The article is excellent. I find that it doesn't mention one other effect that people often ignore. I live in the Cambridge/Somerville area of MA where there is huge demand for upscale housing by people moving to the area to work mainly, but not only, in the pharmaceutical industry. In Cambridge and to a lesser extent (so far) Somerville there is also a huge NIMBY population who attempt to block all new market rate housing exactly as described in the article. Obviously, that doesn't reduce the demand, so what do people looking for housing do? They do the rational next best thing which is to buy modest housing, which in Somerville at least is still relatively plentiful, and put a couple of hundred thousand dollars into fixing them up to the standards that they desire. After a few dozen of these around places like Union and Davis Squares, you end up with neighborhoods that formerly had Fords and Toyotas out front to Audis, Lexus and BMWs.
The upshot is that these new neighborhoods are now unaffordable by most people's standards exactly like housing NIMBYs would have been trying to prevent. One can imagine that if "luxury" housing had been allowed how much less demand would have been put on the existing stock.
So I don't think you know that housing is not a strictly market based enterprise. Its "strict market" is interfered with on numerous levels. First, is incentives to builders, there's the cheating of the contractors (best exemplified by Trump who would not pay his contractors and then also sue them out of business), then there are buyers who buy for investment and never live there (including many billions of dollars of drug or other illegal profit money), last but not least are corporate investments, also involving billions of dollars.
So new construction was NEVER a strict market and still isn't and therefore doesn't follow strict market principles, even in the aggregate. Look at someplace like Vancouver, where market rents and housing prices increased exponentially (and far above affordability) when literally tens of thousands of new units went up (one single development was about 50,000 new condos in 1990, something like 1/2 of which stood empty until last year) bought mostly by investors from Hong Kong who were creating a bolt hole for themselves should Hong Kong ever get really awful (like it is now under Xi Jinping). The market for this housing was never Vancouver itself but really it was Hong Kong with Hong Kong prices and layouts and other concerns (like light and density).
Got to consider two exceptions, and one further induction effect;
1. Sometimes, the swanky new build ends up as second homes, airbnbs, buy to lets, or even buy-to-hold-empty (often foreign buyers wanting a bolthole). And sometimes the developer spends a ton of money on marketing the new builds to oursiders, generating new demand in the area. These dont help the locals wanting a place to live. Thinking of eg cornwall and london, in the uk, as examples of these factors.
2. What was on the site beforehand, and what were the alternative uses? If the answers involve affordableish housing then yes the swanky new stuff is stiffing the locals who just eant a place to live.
3. Because swanky new developments do induce some demand from higher income groups, they support the influx of related retail, leisure, services, employment. Which makes them more of a target for further swanky new developments. Developers keen to grab bits of older housing stock, spare land, and public spaces. More pressure on the modes of land use that were supporting the locals who have been there for decades. More traffic and poorer air quality for the incumbents too. Bit of a feedback loop. Examples abound across london.
Outside of these caveats i'd agree that development can be positive for local communities eg when more housing means more demand for retail and leisure and means that the village or town can support a few more shops, bars, eateries, and leisure facilities.
This is a fine post but one of the difficulties in assessing the thesis is that the lag time between building and downward rent pressure can be on the order of years. In our area of Bethesda, MD , building of new dwelling units (90% or more are apartments) has been ongoing for the past dozen years. From our balcony, we have views of four building cranes for new apartment construction. It takes 18 - 24 months from groundbreaking (in all cases these buildings have required demolition of older buildings that in some cases was older housing) to completion.
On our street there are over a dozen four story apartment buildings constructed in the early 1950s (I'm not sure about the total occupancy) that fall in the "moderate rent" category. These are all slated for demolition with high rise units to be built, but this will be a 13 year process so we can't know what the impact on rents will be for some time.
Our county does have a regulation requiring a certain number of "affordable" units to be part of new construction . This can also ameliorate rent prices. I'm in general agreement with the points Noah made in his post but don't think the evidence, at least in the experience of our area, is totally rock solid.
