[cyberax]

This is not "economy of scale", this is more of a "pollution death spiral" and "unpriced externalities".

A factory that dumps raw waste into rivers is more efficient and can produce widgets more cheaply than a factory that takes care to clean up the waste.

A firm in a dense city core can get talent more easily, and it'll have advantage (on average, in the long run) over a firm that has a fully remote workforce. The negative externalities (living in poor conditions in an expensive city) are offloaded onto the workers. E.g. the housing square footage keeps decreasing: https://www.visualcapitalist.com/charted-the-growing-gap-bet...

[JumpCrisscross]

> negative externalities (living in poor conditions in an expensive city) are offloaded onto the workers

Cities are more efficient in practically every way than subsidized rural developments. It’s really weird to flip that around as an externality. (Disclaimer: I moved from New York to Wyoming. Thanks for your subsidies, I guess.)

There absolutely are jobs in remote places. But the people there aren’t as valuable as someone who will bump into like-minded colleagues and cultural expression as part of their existence in a cluster.

[cyberax]

> Cities are more efficient in practically every way than subsidized rural developments

No, they're not. It's the other way around, in fact. Suburbs subsidize cities (on average, again).

The dense city cores produce most of the _corporate_ income tax, because that's where the companies are headquartered. But most of _personal_ income tax comes from suburbs. This is only now getting close to flipping around.

Cities also have YUUUUGE expenses that simply don't happen in sparse areas. E.g. infrastructure like water or sewer is wildly expensive in cities because of the planning overhead. Case in point: San Francisco spent almost half a billion rebuilding a few blocks of road.

Cities also require EXPENSIVE public transit. One ride on a bus/subway in the US costs around $20. I'm talking about the true cost (number of trips / total expenses), not the farebox cost. And with capex it's almost incalculable, with crazy numbers like $50 per ride for Seattle.

The TLDR; version:

1. Infrastructure in suburbs might end up being a bit more expensive on a per-capita basis. Or it might not, depending on the way you do accounting.

2. In any case, this difference is not at all significant.

3. Suburbs are _not_ subsidized, and in fact generate most of the wealth in the US.

[kasey_junk]

Do you have any citations on that? Very curious because it’s a very heterodox viewpoint you are expressing.

[cyberax]

The thing is, I don't actually know good sources that aggregate all this data. I'm collecting all the references and I'm planning to write a book.

[JumpCrisscross]

If you have a refutation of Triumph of the City, it will be groundbreaking. Until it’s published, throwing in like one credible source would make your comments more serious.

[cyberax]

I read "Triumph of the City" about 15 years ago. It's a well-researched book, and it literally makes the same points as I do. In particular, it describes the density death spiral much better than I can. It's a bit outdated regarding the carbon footprint because it was written before the EVs and self-driving but that's a fairly minor point.

Its biggest problem is that it was written before the negatives of the density death spiral became apparent.

The cities in the early 2000-s were the main benefactors of the global crime collapse. So suddenly, Times Square in NYC transformed from a blighted area into the world's most valuable real estate. How could you be negative about urbanism seeing that? Housing in cities was also relatively cheap because cities used to be less desirable than suburbs.

If I were to summarize, "Triumph of the City" is factually correct. Just as the statement: "coal power plants are a triumph of engineering, providing cheap, reliable, and accessible power".

Edit: another important thing that he's missing is the end of the population growth. So now it's a nearly zero-sum game: a new housing unit in Manhattan means one more dilapidated house in rural Ohio.

[cman1444]

That article has cherrypicked dates.

https://fred.stlouisfed.org/series/COMPSFLAM1FQ

Also, firm in a dense city can get better talent, but those negative externalities (and I would argue these externalities are not clearly caused by the mployer) are made up for through increased salaries the firm must pay compared to areas with lower living costs.

[cyberax]

Care to explain the cherry-picking? The average square footage was growing steadily (except for the 2008 crisis) and the inflection point happened around 2012 when the death spiral started in earnest.

It has been on the downward trend since then.

> Also, firm in a dense city can get better talent, but those negative externalities (and I would argue these externalities are not clearly caused by the mployer) are made up for through increased salaries the firm must pay compared to areas with lower living costs.

Except that the increased salary does NOT compensate for the increased cost of living. And firms absolutely don't pay for their impact on infrastructure.

My favorite example: Seattle. Each household will end up paying around $150k in additional taxes for the new transit that will benefit primarily dense office holders in the downtown.

[cman1444]

> Care to explain the cherry-picking? The average square footage was growing steadily (except for the 2008 crisis) and the inflection point happened around 2012 when the death spiral started in earnest.

It is clearly cherry picked because the graph begins in 2015, which (as you can see in the one I linked) is the exact peak of the new build house sizes. A more honest article would show a larger date range.

Additionally, I feel it makes sense that new builds are getting smaller. People are generally having fewer children these days, so there shouldn't be as much need for huge houses.

> Except that the increased salary does NOT compensate for the increased cost of living. And firms absolutely don't pay for their impact on infrastructure.

It may not entirely make up for the cost of living, but in most cases I would say it certainly makes a big dent. And the rest of the difference can just be explained by the fact that people want to live in cities. So shouldn't it make sense that the cost of living is higher in these areas given that there is more demand for them?

