But short-term it'll hurt availability. One of the most common ways to enshittify cities is buying an SFH, demolishing it, and plopping a 3-4 apartment complex in its place.
I don’t think cities should force density, but there’s no reason to go the other way either as long as growth is managed properly in terms of traffic, transit, and infrastructure.
Nope. The ONLY way to get cheaper housing is by reducing the city population.
> Supply and demand.
Sigh. No. You assume that the demand is fixed. It's not. By building new housing, you _increase_ the demand. And always faster than you can satisfy it.
I just love this example:
Forbes 2016 - "Tokyo's Affordable Housing Strategy: Build, Build, Build", "The Great Urban Myth: 'Cities Can't Build Their Way To Affordable Housing'"
Reuters 2023 - "Surging Tokyo property prices squeeze out young professionals",
Japandaily 2025: "Housing Crisis: Families Struggle to Buy Homes in Tokyo" ( https://japandaily.jp/housing-crisis-families-struggle-to-bu... )
> Rentals in a dense complex are cheaper than standalone rentals or duplexes, townhomes are cheaper than SFH, etc.
Try to find a city where dense housing made it cheaper.
The problem with this is that you can’t forcibly reduce population. Halting growth in one area just displaces it. This can be a good trade if the area into which growth is displaced is underdeveloped with few other potential uses. But it can also create Atlanta or Houston style sprawl, or destroy natural areas that would be better to preserve.
Japan is an unusual example because most job opportunities are in a handful of areas. You can get impossibly cheap properties in the countryside but there are few jobs or young people. Properties in the city keep getting more expensive because of demand and because building codes are constantly evolving to keep up with new developments in earthquake protection. There are older urban properties available for cheap outside of trendy areas, but the cost of renovation is often too high to be worth it. There’s also a cultural stigma around older properties.
We don't need to. The population growth in the Western world is mostly over, the US will likely peak at barely +10% to the current population numbers. Europe is likely already at or near the peak number.
And Tokyo managed to get a bubble within a country with a _falling_ population. Not just stagnant like in Europe, but actually numerically decreasing.
> Halting growth in one area just displaces it.
That's EXACTLY what we need. The US already has 1.1 houses per household, except that they are not where the demand is.
So the fix is to shift the demand, not try to satisfy it. Ironically, this is exactly the same method that urbanists propose to fight congestion: instead of just adding more lanes to busy roads, you shift people to other modes of transportation.
How can this be done? Exactly like we did it with pollution: tax negative externalities, incentivize clean technology. For cities: tax dense office space (cap-and-trade can also work), incentivize work-from-home, incentivize offices in less dense cities, etc.
This doesn't even have to be a huge policy shift.
If the combination of AI, declining returns on service and knowledge-based work, and national security priorities create a resurgence in manufacturing then factory towns may make a comeback.
During the pandemic, some research had shown that something like 70-80% of jobs can be remote.
The kicker is that remote work is less efficient. So in the long run and on average, companies with in-office jobs outperform fully remote companies. This is just like the situation with pollution: a company that spends money on waste recovery is less competitive than a company that can just dump toxic sludge into a nearby river.
And just like with pollution, centralized regulatory changes are needed so that all companies are affected similarly.
> The policies you suggest also seem difficult politically since most environmentalists seem to want more density, not less.
Yet they are misguided because they keep looking at the very tip of an iceberg. It's just one example out of many (see: nuclear power plants).
To give an example, light rail is more efficient than individual EVs. So it's great that Seattle is building light rail, right? It'll result in fewer "headline" CO2 emissions.
But then you realize that Seattle is going to spend $180B ("B", as in "billion") to build about 50 miles of tracks. It's more than the yearly GDP of 130 countries! All these resources could have been spent on something else, perhaps on building more renewable generation.
In the real world, though, a Georgist style LVT probably has about as much chance to be enacted as any of these other policies. Unfortunately I think we’re going to run the current system into the ground.
I have not run the numbers, but intuitively it seems that this tax will kill the SFH long before it starts affecting the dense office space.
> In the real world, though, a Georgist style LVT probably has about as much chance to be enacted as any of these other policies. Unfortunately I think we’re going to run the current system into the ground.
I think there is a real chance, there is this current of massive dissatisfaction. People _feel_ that something is just not right with the current situation. With the populists proposing the usual easy solutions: "it's all immigrants, ICE them out" or "it's all fault of the end stage capitalism, we need socialized grocery stores".
