Hacker News Reader

Comment by datadata | Hacker News Reader

How would you tie housing costs to cars?

Car-oriented development seems to scale efficiently - bringing in a lot of new, cheap land - up until you hit the level of traffic that people will tolerate. Then it's done. New neighborhoods take too long to commute from. Densifying existing neighborhoods threatens parking, which people who live there still need, because they still live in an overall car-dependent metro. Overlaying public transit doesn't do much either, since people's origins and destinations are evenly dispersed instead of clustering around stations like they would in a real public transit city. Without transportation capacity there's no growth, and without growth it's a zero sum competition for existing homes.

You can get a whole hell of a lot further before you hit the physical limits of public transit network architectures.

Housing density becomes a problem above a certain level if everyone expects to use a car. Parking and traffic become annoying if not crippling, so communities widely support and enact bans on denser residential development.

Restricting supply like this can be fine if demand doesn't grow, but it can be a huge problem for housing affordability if your local economy is booming.

Easy, supply and demand. Before cars, dense housing was the norm, but cars popularized spreading everything out as much as possible so you have room for wide roads and lots of parking.

Less density -> less housing overall -> decreased supply -> increased prices

Less density doesn't imply less housing, especially when empty land is cheap, as it typically is in the USA. You just use more space.

I hate cars, and I hate how expensive housing is, but as far as I can tell, housing is even more expensive in places (in the USA anyway) where cars are less necessary.

This is mostly due to supply and demand. Disregarding market forces, low density housing is some of the most expensive out there. This is due to higher per-unit construction, utility, transportation, and infrastructure costs.

Unfortunately, low density construction is the default in the USA. High density housing gets bogged down in reviews and permitting, and medium density housing is effectively illegal in most cities in the country.

If we had actually been building medium and high density housing over the past century, the housing problems we're experiencing today wouldn't be nearly as bad. But that housing is sadly rare, and we now have to play catch-up.

The problem is that the supply/demand argument is that the demand largely isn't from people, it's from investment firms. And investment firms don't give two hoots about neighborhoods, density, cars, or anything other than the property's value appreciating over time (which it will, since we can't (without spending gobs of money) create more land).

Also, my personal opinion: I never want to live in a dense, or even semi-dense population center again. Living in a single family detached home with a small yard is fantastic. I can use speakers.

Condos are much worse speculation vehicles compared to houses, since most of what you own is (depreciating) structure compared to (appreciating) land. To the extent that you want housing for people vs. speculators, you want multifamily.

Do you have a citation for this? My intuition is that the construction of low density track housing would have far less regulatory burden than high density housing. It also has a (more) robust labor force in the US especially compared with high density construction, a framing carpenter does not a steel fitter make.

> High density housing gets bogged down in reviews and permitting,

Yes, because it's more high stakes and should be subject to a greater level of scrutiny than a 1-2 story single family dwelling. Check out the Millennium tower debacle or that apartment building that collapsed in Florida if you think otherwise.

> Do you have a citation for this?

I don't have a citation for construction costs, and I will concede that there are many factors that may make some low density developments cheaper per unit then some higher density development.

However, construction is only part of the picture. Low density development imposes terrible financial costs for cities that contain them. That's because the taxable value per square mile is much lower, but infrastructure (sewer, roads, etc) has the same per-mile maintenance and replacement costs. The infrastructure maintenance costs in many low density developments is often greater then the amount of tax revenue the development generates.

For more information on the hidden costs of low density development, I can't recommend this video [1] by Not Just Bikes enough.

[1] https://youtu.be/7IsMeKl-Sv0

> has the same per-mile maintenance and replacement costs.

No it doesn't, the cost to tear up a street in a high density area is much higher than it is in a low density area, especially if we are looking at hidden economic costs.

You're comparing only the densest places (i.e. expensive places with mostly apartments) to only the least dense places, giant houses on big lots. Look at a pre-1950s / railroad neighborhood of basically any smaller/older city in the US, you'll often find a walkable, affordable neighborhood of single family homes.

Also - cheap land does not mean that it's cheap to build and maintain. Almost all of the infrastructure required to build (sewer, power, roads, etc) scale up with area, not density.

You may want to look into housing prices in european or asian cities :)

Less drivability -> harder to move out to suburbia -> more demand for in-city housing -> increased prices

You should spend some time and look at what housing prior to cars was really like as you are very wrong that "dense housing was the norm".

Prior to cars, there were large "estate" houses with separate carriage houses for the horses. These homes often sat on very large plots of land. Prior to that, was old farm houses, where a neighbour could be a KM away?

You seem to be thinking of "row housing" which at least in north america was more a post-WWII design?

> Prior to cars, there were large "estate" houses with separate carriage houses for the horses.

Maybe for the rich? I don't think most people could afford that.

> You seem to be thinking of "row housing" which at least in north america was more a post-WWII design?

They are substantially older than that: https://philadelphiaencyclopedia.org/essays/row-houses/

Not to mention the increased cost of maintaining the infrastructure for all of that suburban sprawl. The farther apart things are the more asphalt, water and sewage pipes, fire department coverage, school buses and drivers, street lamps, etc. are required. That money has to come from somewhere.

Strong Towns and Not Just Bikes has a great series on this: https://youtube.com/playlist?list=PLJp5q-R0lZ0_FCUbeVWK6OGLN...

I'm working on a team to make alternative YouTube recommendations. Here is our list of similar channels for Not Just Bikes which includes dozens of urban design and transit channels like Oh the Urbanity! and Shifter:

https://channelgalaxy.com/id%3DUC0intLFzLaudFG-xAvUEO-A/

> Less density -> less housing overall -> decreased supply -> increased prices

The arrows (logical implication) might be questionable, but it's still true that housing and road infrastructure compete for space.

Subterranean parking costs $40-70k per stall

A developer can only build if they can achieve a 5-6% return on total costs

So 5% of $50k = $2,500 of income, or assuming 30% expenses, $2,500 / 70% margin = $3,500 of income needed per stall per year, or almost $300 per month per stall of underground parking provided

Surface parking is less expensive, like $10-20k/stall but you get the point.

Where do the surface parking costs ($10-20k/stall) come from? That's astonishing to me -- I would guess that a dirt driveway can be almost free, a gravel driveway can be close to free, and a paved driveway costs a few hundred per car space.

Land value. Where I live, 80% of the cost of our property is the land itself, and only 20% is the building or "improvement".

Reply on news.ycombinator.com