if you can only raise your rent 3% a year, but inflation is 8%, and your costs are going up probably quickly as well, you're just guaranteed to fall behind over time, Mike?

Michael Manville 20:42
Yeah, I was just gonna say, I mean, listening to Richard and Sahil you can you can start to see a kind of a bad recipe forming, right, you know, because one thing, you know, to Richard's point about the first generation rent controls, especially in the United States, is that they really were part of a comprehensive set of price controls designed to prevent inflation during the war. And so it was the case that yes, he was a landlord couldn't collect much rent but also when you and his landlord went to buy something from a contractor to fix your unit, they could only charge you so much. And when one thing that happened to rent control, the post war United States was that all those other price controls disappeared, and in New York City, rent control stayed, and so suddenly, it was just like you were a landlord, and you alone, face to price control, and all your other product costs would rise. And so if you have a situation like that, and then combine it to Sahil's point with after five years, you know, you finally get to raise the rent to catch up with expenses, and then your tenant does have the option of just saying, "well, I'm not going to pay it" and can probably squeeze another two years of occupancy out. I mean, that really does, you know, to Rich's point suggests, well, I'm just going to hold on to this as an asset, but I'm not going to get any income off it, and and of course, that's completely counter counterproductive. I mean, you really can see how this would these two things would combined to generate this paradox you're talking about.

Shane Phillips 22:12
And one other thing I wanted to note, just a paper that I read a couple of years ago, also about Mumbai, it's called 'A decline of rental housing in India, the case of Mumbai', it's by Ten Dell and co authors. What they found is that from 1961, to 2011, the stock of owner occupied housing in Mumbai, which has these very hard rent controls, grew by about 2500%, so grew by 25 times, and at the same time, the rental stock increased by less than 5%. So this was a very, very severe restriction that made building or expanding rental housing completely unappealing, there was no reason whatsoever to do it, and so we'll include that article in the show notes as well. So another distinction I think, might be helpful here is between frictional vacancies and quasi permanent vacancies, what is the difference between those and why would we expect one to have different impacts on housing prices than the other?

Richard Green 23:13
So vacancies are never zero, because there are always frictions in the housing market. And it's, let's just start with a very simple one. So we would never expect the vacancy rate to go below 3%. Now, as it happens, I'll bring it back to Southern California, we have lots of markets in Southern California that are below 3%, which will tell you how tight we are. And the reason for that is quite simple, is the typical renter stays in their unit for about three years, and it takes a month or two after one tenant leaves to sort of tidy the place up and market the place. And this is even in a very well functioning market, you would expect a unit to have one month out of 36 being vacancy being vacant just because of turnover. There's nothing wrong with that. That's just how things are and so that by itself means you will not get a zero vacancy rate. The other thing is sometimes you will have mismatches between the units available and the kind of units people want. So let's say we're in a market with lots of single people, okay, you might find that the one bedroom and studio market has a vacancy rate of 2% but the three bedroom market has a vacancy rate of 7%. And it's just because there aren't households match or it could be the reverse, of course. But the point is, if the typology is not the same as household desire, you could get some vacancy as a result of that as part of natural forces. And then the other thing that goes on is sometimes developers will get a little ahead of themselves. And so you'll find that sometimes the fringes and you actually do see this near the fringes will have higher vacancy rates for the Center City. If, and you put all that together and you get some vacancy, so those are just market phenomena that happen.

Shane Phillips 25:05
Those would be the frictional vacancies,

Richard Green 25:08
Yes, these are frictions that just happen in the market. By quasi permanent, we mean, you know, somebody owns a building, and they say, You know what, we're just not going to rent it out, it's not worth it. And that's because of a regulatory regime in which people are not permitted to make a profit. And, you know, we can talk about excessive or not, I'm talking about, they're not permitted to make any profit at all. But again, they see housing as a way to preserve capital and so they buy housing, just leave it empty, and that's what we mean by the more quasi permanent.

Shane Phillips 25:42
And this would also include vacation homes, second homes, this kind of thing, maybe not as much of a problem in India, but like, they would also fall under that header of quasi, just so we're clear?

Richard Green 25:52
Yesh

Shane Phillips 25:53
And why would we expect those to have different impacts on housing prices?

