> I call water safe, but you may say it's a tacit admission that you could drink 5 gallons in an hour an experience zero effects. I wouldn't call that an honest rebuttal

I meant it as an honest rebuttal, and I largely disagree that your analogy applies.

Humans obviously die from hyponatremia if they drink massive amounts of water: in your example, 5 gallons of water, which is 5X-10X someone's daily consumption in a short period of time. OK, I agree.

I don't think the theoretical example of increasing the housing stock in NYC by 54% is likely to lead to serious negative effects, especially if it happens over a period of several years, because the city has astronomically high rents and even higher house prices [1]. I believe the culprit is primarily zoning and other restrictions on building, which I defended in another comment upthread.

I do think radically relaxing zoning laws and regulations could theoretically lead to such an increase in supply within a period of, say, ten years. It is politically infeasible and would require basically every voter to understand that supply and demand pricing works in the housing market, which they don't.

The reason I used that exact number—which was deliberately chosen to be large—is that I simply disagree with your claim, if I am understanding it correctly, and I felt a larger number would drive the point home more.

I interpreted your claim as: adding marginal units to the housing market in NYC will not change the price of housing, relative to the universe in which we do not add more units, because there is alway a marginal person waiting in the wings who would like to move to NYC but hasn't yet.

The thing is, what is actually preventing that marginal person from moving here? There is some vacancy here in NYC, even if it's low at 1.4%. Just adding one more unit with no change in price anywhere should not affect anything. In order for that person to actually go from "I am staying where I am" to "I am moving to NYC," the thing that has to change is the value of that apartment: it must breach the threshold at which point they'd move to NYC. I claim this predominantly happens through increased supply driving down price for a given unit compared to the universe where supply is held constant, which happens with basically any additional housing (via the filtering effect). If the increased supply held rents steady, there is no reason for that in-migration. This is the underlying mechanism by which all supply and demand works from my understanding; I don't see a reason why it should be different solely for housing in exactly one metropolitan area.

Obviously individual decision making for a given person is not this simplistic, but at a population level I think the desirability vs. price calculation holds and the increase in supply decreasing price holds, which is why metro areas that have built a lot (e.g. Tokyo, Austin, Nashville) have even seen falling housing prices, while NYC and SF see rising rents.

I picked a large number of units in my initial claim, but I basically do not think the logic changes for a small number of units either. Adding 100 units relative to baseline would nudge it down a smaller, but still meaningful amount.

[0] https://rentguidelinesboard.cityofnewyork.us/wp-content/uplo...

[1] https://www.apartments.com/rent-market-trends/new-york-ny/#:...

(In a well-functioning real estate market there are some units offline due to vacancy, turnover, and renovation, so we don't expect 0%. High vacancies of e.g. 10% suggest insufficient demand for the supply.)

It was a quick google search and the value was taken from the Gemini result, which is not great practice. Thank you for asking for citations. Unironically! Good discussion on this site matters to me, and I want my positions to have strong foundations.

I looked into it more deeply and have found the actual numbers. The 1910 density was more like 54% higher.

Manhattan reached its peak population of ~2.3M around 1910, with a built-up area on the island of ~4,000 hectares=15.44 square miles. This gives a population density of 149k people per built-up square mile. This is per NYU professor Angel Shlomo's 2015 paper, which seems to be the most in-depth analysis I could find ([0]). We use built-up area because north Manhattan still had large tracts of uninhabited farmland at the time; this area should be excluded to better represent the experience of the residents.

Manhattan's current population is ~1.66M (per [1]), and I am estimating the current built-up area at 4466 hectares=17.24 square miles, as this was the exact built-up area given in [0] for 2013. I am assuming Manhattan was, then and now, basically fully built-up and therefore the area is unchanged. (If I'm wrong, then I would guess there's actually more built-up area now, and therefore the population density is lower, strengthening my POV.)

This gives a current Manhattan population density of 96.3k residents per built square mile. This means the peak density in 1910 was actually (149k/96.3k=)154.7% what it is today! My initial Google estimate was undershooting it.

Ergo Manhattan today is nowhere near its historic density.

From all this, I think it is clear that Manhattan and NYC are nowhere near their carrying capacity. There are other safe, clean metros on planet earth with much higher density and lower housing costs (e.g. Seoul), and the thing preventing greater density and lower costs isn't some fundamental upper bound or even economic incentives: it's just zoning.