England is the perfect counter to those saying if we force people to build sub-market rate, but don’t expand the significantly stock in area with high demand. After several years the properties and up at market rate, which has only continued to go up. This holds even for public housing due to ‘right to buy’
There are cities that build a lot where rents stay cheap.
There are cities that don't built a lot where rents skyrocket.
There are a few superstar cities that build a lot and yet get more expensive anyways because their job market is so hot that they are having trouble keeping up. This is legitimately a problem, but it's a much better problem to have than having your key local industry collapse, like what happened to Detroit.
There is literally no such thing as a city that's cheap but doesn't build a lot. _Even Detroit_ actually makes it relatively easy to build (tearing down abandoned units to do new construction), and that combined with the cheap underlying land makes housing very cheap.
Actually I thought Noah Smith's article was interesting and informative. I am going to read it again later today.
But I thought and still think the issue is more complex than as presented.
Where I live we have zoning. Some seem to think they should have some control over what is in their back yard. I am one of them.
I live in a quasi-rural area and we want to keep it that way. But the demographics are changing and the prior prevailing opinion may change. One has to accept that. Sometimes what you do not want turns out later to be something you would not want to part with.
A discussion board should not be set up so that dissenting opinions are discouraged. Generally we benefit from discussion. Cause and effect are always difficult to deal with.
I think that part of the issue is that the USA makes it so easy to anonymously park one’s money in real estate, so, at least from what I understand to occur in NYC, despite luxury apartments being built and bought, they languish empty because the wealthy have purchased them just as an item to hold like a piece of art until it appreciates in value, and maybe occasionally use in lieu of hotel, but otherwise just keep off the market. New luxury housing then does not cause the rents to rise, but there are perverse incentives to build more luxury housing than needed which, because city land use is inevitably a zero-sum game, reduces supply of affordable housing.
It is late at night and I barely have the concentration to post this, plus I didn't thoroughly read the article yet, but... this is one of the very, very rare occasions where I differ from Mr. Smith and actually seem to see his easy faults while describing the arguments in place. So I will surely have to come round again and correct myself.... or not.
I've been personally running from gentrification forces for almost three decades now, on a "free market" country (Chile) where certain kinds of growth do get impulses at certain phases in time and so have experienced some of these arguments in the flesh.
So, when Noah says " If you’re a working-class person, and you see a big new shiny glass apartment tower going up a block away, it makes perfect sense to be afraid that rents are about to rise.".
And yes, that is part of the gentrification equation. Not the main cause yes, but part, because higher income renters are not going to live in the neighborhood as it is/was. Firstly renovations galore start, and a shift in priorities cascades; new flashy small shops appear, little cafes then followed by restaurants and tipping points are reached when existing apartment buildings get a % of new owners enough to change the communal policies.
So it is not just umbrellas opening to the coming rain, but a whole other moving factors at play. Like the sense of security and thus having 24 hours guards in some streets do improve their attractiveness (in fact becoming causal then). Plus existing characteristics come into value as well like closeness with transport (metro), to city services and soon that 15 minutes city concept, or even 30 or 45 mins if you move on bike (like myself), become main causes that promote gentrification and hence rises on rent faster there than in other places of the same metropolis.
The question is if those are real existing demand swings are promoted, incentivized ones. Precisely there is where the free-market becomes conducted and steered by partial actors, usually those with the loudest voice. Nimbys or Yimbys may alter the demand and supply too. Marketing and power moves being the not-so-hidden variables both (S & D) actually include in their completer formulation.
Reading further got me again having a revisionist angle on this article, but in fairness to its theoretical coherence (as usual for Noah) I will leave this here for now.