Richard Green 25:59
So the quasi permanent stuff, basically, as you're moving stock, you're just removing supply that's not being used. And as I don't remember, there's gender microset so it's just counterproductive, right, particularly if it's induced by regulation, and you could, you know, we can have arguments about whether second houses are okay for people or not, but let's not talk about that today. But the whereas the the frictional stuff, is that's just a natural phenomenon in the market. And, you know, I sometimes see housing advocates say, Well, you know, the vacancy rate in Los Angeles is 3%. And there are, you know, helped me with this math here.

Shane Phillips 26:37
60,000, unhoused people, that kind of thing.

Richard Green 26:41
Yeah, and so they say, why don't we just put all these unhoused people and it's, you know, is is that's just stuff that's waiting for the next renter who will be in there within two to four weeks and Right, right. So that does not reflect a surplus of housing, that reflects the natural workings of the housing market,

Shane Phillips 27:01
Worth noting that the overall vacancy rate in Los Angeles is at or near historic lows, despite those numbers. So we can now move on to kind of a little more of the context here in India, how did rent control come about in India. It's been there for a long time. Rent controls, of course, have popped up all over the world different times. So there's clearly something to them some problem or need that they are responding to, that people feel like they're an appropriate response to. They seem to be especially popular during times of war and very high inflation, maybe as a kind of stopgap measure where you don't have enough resources to build at the time, or you just can't build fast enough to keep up with rising demand. So for the benefit of our audience, what is the positive case for rent controls? What is your steel man argument?

Richard Green 27:50
Yeah, so I think Sahil wants me to take this one, so I will. Rent control is really good for people who are in rent controlled units, they are beneficiaries of not seeing them, essentially, what you've done is you've transferred a property right from the owner to the incumbent, when you have rent control. I mean, we think of you know, fee simple interest as being a bundle of sticks, and when you have rent control, some of those sticks are taken away from the owner and give it to the tenant. And there are I'm sure a number of people in your audience who thinks that's an absolutely wonderful thing, and it's absolutely true. I mean, it's an end, you look at, like Rebecca Diamonds, heavily cited paper about rent control in the Bay Area, she makes no bones about it is the the people who are in place benefit a lot from rent control. The problem is that the people who are not in place, do not benefit from rent control and might be harmed by it. In fact, a lot of us think they are harmed by it. And so you get into this whole philosophical issue of how do you determine fairness? And you know, Mike, I heard you talk about this very eloquently last week, is there's sort of what we'll call incumbency bias, which is, people who are there we tend to value more than people who aren't physically. But you also talked about I really liked this temporally is people who are already on earth, we tend to give greater value to than the people who are going to follow us. And so if we don't have enough housing for people in the future, well, they're not here yet. So why do we worry about it? And so you get into this whole argument about, you know, what constitutes fairness and your views about whether there should be a preference for incumbency or not, is critical to how you think about things so, but there's no doubt that the plus two it is it is good for people who happen to be in rent controlled units.

Shane Phillips 29:52
So let's let's give a quick overview of the methodology for this paper. I'm going to give my explanation of it you can tell me How much is right how much is wrong? As I understand it, you looked at states and districts throughout India, and you group them by whether or not they made changes to the rent control laws during a specific time period. Places that change their laws in a more pro landlord direction are in one group, and places that just didn't make those changes are in another. And you're evaluating how vacancy trends differed between the two groups. The presumption here is that if the places that change their rent control policies exhibited different vacancy trends, like a faster fall or slower rise in vacancies, that it could be attributed to the policy change. So is that on track so far, and what else should we know about your research approach here?

Sahil Gandhi 30:42
That's broadly correct. But before we get into the methodology, we spent a lot of time collecting the rent tax and going over them. So we literally had to find these acts and go through them and get the relevant clauses and the data from it.

Shane Phillips 31:00
Because it varies for every single state and district and so forth, right?