> My favorite example: Seattle. Each household will end up paying around $150k in additional taxes for the new transit that will benefit primarily dense office holders in the downtown.

I think it's strange to frame public transit as primarily benefiting dense office-holders. There are tons of other entities in cities besides just companies in offices. It's a public good, and transit generally benefits everyone. Citizens, companies, government, educational institutions, etc.

Also please link to where I can read more about the $150k number.

[cyberax]

> It is clearly cherry picked because the graph begins in 2015, which (as you can see in the one I linked) is the exact peak of the new build house sizes.

I mean... That was exactly the point.

> A more honest article would show a larger date range.

Sure. It provides an even better illustration about how the runaway urbanism resulted in breaking the previous trend of increasing living area.

> Additionally, I feel it makes sense that new builds are getting smaller. People are generally having fewer children these days, so there shouldn't be as much need for huge houses.

You might have cause and effect confused here.

> I think it's strange to frame public transit as primarily benefiting dense office-holders.

Why is it strange? In the counter-factual world without the dense office core, the new transit would have been useless.

> There are tons of other entities in cities besides just companies in offices.

And none of them requires the amount of traffic that justifies $150k per household.

> Also please link to where I can read more about the $150k number.

https://mynorthwest.com/kiro-opinion/sound-transit-3-34-5-bi...

[cman1444]

>I mean... That was exactly the point.

Great. So we agree it is cherry picked.

> Sure. It provides an even better illustration about how the runaway urbanism resulted in breaking the previous trend of increasing living area.

I don't follow the logic here. Why would 2015 be the tipping point? What's so special about that year? The US has been increasingly urbanizing since WW2.

https://www.macrotrends.net/global-metrics/countries/usa/uni...

> You might have cause and effect confused here.

Just to be clear, you're suggesting that people are having fewer kids (fertility had a recent peak in 2007) because new builds (which are just a small percentage of total housing stock!) have lost a few square feet? I find that to be laughable.

> Why is it strange? In the counter-factual world without the dense office core, the new transit would have been useless.

1. There are more than just offices in CBDs. There are also retail, restaurants, entertainment venues, residential, etc.

2. This ignores the benefit to the worker. And employer/employee relationship goes two-ways. You assign all the benefits to the employers (who gain access to a larger talent pool), while ignoring the fact that this benefits workers too by increasing the pool of jobs they can reliably and comfortably commute to.

3. Economic gains for employers are not purely extractive. A thriving and dynamic centeral business district helps provide income for residents, drive increased investment, and foster innovation.

> And none of them requires the amount of traffic that justifies $150k per household.

Sports venues. Concerts. Festivals. Parades. Any large gathering of people. Not to mention that it's still a good thing to have for general use to go anywhere, even at non-max-capacity times.

https://mynorthwest.com/kiro-opinion/sound-transit-3-34-5-bi...

Not seeing that $150k per household number anywhere in that article. Looks like $10k per person is the number they come up with. Expensive to be sure, but not quite the number you somehow arrived at.

[cyberax]

> Great. So we agree it is cherry picked.

Got it. You have no idea what you're talking about. "Cherry-picking" means finding parts of data that go against the actual trend but "confirm" your point.

> I don't follow the logic here. Why would 2015 be the tipping point? What's so special about that year? The US has been increasingly urbanizing since WW2.

The tipping point was around 2010-2011, it just took some time to become visible. I love bureaucracy; it's the greatest invention of humanity. I traced the policy changes that started the enshittification in earnest to around that time.

For example, that's the year when Seattle's DoT changed the road scoring method so that a field with trenches would get a perfect score. SF started its war-on-cars around the same time and passed the "Twitter exemption".

> Just to be clear, you're suggesting that people are having fewer kids (fertility had a recent peak in 2007) because new builds (which are just a small percentage of total housing stock!) have lost a few square feet? I find that to be laughable.

People are having fewer kids because inner cities are inimical to having _multiple_ children. One child is fine, but 3 children are almost impossible without a parent staying at home.

It's not the _sole_ reason, but it very much is _a_ reason. And you can trace that correlation throughout all the developed world, including the bike hellscape of the Netherlands. And transit availability only _strengthens_ the correlation.

> Sports venues. Concerts. Festivals. Parades. Any large gathering of people.

So basically: nothing. Nothing here justifies the ginormous expense and the associated deterioration in the quality of life.

> 1. There are more than just offices in CBDs. There are also retail, restaurants, entertainment venues, residential, etc.

No, there is not. Entertainment venues, restaraunts, etc. can just as much stay distributed through the city.

> 2. This ignores the benefit to the worker. And employer/employee relationship goes two-ways. You assign all the benefits to the employers (who gain access to a larger talent pool), while ignoring the fact that this benefits workers too by increasing the pool of jobs they can reliably and comfortably commute to.

Ha. I have numbers for that ( https://www.nature.com/articles/s42949-021-00020-2 ). A worker in the 80-s US could access 2x workplaces compared to their counterpart in Europe within a 30-minute commute. Because of cars.

If you shackle yourself to transit, your world rapidly shrinks into tiny "beads" of accessible area outside of stations. While a car provides you true point-to-point travel freedom.

> Not seeing that $150k per household number anywhere in that article. Looks like $10k per person is the number they come up with.

This is the _additional_ expense per _person_. I have calculations from the recent estimates.