I’m a big fan of most of your articles but I think this article is attacking a straw man. No one (or at least no one smart) is saying that luxury apartments cause rent to go up. The argument misses the fact that high end real estate and low end real estate exist in separate markets. They’re not substitute goods except at the margins
The argument is that if you could increase supply in either the low-cost or high-cost housing, you would lower the market price in that market. Adding luxury housing makes luxury housing a little more affordable, but does not affect low end housing — and does not increase the quantity of housing units exchanged in the low end market.
What do you think the impact would be of building below-market housing?
Nice post. Bur, I’m perplexed by the DC Multifamily market. There has been incredible multi family construction since 2020, between 11,000 and 12,000 units per year in a market, about 34,000 units, where the population actually declined from by 2.6 percent (about 20,000) between 2020 and 2022. Yet average rents have barely moved and there is still pretty robust construction. I’ll grant that older buildings have probably gotten somewhat cheaper, but this large increase in housing units hasn’t had much effect on rents (particularly in the upper end). There is considerable ProPublica had a recent piece about possible collusion in rent setting, which could be part of the puzzle. https://www.propublica.org/article/yieldstar-rent-increase-realpage-rent
Anyway the situation is much different than my experience in Denver in the early 1970’s. In that period Denver experienced a multi family construction boom partially caused by the government ceded a defunct military base to the city, who then sold it yo developers. These new properties came on line within a few months and rents collapsed. Vacancy rates went so high that several properties were held off the market for while. Rent concessions for one year leases were very common, free first month and last month rent, a free bicycle tossed in or a bottle of wine when rent was paid on time. Nothing approaching this has time around in DC.
Obviously, this isn’t a well specified analysis but I remain confused. It seems to me that the large increase in housing stock combined with a population decline should have had a larger effect on rents.
Excellent, well researched article. Thank you.
Big fan of getting these types of posts every 9 months or so with updated research links ;)
Very simple truth: if “luxury apartments” aren’t built than the rich will simply compete for and live the “affordable apartments”
The article is excellent. I find that it doesn't mention one other effect that people often ignore. I live in the Cambridge/Somerville area of MA where there is huge demand for upscale housing by people moving to the area to work mainly, but not only, in the pharmaceutical industry. In Cambridge and to a lesser extent (so far) Somerville there is also a huge NIMBY population who attempt to block all new market rate housing exactly as described in the article. Obviously, that doesn't reduce the demand, so what do people looking for housing do? They do the rational next best thing which is to buy modest housing, which in Somerville at least is still relatively plentiful, and put a couple of hundred thousand dollars into fixing them up to the standards that they desire. After a few dozen of these around places like Union and Davis Squares, you end up with neighborhoods that formerly had Fords and Toyotas out front to Audis, Lexus and BMWs.
The upshot is that these new neighborhoods are now unaffordable by most people's standards exactly like housing NIMBYs would have been trying to prevent. One can imagine that if "luxury" housing had been allowed how much less demand would have been put on the existing stock.
Really good. But:
1. People are selfish
2. How do we incentivize the NIMBYs to change?
As in - what’s in it for them?
Loved this. It's a problem in the UK as well.
So I don't think you know that housing is not a strictly market based enterprise. Its "strict market" is interfered with on numerous levels. First, is incentives to builders, there's the cheating of the contractors (best exemplified by Trump who would not pay his contractors and then also sue them out of business), then there are buyers who buy for investment and never live there (including many billions of dollars of drug or other illegal profit money), last but not least are corporate investments, also involving billions of dollars.
So new construction was NEVER a strict market and still isn't and therefore doesn't follow strict market principles, even in the aggregate. Look at someplace like Vancouver, where market rents and housing prices increased exponentially (and far above affordability) when literally tens of thousands of new units went up (one single development was about 50,000 new condos in 1990, something like 1/2 of which stood empty until last year) bought mostly by investors from Hong Kong who were creating a bolt hole for themselves should Hong Kong ever get really awful (like it is now under Xi Jinping). The market for this housing was never Vancouver itself but really it was Hong Kong with Hong Kong prices and layouts and other concerns (like light and density).