Sahil Gandhi 31:04
Oh, plenty of the state. Yes. And we managed to get a panel for 2001 and 2011. So the clauses that are applicable for 2001 census data and the clauses that are applicable to 2011. So we had two data points, we had the Census of India for 2001, and 2011, and we also had rent control clauses for the two years, which are applicable for the two years. So we exploited changes in the rent control law to see the impact on vacancy rates between 2001 and 2011. We have states that change the rent control laws, which are our treatment states, and we have states that did not change the law, which are what control states, most changes happen from a pro tenant one to pro landlord one, we then use a panel data to run a two way fixed effects regression to see if places that change their rent control laws move differently from the control group as you correctly mentioned Shane.

Shane Phillips 32:03
And you looked at several components of rent control policies to see which of these were actually associated with with changes in vacancy, these included the number of months that non payment is allowed before a tenant can be evicted, the age of buildings that are covered by rent control, whether landlords can evict tenants who aren't occupying the unit that they're leasing, which is an interesting one to me, and how restrictive the limitations on rent increases are. What you found is that it's only really changes to that last one, the limitations on rent increases that had a large and statistically significant impact on vacancies. Since limits on allowable rent increases seem to be the important factor here, can you explain for us how you distinguish between more and less restrictive policies on that dimension? As you define them in your study, what is a what is a pro tenant rent increase policy, and what is a pro landlord one, at least in the Indian context?

Sahil Gandhi 33:00
Right, so when we went through these clauses, we realized that, with respect to rent provision, there were four important things that allowed a landlord to change rents. The first one is when there's an increase in market value of the property, and after five years, after five-year period, some parts of the market value can get reflected in an increase in rent. The second thing is, landlords are allowed to increase their rents periodically, there are caps on this increase, as I mentioned earlier, the third one is landlords can increase rents if the tax goes up. And the fourth one is, rents can be revised only when there is a physical improvement. So there are caps on all of these, and we had to take a call on which of this is a pro-tenant one, which of this is a pro-landlord, one, and we decided that if a landlord is allowed to increase rents only due to physical improvement, that is a pro-tenant rent revision. And all the other three are pro landlord rent revisions, we find places that make changes in rent revisions to become pro landlord see a decrease in vacancy rates. Karnataka is one of the states that made this change. And we find that a pro-landlord policy move will reduce vacancy by three percentage points, when the average vacancy is around 12% or a three percentage point reduction is quite a lot.

Shane Phillips 34:37
Yeah, that is really huge. So you explained, you know, there are these different components of the law that are determining an end of the rent revision component of the law specifically, that you're deciding our pro landlord or pro tenant. We had a conversation about inclusionary zoning with Emily Hamilton a few episodes ago, and she created her own metric for how restrictive these iz laws are, and it also has a lot of different things she has to balance. It's tough to know where to draw the lines. You talked a little bit about it but I'm curious, you know, what methodological or theoretical concerns might have come up as you were deciding, like, here's where we're going to draw the line, you know, what might you be missing? What problems could it cause for your models based on what you what the decisions that you made here?

Sahil Gandhi 35:24
Right, so that's a great question. And as I mentioned, that there are four clauses that comes within the rent revision in our paper, and, and we had a hard time to decide where the increase in rents due to an increase in tax scores is that theory, pro landlord, or is that kind of to tenant because it is finally a cost on the landlord that is just getting passed on to the tenant. So we weren't sure about that. So what we did finally, is that we started playing around with what goes into a pro tenant and a pro landlord. And these, these results are there in the appendix of the paper

Shane Phillips 36:04
To kind of like shifted, shifted the goalposts a little bit and exacting thresholds.

Sahil Gandhi 36:08
And we also made, we also used a dummy for all four of them, the all four of the rent revision clauses. So we started playing around, incorporating all the four rent revision clauses in our model, which is right at the end in the appendix of the paper. And we find that truly only one of it has an impact on vacancy rates in a negative way, which is, if landlords are allowed to increase rents based on an increase in market value, then that leads to a reduction in vacancy rates. When we ran this as four different regressions and included each of these clauses as an independent variable separately, we find that the market value, an increase in rents due to an increase in market value, reduces vacancy rate.