What has changed since 1910? New York City has much better infrastructure that would allow even greater density in the city, outer boroughs, and surrounding area without negatively impacting quality of life. Modern building techniques allow us to build far more than six stories high without imperiling residents in tenement housing. There is a robust subway and rail network that allows commuters to come from other, even less dense areas. I think we could build to these historic densities much more safely and pleasantly now, which is also the opinion of the authors of [0].

> And there is never going to be enough supply.

I simply do not understand your rationale. We have not seriously tried to provide enough supply, due to the zoning constraints.

In 2023 there were ~3.7M housing units in NYC ([2]). We are adding ~9,450 units every quarter ([3]), for a total of ~38k units per year, a yearly increase of about 1%. This is a drastic undershooting of NYC's own goal to build 500,000 new homes from 2022 to 2032, which would look more like 13k to 14k new units per quarter ([3])—and some groups estimate that the housing shortage is even more than that 500k units. It takes 3.4 years on average to build a unit in NYC and more than four years to build an apartment in Manhattan ([3]), which is absurdly slow, and the vacancy rate is an absurdly low 1.4% where ~5% is the marker for a healthy real estate market ([2]+general knowledge).

Manhattan (and NYC in general) has onerous zoning and review requirements that prevent the adequate building of new housing to meet demand. Of the existing building stock, fully 40% of it would not be permitted nowadays, primarily because they are "too tall" or "have too many apartments," "too many businesses," or any number of other absurd requirements in a city with extremely high housing costs ([4]). There are even regulations around having enough parking spots in parts of Manhattan, a place with ample public transit, lots of street parking, and subsidized van service for the disabled. These buildings are the backbone of our urban core; they provide housing for hundreds of thousands of people and commercial space that drives much of the NYC metro area's economy, the GDP of which is over $2T and makes up ~9% of the US economy ([5]).

Conversely, there are many land uses in Manhattan that fit these zoning requirements and are ridiculous and uneconomical uses of land. For example: a small parking lot structure in the West Village (of which there are a depressing number) is a valid usage of land, while an eight-floor apartment building with an elevator and retail space on the bottom is "too tall" and disallowed because it would exceed the 80-foot limit in the village. How does another small parking lot serve a city gripped by housing crisis that is caused by a shortage of units—which also has excellent rail connection and plenty of parking already—better than an apartment building? The apartment building could even be on top of a parking garage if we wanted it to, but adding the residential space and/or more commercial space (which raises incomes via agglomeration and therefore improves affordability by higher salaries relative to housing) is not allowed because it runs afoul of height restrictions.

This is irrational policy that directly contributes to the housing crisis.

> Your only choice is to Detroitify your city.

It's interesting that you claim densifying Manhattan will turn it into Detroit.

I claim that Detroit was hollowed out precisely by anti-density and pro-suburbanization practices. I'm lazy and have been typing forever at this point, so: https://en.wikipedia.org/wiki/Decline_of_Detroit#Population_....

When you send factories to the suburbs and then put highways straight through the neighborhoods in your city where people actually live and work, that's not densification leading to bad outcomes. It is detonating a city via discriminatory urban policy and then falsely claiming it is the natural end result for those pesky cities and their residents.

[0] https://marroninstitute.nyu.edu/uploads/content/Manhattan_De...

[1] https://censusreporter.org/profiles/06000US3606144919-manhat...

[2] https://www.nyc.gov/content/tenantprotection/pages/fast-fact...

[3] https://www.nytimes.com/2025/12/23/nyregion/nyc-housing-deve...

[4] https://www.nytimes.com/interactive/2016/05/19/upshot/forty-...

[5] https://edc.nyc/sites/default/files/2025-12/NYCEDC-2025-Stat...

(This is a more readable but much less detailed intro to [0]: https://urbanomnibus.net/2014/10/the-rise-and-fall-of-manhat...).

Would love to see some justification for this claim, which tacitly suggests that adding, say, 2M extra apartments over baseline would have zero effect on our housing market.

The market prices full one-bedrooms at less than this in most of Manhattan. Flophouse beds would cost a fraction of this and they would get cheaper the more of them you have. They’d also slow the growth of rent price in NYC.