Interesting piece.
I'd agree as a general case.
Got to consider two exceptions, and one further induction effect;
1. Sometimes, the swanky new build ends up as second homes, airbnbs, buy to lets, or even buy-to-hold-empty (often foreign buyers wanting a bolthole). And sometimes the developer spends a ton of money on marketing the new builds to oursiders, generating new demand in the area. These dont help the locals wanting a place to live. Thinking of eg cornwall and london, in the uk, as examples of these factors.
2. What was on the site beforehand, and what were the alternative uses? If the answers involve affordableish housing then yes the swanky new stuff is stiffing the locals who just eant a place to live.
3. Because swanky new developments do induce some demand from higher income groups, they support the influx of related retail, leisure, services, employment. Which makes them more of a target for further swanky new developments. Developers keen to grab bits of older housing stock, spare land, and public spaces. More pressure on the modes of land use that were supporting the locals who have been there for decades. More traffic and poorer air quality for the incumbents too. Bit of a feedback loop. Examples abound across london.
Outside of these caveats i'd agree that development can be positive for local communities eg when more housing means more demand for retail and leisure and means that the village or town can support a few more shops, bars, eateries, and leisure facilities.
This is a fine post but one of the difficulties in assessing the thesis is that the lag time between building and downward rent pressure can be on the order of years. In our area of Bethesda, MD , building of new dwelling units (90% or more are apartments) has been ongoing for the past dozen years. From our balcony, we have views of four building cranes for new apartment construction. It takes 18 - 24 months from groundbreaking (in all cases these buildings have required demolition of older buildings that in some cases was older housing) to completion.
On our street there are over a dozen four story apartment buildings constructed in the early 1950s (I'm not sure about the total occupancy) that fall in the "moderate rent" category. These are all slated for demolition with high rise units to be built, but this will be a 13 year process so we can't know what the impact on rents will be for some time.
Our county does have a regulation requiring a certain number of "affordable" units to be part of new construction . This can also ameliorate rent prices. I'm in general agreement with the points Noah made in his post but don't think the evidence, at least in the experience of our area, is totally rock solid.
England is the perfect counter to those saying if we force people to build sub-market rate, but don’t expand the significantly stock in area with high demand. After several years the properties and up at market rate, which has only continued to go up. This holds even for public housing due to ‘right to buy’
One graph that should be included in this kind of review: https://twitter.com/JeremiahDJohns/status/1615352365040766976
There are cities that build a lot where rents stay cheap.
There are cities that don't built a lot where rents skyrocket.
There are a few superstar cities that build a lot and yet get more expensive anyways because their job market is so hot that they are having trouble keeping up. This is legitimately a problem, but it's a much better problem to have than having your key local industry collapse, like what happened to Detroit.
There is literally no such thing as a city that's cheap but doesn't build a lot. _Even Detroit_ actually makes it relatively easy to build (tearing down abandoned units to do new construction), and that combined with the cheap underlying land makes housing very cheap.
Actually I thought Noah Smith's article was interesting and informative. I am going to read it again later today.
But I thought and still think the issue is more complex than as presented.
Where I live we have zoning. Some seem to think they should have some control over what is in their back yard. I am one of them.
I live in a quasi-rural area and we want to keep it that way. But the demographics are changing and the prior prevailing opinion may change. One has to accept that. Sometimes what you do not want turns out later to be something you would not want to part with.
A discussion board should not be set up so that dissenting opinions are discouraged. Generally we benefit from discussion. Cause and effect are always difficult to deal with.
I think that part of the issue is that the USA makes it so easy to anonymously park one’s money in real estate, so, at least from what I understand to occur in NYC, despite luxury apartments being built and bought, they languish empty because the wealthy have purchased them just as an item to hold like a piece of art until it appreciates in value, and maybe occasionally use in lieu of hotel, but otherwise just keep off the market. New luxury housing then does not cause the rents to rise, but there are perverse incentives to build more luxury housing than needed which, because city land use is inevitably a zero-sum game, reduces supply of affordable housing.