Shane Phillips 37:03
Okay, in the article, you discuss how different places have different "natural vacancy rates", and how those rates depend on many things, but one of them is the existence or absence of rent control. It reminded me of something that I often argue, which is that rent control actually, I think, increases the local demand for housing in a certain way. If rent control keeps people in their homes longer, on average, which it does seem to do, then it necessarily reduces the number of homes available for everyone else, there's not as many vacancies coming up. And so if we don't pair it with policies to increase homebuilding, we're really committing ourselves to lower vacancy rates and housing prices rising more quickly. I think it's actually a decent policy to you know, let people stay in their homes longer, and if rent control allows them to do that, then maybe that's, that's okay, and maybe that's generally a good thing. But I, at the same time, do think we have to acknowledge that it does have impacts on the housing market writ large, and ultimately, all of the people in it. What are your thoughts, both of you, either of you on that interaction between rent control and housing production? I think there's there's a lot of things that could be said here but one thing I find kind of frustrating about some of the policy debates out there is how things get framed, as you know, benefiting existing tenants or future tenants or people who don't want to move versus those who do want to move but in reality, most of us are every one of those people at different times in our lives, and it seems that we need to do maybe a better job of reflecting that in our policies and rhetoric. I guess

Richard Green 38:44
I'll go first. So first of all, I agree with you, and one of the things that's striking to me is if we look at, I'm going to come back to the United States for a minute, if we look at CPI rent data, going back to 1998,

Shane Phillips 38:57
Which is CPI just being Consumer Price Index, inflation rate.

Richard Green 39:01
Yeah, yeah, the rental component of the Consumer Price Index, going back to 1998, which is sort of when the series as it currently is began(I'm not trying to cherry pick a beginning date there), the three cities with the fastest growing rental component of CPI have been New York, San Francisco, and Los Angeles, which comes in at number one, by the way, all cities with rent control. Now, it's what I mean sometimes economic theory really works. So what it says is if you have a basically a dual market, one that's rent-controlled, and one that's not what happens is the market that's not rent controlled, you get higher prices than you would in equilibrium. The reason you just gave is is you don't just have the sort of natural churning in the market that you would if you didn't have rent control. But the other thing is there's a political economy aspect of it is it so is the part that I find particularly frustrating. So, in principle, you could actually convince me that second-generation rent control is okay. And it goes back to an article by Richard Arnott, the Journal of Economic Perspectives, and he basically says, "Well, if landlords are earning what we call Ricardian rents, which is to say they don't do anything, they're not really good landlords, they're not keeping up their buildings, but just by being the lucky people who are sitting in the right place, they can take up their rents. Okay, well, that's not particularly economically productive, either". They basically have a monopoly position or some sort of non-competitive position. But you know, I say to myself, what's causing the monopoly rents or the Ricardian rents, well, a lot of it is the land use regulation. Very often, the thing that creates Ricardian rents is a regulatory regime that prevents competition in the housing market, because it prevents new construction. And so the Ricardian rents are a function of the regulation, and so then the regulator's say, "Oh, these Ricardian rents are bad so let's cap what they can be!" when maybe the right thing to do is not discourage construction in the first place. And we could see examples of markets that have done a pretty good job of keeping their housing, pretty affordable. Chicago would be one, Minneapolis would be one, even though they haven't, you know, we're not going to talk about economies that are shrinking, that that would be a straw man to talk about. Well, Detroit is cheap. But you know, Chicago, Minneapolis, these are places with very vibrant economies, and their housing is reasonably inexpensive, and it's reasonably easy to build in those cities. And so if elected officials say, "You know what, we're going to build like crazy, and in the short run, we're going to have some second-generation rent control until we get to where we need to go", I can actually live with that as a policy but it's how electeds use rent control as a cudgel to say, see, I'm doing something about the housing crisis, when in fact, they're just moving things around and not fundamentally solving the issue.

Shane Phillips 42:13
Right, they're sort of creating the conditions for landlords to collect these Ricardian rents that they didn't earn, and then getting mad at them for doing so, and still not changing the actual conditions at a fundamental level, exactly, by which those rents those rents are rising.