Let’s take your example at face value. Suppose Manhattan added one million beds to its existing ~3.8M bedroom housing stock via the “missing middle” housing that you described, perhaps over the next ten years. These might have small private bedrooms and shared kitchen/office/bathroom facilities. They might even include dorms, but I’ll focus on single-room occupancy units. You’d get the space for this from some mix of re-developing office buildings or upzoning or re-developing low-slung buildings.

The rent of these single-room occupancy units would be a fraction of the rent of a normal place. The people living there would be far less rent-burdened than they would have been otherwise, freeing up more income for consumption or savings/investment, boosting economic activity. Some of them would be new residents, whose income taxes (if they pay them) and spent dollars/sales taxes would be a net benefit to the city budget. Some of them would otherwise become transiently homeless due to affordability concerns, which would be destabilizing for them and expensive for the city due to homeless program spending.

Others would be people who currently live in apartments but would move to these units because they prefer cheaper rent, greater privacy (they might be sharing a room today), a newer building, or the greater efficiency of having multiple bathrooms. Maybe right now they are sharing e.g. a four-bedroom one-bathroom apartment with three strangers in Hell’s Kitchen for $1400/month. These people would otherwise be in the housing market for a full apartment, and removing them leaves more full apartments for people who want to occupy them, either alone or with roommates. Ergo we get downward pressure on full apartment rents.

The flophouses and dorms and SROs were a key part of the housing market that kept Manhattan more affordable and therefore livable in the 20th century, when density was up to 40% greater than it is now. It was deeply shortsighted to get rid of them. The idea that we should downzone even further makes no sense to me; you get to decrease affordability and decrease the economic benefits of agglomeration to the local economy at the same time, all so… there are fewer people on the subway, I guess? I disagree with the “too much density” argument on its face anyway. Density has clear economic benefits via agglomeration and productivity gains; diverse and dense housing stock via upzoning increases affordability via supply/demand and filtering effects; and the way you manage density is through appropriate infrastructure spending on housing, services, and public spaces which—you guessed it!-becomes cheaper per person the denser you build. Seoul has twice the density of Manhattan.

I live here. The thing making Manhattan unlivable is that a one-bedroom is $4500 in the east village due to not enough supply. Fix the housing costs by building more of any and all kinds of housing and then we can deal with the other problems via better governance and increased tax revenues. There’s nothing we can do otherwise that isn’t just rationing or some other bandaid solution.

There is an unbelievable shortage of housing that is solvable only by increasing supply and building upwards. It’s not even single-family homes; why are there any one-story buildings in the lower east side?

[0] https://hsph.harvard.edu/news/usaid-shutdown-has-led-to-hund...

Comments URL: https://news.ycombinator.com/item?id=47317347

Points: 9

# Comments: 3

Every additional coping mechanism you add, that works well for you, provides defense in depth.

If so, that is a risk that is very dependent on local laws. I would be much more cautious about seeking mental health care in Florida, where the Baker Act makes it very easy, than in Connecticut [0]. The risk of being involuntarily committed also, of course, has to be balanced against the risk of forgoing mental health care when you need it and then injuring or killing yourself.

[0] https://psychiatryonline.org/doi/10.1176/appi.ps.201900477

[0] https://www.bls.gov/news.release/cpi.nr0.htm [1] https://tradingeconomics.com/united-states/gdp-growth

1) The US was rebuilt after WW2 to be highly car-centric, with disastrous results for pedestrian safety. It is actually more dangerous for children to move in public because of this automobile centricity in >95% of US spaces, which drives cultural shifts

2) US media is highly sensationalist; Fox News is popular and generally serves to make viewers frightened of the world

3) The US has 340M people to Ireland’s 5.6M people, so from a base rate, we’d expect ~60x more lurid headlines to come out of the US

I understand asking $700 for this feels icky, but the only reason they can do it is because of the insane scarcity of housing in SF. If we want this to not be a competitive option, the way forward is just to build more housing - like these people are doing.

Comments URL: https://news.ycombinator.com/item?id=41799761

Points: 4

# Comments: 0

What an awful coincidence. I can’t even imagine how it must feel to have a freak technical accident deprive you of seeing your father for the last time.