Well yes, but that's not at all something intrinsic to the process, it's just a dysfunction that Seattle is suffering — along with most of the rest of North America.
In 2002, Toronto paid less than $1b to build the 5.5km of https://en.wikipedia.org/wiki/Line_4_Sheppard , which is fully underground subway running on a custom gauge for historical reasons.
Last year, we finished https://en.wikipedia.org/wiki/Line_6_Finch_West , which is light rail running at-grade on standard gauge. It's not even twice as long but cost 3.6x as much. For light rail. And it's apparently running well below design speed and at least initially with terrible signal priority.
Currently we are building https://en.wikipedia.org/wiki/Line_5_Eglinton, just over five times as long. It's light rail at a combination of underground, at grade and elevated; current projected total cost is $17.5b. So, three and a half times as much per unit distance, for what is supposed to be a considerably less expensive option.
Prices for other things have not gone so crazy in that time frame. But yeah, it isn't costing us anywhere near 3.6b USD/mile (about 3.1b CAD/km) for light rail. Yet.
Not just light rail, but even regular water and sewer. San Francisco spent half a decade repaving a few blocks (Van Ness bus rapid transit) because they had to slowly dig through unmapped ancient infrastructure.
Even in the case of Toronto, you're looking at amounts that can buy each incremental rider a new house. In case of Seattle, each household is going to pay around $150k for that rail.
It's simply ridiculous.
Are there cities where replacing denser housing with single-family homes made housing cheaper?
A really good example is Copenhagen, the world's most liveable city. Its current population is still _less_ than during the 1970's peak: https://www.macrotrends.net/global-metrics/cities/20894/cope... The driver for the decrease was suburban migration, as cars became more accessible.
The US _itself_ is a great example. The suburban development created cheap housing for the rapidly growing population in 60-s.
Obviously housing built outside the city will be cheaper than within it, and that might work for people who are fine living just anywhere (like away from the city), but the hypothesis is that the prices within the city increase when more housing (denser housing) is added within the city, right?
It's just hard to imagine how replacing a 300-unit occupied building downtown with 4-8 single-family-homes would result in the SFHs being cheaper than a unit in the skyscraper was.
0: https://www.courthousenews.com/copenhagen-housing-prices-dou...
I don't have all the information on Copenhagen yet. The stats from 1970-s are not available online, so I commissioned someone to get the data from the archives.
The available data basically shows that prices were stagnant during the 70-80-s and started rising in the 90-s.
> After all, building more housing in the city isn't mutually exclusive with building housing outside the city.
I think it is mutually exclusive, exactly because of the population growth (the lack thereof). Each dense apartment in a city core means one less house in a rural area somewhere.
Japan, that I gave as an example, has literally free houses that anyone can get for nothing but the government real estate transaction fees. Just 3-4 hours away from Tokyo.
Is there any evidence of this? The demand for living in the city (rather than outside of it) already exceeds the supply.
> Japan, that I gave as an example, has literally free houses that anyone can get for nothing but the government real estate transaction fees. Just 3-4 hours away from Tokyo
Yeah, but that isn't housing in the city, it's housing in a place that isn't the city. Even setting aside the cultural differences of Japan, where homes are rebuilt every couple decades, and even setting aside that it has a declining population: a place where people don't want to live will naturally have lower prices than a place where they do.
It kind of sounds like when you're talking about reducing the density of a city, you're actually referring to keeping the density of the city the same, and building SFHs in places that aren't the city (which actually increases the density of those non-city places). Is that what you mean? Or did Tokyo replace skyscrapers with SFHs? And were those SFHs within the city cheaper than a unit in the skyscrapers was?
But that's not true, is it? Most people in the US want to live in suburban SFHs, yet they are often forced to live in apartments. But that's not a viable option for them because the jobs are only available in dense cities.
> It kind of sounds like when you're talking about reducing the density of a city, you're actually referring to keeping the density of the city the same, and building SFHs in places that aren't the city
Correct.
> which actually increases the density of those non-city places
And technically increases the housing price there from zero to some value, just as predicted :)
> But that's not true, is it?
It certainly is: look at the price of housing in New York City, then look at the price of housing in Newark, New Jersey. Many people want to live in the former, but must settle for the latter, due to lack of affordable NYC housing. Then look at the price of housing in Ainsworth, Nebraska: Even cheaper, because people want to live there even less.
Or look at your own example: People want to live in Tokyo more than they want to live 3-4 hours outside of Tokyo, hence the pricing for the latter is lower.