Michael Manville 42:30
I think, you know, everything Richard said I agree with and I would just add to it that, you know, in principle, you could have second generation rent control with robust building and zoning deregulation, and so forth, and I think the reason people like me are skeptical of that is just that in practice, that doesn't happen. Right, that and I think if you were to go back to the question that you posed to Richard earlier, like, what's the positive case for rent control? I would say that Richard is absolutely right, it's a huge advantage to incumbent tenants and rent-controlled apartments, and in full disclosure, I am one, and that should tell you something about rent control, that a tenured professor at the University of California gets a housing subsidy courtesy of the City of West Hollywood. But in addition, it's wonderful for elected officials. Right, because I think the reason rent control becomes popular in some places, is that robust building is unpopular. Taxation to raise housing subsidies is unpopular. And so if you can seem to be accomplishing something with the stroke of a pen, that is a wonderful escape valve for a city council member, right? You can say, we're going to keep rents down but I'm actually not changing the neighborhood, or the tax bill of the typical voter. And then I think the evidence to this is, Richard and I were talking about this last week, you know, look at New York City's rent control law. Just like the actual law, it is an emergency rent control law, passed in 1948 to be ended, when the vacancy rate exceeds a certain amount. Look at Berkeley's rent control law, it is Berkeley's law that when the vacancy rate in Berkeley, crosses, whatever it is 6% for six months, rent control is repealed. It has never been repealed. So in principle, yes, roll a rent control law together with a robust building program, and practice what cities treat rent control as an excuse not to build.

Richard Green 44:26
If I could just follow up briefly on Mike's point to on who benefits and who doesn't that feel. So one of the things I did is I once plotted the incomes of people in rent stabilized and non rent stabilized buildings in San Francisco, Oakland, Los Angeles, and I think maybe one other place, and they're the same. There is no targeting. There is no even slight movement of benefits toward low income people relative to high income people who are renters. Now, to be fair, apparently in Denmark, they say "well, that's a feature not a bug, because it means low-income people in high-income people live close together" so I do want to acknowledge that but if we actually care about helping the people who most needed this is not a particularly effective way to do it.

Shane Phillips 45:13
This raises a question, I've had something that comes up when I read a lot of times, and I think the Rebecca Diamond paper about rent control in San Francisco brings us up as well. It sort of does what a lot of economists do, which is say, you know, it'd be more efficient, more effective, better targeted to, you know, just give people housing vouchers rather than enact these price controls. But what do you think about the fact that, you know, at the local level, certainly cities can change their zoning laws to allow more housing and you know, over time, that should help, you know, keep prices lower, but they don't really have the control to just issue a bunch of housing vouchers to anyone who wants them. And so there's this problem of, well, you know, what, what is actually within our power to do even if there is this more optimal policy that could be implemented, but you know, it's out of our hands, the federal government would have to do it.

Richard Green 46:04
Well, I have a bit of a fantasy, which is that would allow local governments to do this. There is an argument against upzoning, for which I have some sympathy, and it goes like this. Suppose you have a landowner with a property that's zoned for four units. And now you upzone it to allow for 120 units, you have just given that land over an enormous windfall, the value of the property will be much higher. And let me just sidebar and people say, so see, you're not helping affordability. But no, it is. So let's say you've tripled the value of the land, but the cost of land per unit of 120, will be much lower than it is for it so it really does help affordability. But nevertheless, again, this is sort of an issue of what people consider fair is just handing over what could imagine millions of dollars to somebody who happens to be in the right place just doesn't seem like an okay thing.

Shane Phillips 46:59
And at least with renters, renters are generally poorer than homeowners with homeowners, it's like property owners, these are much wealthier, much higher income, on average, that's even feels even more unfair.

Richard Green 47:10
And so what I would love to see as a mechanism, that options off air rights, so that people have to pay for the permission to build, but not in a sort of one-off way, not in a sort of spot zoning way that we do in Los Angeles, which I think is really awful, and leads to corruption, all kinds of other things, but rather, a well-organized auction, which we've done for things like oil and gas leases, which we've done for broadcast rights, and so on. And what that could do is generate the revenue necessary to provide this housing assistance to people who need it at the local level. So in a city like Los Angeles, there is an awful lot, and by the way and so in a city like Delhi, there is an awful lot of value that's going on top that could really help the fiscal condition of those cities. And one of the ways that revenue could be used would be to provide rental assistance, it could also help make transit better, which allows density to be more pleasant than it otherwise would be, and so on. But I think by keeping these low levels of density, we're doing damage in all kinds of dimensions, among which we are preventing ourselves from having the revenue we need in order to help people who need that help.