It is late at night and I barely have the concentration to post this, plus I didn't thoroughly read the article yet, but... this is one of the very, very rare occasions where I differ from Mr. Smith and actually seem to see his easy faults while describing the arguments in place. So I will surely have to come round again and correct myself.... or not.
I've been personally running from gentrification forces for almost three decades now, on a "free market" country (Chile) where certain kinds of growth do get impulses at certain phases in time and so have experienced some of these arguments in the flesh.
So, when Noah says " If you’re a working-class person, and you see a big new shiny glass apartment tower going up a block away, it makes perfect sense to be afraid that rents are about to rise.".
And yes, that is part of the gentrification equation. Not the main cause yes, but part, because higher income renters are not going to live in the neighborhood as it is/was. Firstly renovations galore start, and a shift in priorities cascades; new flashy small shops appear, little cafes then followed by restaurants and tipping points are reached when existing apartment buildings get a % of new owners enough to change the communal policies.
So it is not just umbrellas opening to the coming rain, but a whole other moving factors at play. Like the sense of security and thus having 24 hours guards in some streets do improve their attractiveness (in fact becoming causal then). Plus existing characteristics come into value as well like closeness with transport (metro), to city services and soon that 15 minutes city concept, or even 30 or 45 mins if you move on bike (like myself), become main causes that promote gentrification and hence rises on rent faster there than in other places of the same metropolis.
The question is if those are real existing demand swings are promoted, incentivized ones. Precisely there is where the free-market becomes conducted and steered by partial actors, usually those with the loudest voice. Nimbys or Yimbys may alter the demand and supply too. Marketing and power moves being the not-so-hidden variables both (S & D) actually include in their completer formulation.
Reading further got me again having a revisionist angle on this article, but in fairness to its theoretical coherence (as usual for Noah) I will leave this here for now.
Edit; including the morning after "translation".
I’m a big fan of most of your articles but I think this article is attacking a straw man. No one (or at least no one smart) is saying that luxury apartments cause rent to go up. The argument misses the fact that high end real estate and low end real estate exist in separate markets. They’re not substitute goods except at the margins
The argument is that if you could increase supply in either the low-cost or high-cost housing, you would lower the market price in that market. Adding luxury housing makes luxury housing a little more affordable, but does not affect low end housing — and does not increase the quantity of housing units exchanged in the low end market.
What do you think the impact would be of building below-market housing?
Nice post. Bur, I’m perplexed by the DC Multifamily market. There has been incredible multi family construction since 2020, between 11,000 and 12,000 units per year in a market, about 34,000 units, where the population actually declined from by 2.6 percent (about 20,000) between 2020 and 2022. Yet average rents have barely moved and there is still pretty robust construction. I’ll grant that older buildings have probably gotten somewhat cheaper, but this large increase in housing units hasn’t had much effect on rents (particularly in the upper end). There is considerable ProPublica had a recent piece about possible collusion in rent setting, which could be part of the puzzle. https://www.propublica.org/article/yieldstar-rent-increase-realpage-rent
Anyway the situation is much different than my experience in Denver in the early 1970’s. In that period Denver experienced a multi family construction boom partially caused by the government ceded a defunct military base to the city, who then sold it yo developers. These new properties came on line within a few months and rents collapsed. Vacancy rates went so high that several properties were held off the market for while. Rent concessions for one year leases were very common, free first month and last month rent, a free bicycle tossed in or a bottle of wine when rent was paid on time. Nothing approaching this has time around in DC.
Obviously, this isn’t a well specified analysis but I remain confused. It seems to me that the large increase in housing stock combined with a population decline should have had a larger effect on rents.
Thanks for you time
Jack