> Most people in the US want to live in suburban SFHs, yet they are often forced to live in apartments.
In my experience, most people in the US want lots of square footage within the city, and either settle for suburbia to get the square footage they want, or settle for less square footage to get the city living they want. This goes for both renters and buyers, and for both Single- and Multi-Family Housing.
How, then, would increasing the price of in-city housing (by reducing the supply, by replacing denser housing with less-dense housing) allow people to realize their big-city-big-living desires?
Alternatively, how would building housing hours and hours and hours outside the city (where pricing illustrates people don't want to live) allow people to realize their big-city-big-living desires?
That's because they _have_ to work in NYC. Polls show that something like 80-85% of people in the US would prefer to live in suburbs.
> Then look at the price of housing in say, Cheyenne, Wyoming: Even cheaper, because people want to live there even less.
Well, yes. Land is not scarce in Cheyenne, so housing is cheap. But people don't flock there because they don't have any job prospects in WY. See my notes about remote work.
> Or look at your own example: People want to live in Tokyo more than they want to live 3-4 hours outside of Tokyo, hence the pricing for the latter is lower.
How about: "People HAVE to live in Tokyo, because there are no job prospects outside of Tokyo"?
> In my experience, most people in the US want lots of square footage within the city, and either settle for suburbia to get the square footage they want, or settle for less square footage to get the city living they want. This goes for both renters and buyers.
Well, sure. I would love to live in a mansion with a private lake, in the middle of Union Square.
> How, then, would increasing the price of in-city housing (by reducing the supply, by replacing denser housing with less-dense housing) allow them to realize their big-city-big-living desires?
You assume that people _want_ to live in big cities. People want to have access to big city amenities, but not necessarily live there all the time.
That's why suburbs are such a desirable place. Think about this: what if you live 1 hour away from the city core? This still allows you to easily enjoy all the amenities like theaters and shows. Or to periodically go to your favorite ethnic restaraunt. But it's _way_ too far for daily commutes so it's impractical.
> How about: "People HAVE to live in Tokyo, because there are no job prospects outside of Tokyo"?
How about: people don't want to be located 3+ hours outside the city, and neither do many employers? Being fully-employed in the middle of nowhere is still pretty undesirable. This is borne out in housing prices: the people buying houses in Cheyenne are likely just as employed as the people buying houses in Newark or NYC, and yet the market rate for the former is still much lower.
> That's because they _have_ to work in NYC.
Or, because they want to live in NYC: socialization and meetups and camaraderie around every conceivable interest, innumerable dining options within a matter of blocks, some of the best live entertainment options in the world, friends within walking distance, etc. Case in point:
> I would love to live in a mansion with a private lake, in the middle of Union Square.
The reason you can't have that is the cost, which we're talking about ways to decrease.
> That's why suburbs are such a desirable place
The suburbs are desirable in that they are cheaper: that mansion with a private lake would be cheaper in the suburbs than in the city, so some people choose to settle for the suburbs, while others choose to settle for a less-nice place in the city.
> what if you live 1 hour away from the city core? This still allows you to easily enjoy all the amenities like theaters and shows.
1 hour each-way of driving is a lot, and by then you're usually in a different city. 3-4 hours puts you in a different state, and in some places, it puts you in an entirely different country!
Do they? Or do they just rationalize having to live in NYC?
Depending on the question, most people HATE living in cities: https://www.reddit.com/r/dataisbeautiful/comments/1l1ryyq/ci... - and it's not even a close competition.
> socialization and meetups and camaraderie around every conceivable interest, innumerable dining options within a matter of blocks, some of the best live entertainment options in the world, friends within walking distance, etc. Case in point
Well, sure. Everything has trade-offs and people decide which ones are more acceptable for them. When given a choice, I moved out of dense areas. Because I don't actually care about drinking every day, and all other activities are not materially affected by having a 1-hour commute to the downtown.
> 1 hour each-way of driving is a lot, and by then you're usually in a different city. 3-4 hours puts you in a different state, and in some places, it puts you in an entirely different country!
Sure. But we're talking about the US in particular.
It seems much more realistic to freeze growth than reverse it. Even then, growth in surrounding areas or other factors can quickly make the area more desirable and expensive as in SF.
This is the first time I see this bold claim, what evidence do you have to prove it?
The only price decreases happened only during the 2008 crisis and during the pandemic lockdowns, due to local population decreasing.