As almost always, we come back to "yeah, we really got to build some housing here if we're going to solve the fundamental problem here". I do want to make sure, I think we have enough time to do this, to talk about the judges component of your paper, because this sort of just builds on the same concept of insecure property rights from a different angle, you find that places with more judges per 1000 residents, so its population adjusted, they also have lower vacancy rates, and this relationship is consistent and statistically significant across multiple regression models. I'll just ask you first, why do you think that is the case? Or what is your hypothesis for why it is true that when you have more judges, vacancy rates tend to be lower?

Sahil Gandhi 49:17
It takes really long to resolve a dispute in India, and there are a lot of pending judicial cases in district courts in India. So if you increase the number of judges or if it don't use up per capita terms, there are more number of judges, they can go through cases much faster and and that could resolve cases related to the rental dispute that landlord attendants are having much faster, which eventually goes to a condition where landlords have faith that if a tenant is not going to pay rent, we can get rid of the tenant or we can evict the tenant pretty fast. So a strong judicial system creates an environment for a landlord All right, they have confidence that if a tenant does not pay, we can evict them. So we think that in a place where a judicial system is weak, where the number of judges are lower, it takes longer to resolve cases. And in such places, landlords will not rent out, they would rather just have their property weakened.

Shane Phillips 50:22
Yeah. So similar idea. How big an impact was this relative to the pro tenant versus pro landlord rent control laws? And just so I'm clear, is this judges per 1000 residents, like any kind of judge, or is this specifically judges who handle eviction cases?

Sahil Gandhi 50:42
That's a good question. So we unfortunately did not get by we didn't get number of judges through different areas of expertise. We just got district, total number of district judges, and we have 0.04 number of judges per 1000 the population and there is one standard deviation increase in the number of judges, then we can see falls by 0.4 percentage points. So it's much lesser than the impact that land division has

Richard Green 51:17
But still meaningful. Yeah. And none of us, by the way, wants to see people evicted who shouldn't be evicted. Like none of us wants to see people who are paying their rent on time, be evicted, none of us wants to see people who complain about what are genuinely poor living conditions be retaliated against by eviction.

Michael Manville 51:36
One thing I might say just reiterating Richard's point is just that one can be very concerned about inappropriate evictions, retaliation, you know, things like that, and still believe that it is possible to have, in some circumstances, an eviction process that is too slow, right? That is, I think, in some areas of housing discourse, that's like a, "I just got cancelled", right like that's a horrible take, there should be no evictions that should take forever. But as Richard said before, like if we want to have rental stock, then the landlord has to have some assurance that they will actually make a living off that rental stock. And there are situations and of course, everyone is on the spectrum, right. I mean, you go back to right after the financial crisis of 2008, where it didn't take too many judges in Florida to foreclose on people's homes, because they were just running these sort of, you know, factory courts, that's bad. On the other hand, you can search around and find these horror stories in Berkeley and so forth of someone who moves into a house, doesn't want to move out, doesn't want to pay, and the owner is just stuck, not making any money for a year or 18 months or things like that, that's also bad. One message we can draw from this paper is that if you do create a regulatory environment, and it is possible to create such a regulatory environment, where a landlord looks at their property and just says, it's actually better for me that this is empty is less risky, that's counterproductive. And I will just as a final point, just say, this finding that Sahil and Richard have have discovered is not totally out of left field, right. So if you look at the research that was done, when rent control ended in Massachusetts in the 1990s, one thing they found was a falling vacancy rate as a bunch of units were brought back on the market for precisely that reason.

Richard Green 53:31
It does feel like sometimes we get into these debates over if you support rent control, you have to support any kind of rent control. If you think evictions are bad, you have to oppose all evictions of any kind, and I think that just gets you into dangerous and unproductive territory, when it's pretty clear, and I think, you know, Richard and Sahil your work makes clear here, that it just depends - the details really, really matter. And if you get the details wrong, the impacts can be very bad. And if you get the details, right, then you know, maybe the benefits can outweigh the costs and it just it totally depends. I wonder if either of you have anything to add before we close out here,

Sahil Gandhi 54:11
This whole thing about the political economy of rent control change in India. So for example, Maharashtra made an attempt to change the rent control act in the state, but however, so, we know that Maharashtra is the first generation rent control and there are tenants who are happy to be in the house. And so, they never want to let go of the unit. But there is a market that has emerged where they sell their tenancy rights to another tenant informally and they sell it at 70% of the market value which is called the key money or poverty in in Maharashtra. So a lot of rent control property has seen ownership change, where there is a new tenant and that change has been made in the landlords book the ledger where this name has become the other name. Now this, these new tenants who've paid 70% of the market value have kind of played a market value, and they are the biggest group that stops any rent control change because they have paid the market value in South Bombay. So when there is any change in the rent code, when when the government wants to change the act, these people try their best to stop any change in the Rent Control Act. And they lobby with the state government, ministers....

Shane Phillips 55:40
.... because, you know, they were kind of victimized by it in a way, but now that they've paid their money, they're like, "Well, I've, I've bought my way in, and now if you take it away, I'll have spent all that money, and I won't get these protections anymore".

Sahil Gandhi 55:52
Exactly, yeah, exactly! So these people are the ones who will add the new tenants who have paid 70% of the market value, and....

Shane Phillips 56:01
It kind of reminds me of homeowners in a way, where it's like, you know, you buy your home, and you don't want the value to fall, and you do all these things to prevent, you know, competition and so forth.

Sahil Gandhi 56:12
Right, so whenever there is a new change, there has to be a way of dealing with these tenants. And it's difficult to do that because of the Act says that the tenants who bought their way through the key money will not get affected. Everyone will start saying that we are the ones who paid the key money, and it's difficult for the landlord to also back these. So there is this whole political economy of the rent control change that's playing out in major cities. And that's being a major hindrance to any change.

Michael Manville 56:48
That's a great point. And I think, you know, key money has long been an aspect of New York City's rent controlled market, and to a lesser extent in California. But I think the two things to take away from that are that you really, you can only make your key money back if your time horizon is long because it's a big upfront payment to the rent-controlled or the slightly above rent-controlled price is only lower if you can stay five, six years or something like that. And it also just goes back to, you know, a recurring issue with rent control, which is just that it does breed black markets, most price controls do and low income people, they're very disadvantaged in black markets, right? It's hard for low-income person to come up with market rent in New York, it's impossible for them to come up with a huge cash bribe to get them into a unit. Right, the rent control is for, in many ways, higher income people who can pay these big lump sums.

Richard Green 57:40
This is actually a feature or was at least a feature of a lot of the Indian economy. I remember the first time I went to Mumbai, which was I think, 2004-2005, the taxi meters were set at prices that were so low that it would it would basically cost you like 75 cents to get from the middle of Mumbai, up to the airport. Of course you can't drive a taxi and charge 75 cents and get people that distance, you lose money and so when you get in the cab, you make a deal. And so there was the meter ran but it was nothing to do with what you actually paid. So is that still the case or has that changed?

Sahil Gandhi 58:23
I'm actually surprised to hear that to be honest, because always taxi service pretty good. So I don't know if I...

Shane Phillips 58:31
Just got scammed, Richard.

Sahil Gandhi 58:34
That's what I mean. Yeah.

Richard Green 58:36
I still thought I didn't pay very much for how far I was taken. But maybe I was taken in more ways than one.

Shane Phillips 58:46
Okay, well, we will close it out there. As promised, we will do our best to have a another episode that gets into second generation and in more detail and its impacts as well. But Sahil and Richard, thank you so much for coming on the podcast today.

Sahil Gandhi 59:00
Thank you so much for having us, Shane and Mike

Richard Green 59:03
Yes, thank you. It's been a pleasure.

Shane Phillips 59:05
Thanks, guys.

You can read more about Sahil and Richards research on our website. lewis.ucla.edu show notes and a transcript of the interview are there too. The UCLA Lewis Center is on Facebook and Twitter. I'm on Twitter at Shane D. Phillips. And Mike is at Michael Manville six. Thanks again for listening. We'll see you next time.