This doesn't take into account something mentioned in the article - that increased housing supply is going to attract MORE of the highly skilled workers that were driving up costs, and those people are going to be the ones filling up the houses, not poorer people.
This is a form of induced demand.
What really you are saying is we don't want to build enough for all the unmet demand at the current price point. This is a perfectly fine thing. Roads are expensive, if we can get people to not drive someplace because the current roads are too crowded that might be a worthwhile thing. Unlimited road expansion quickly requires expensive bridges.
Building more houses actually starts to fix the road problem because dense enough houses makes public transportation practical.
Yes rich folks will move into San Francisco at first. However their old house isn't in middle of nowhere, it is a longer commute away from San Francisco, and that house is available.
Here is a report from UC Davis that shows otherwise: http://www.dot.ca.gov/research/researchreports/reports/2015/...
But big picture: more people got more of what they wanted, which to drive farther (without more congestion).
If they built an extra lane on every road between me and the pizza shop, I'd go the pizza shop more. I dont' go to the pizza shop as often as I'd like because I hate traffic, if there were less traffic, I'd be better off. But I won't drive any more than I'd like to, I think we could build more roads and build that capaity for more people.
There’s certainly a place for cars in city planning (lower density areas), but there’s also a point where planning focuses too much on cars to the detriment of everyone else.
Only for specific definitions of "efficient". You get higher throughput, and with that higher latency.
Not everyone needs a whole car, though. Motorcycles, for example, would be a more efficient use of roads, increasing throughput without sacrificing latency, and also decreasing the parking footprint necessary per capita (these are, in fact, the two primary reasons why I'm strongly considering getting an M-class endorsement and buying a motorcycle, or at the very least getting a motorized scooter that doesn't require said M-class endorsement; the tertiary reason of "motorcycles are fucking badass" certainly doesn't hurt the former option, either). Should be no surprise that in most dense cities you see outside the US, motorcycles and scooters compose a significant portion of road traffic.
If you have to drive a car to a transit hub, wait 10-15 minutes for a vehicle, transfer multiple times (with wait times for each), then walk more than ~5 blocks to your final destination it becomes a shitshow. Los Angeles is a perfect example of this.
LA being antithetical to transit is... really a stereotype. The entire city (and even much of the county) is built along the old streetcar lines and the red car network (the largest streetcar system in the world back in the 30s!), which existed until the 60s. It’s a city built for transit and a city where transit is being heavily invested in again — because only relying upon cars turned out to be a failed experiment.
As a fun tidbit of things to come:
By the 2028 Olympics, LA is hoping to have completed a metro line connecting the Valley to the Westside via the Sepulveda Pass within 12 minutes — a journey which takes the average driver during rush-hour 55-65 minutes to complete on an ever-expanding Highway. Once it’s finished, it’s projected to be the busiest single rail line in the US, surpassing any in New York City even.
It's potentially anywhere on Earth—people moving into San Francisco are not limited to people currently within commuting distance of San Francisco.
> and that house is available.
No, rich people can and do own multiple homes without renting any of them out, and are known to just buy another one if one meeting their interest becomes available. There is no fixed limit on the number of homes a person can retain exclusive use of.
How big of a problem is that really? Is there that many people owning an absurd amount of homes everywhere?
This keeps happening for any practical width of road, because in practice you can't really have a 30 lane wide highway. Automotive travel is inefficient enough that even at suburban density you can always congest any road network you build around it.
This does also apply to housing - even at the highest densities you can get, rivaling Hong Kong, there are too many people in the world for anyone city to ever build enough housing to legitimately run out - density itself creates economic multipliers that attract business, and if you were to have some of the highest density in America (which is largely the claim of NY) you would have effectively unlimited attractants for capital to fund immigrants to move into said city and populate it. In the scenario where you "overbuild" high density housing to the point where the next poorest local cannot afford it at its upkeep costs you would attract foreign immigration for the cheap housing with ample employment opportunity.
Note that that last scenario assumes all other variables are held constant - desirability, transit supply per-person, availability of public services like fire and police, etc. Missing other variables can make high density areas undesirable for capital and thus prevent the chain reaction.
Roads can go up (or down) quite a bit though. It is expensive (much more expensive than adding more lanes) which is why we rarely do it. Even when/where we have, we would have to go more than we have to get ahead.
My point is you can get ahead of induced demand. I know a farmer who planted right across a public road - technically illegal, but the neighbors didn't use that road anyway. When you get in a city is is harder - by enough that it isn't worth it - but you can. (Cities should of course have more public transit which gets ahead much quicker).
You are correct that cities building - at affordable rates - do attract people. However if the city allows building over the long term we know from existing examples it gets out of it. Chicago and Huston don't have housing price problems because you can build there.
Now we do need to be clear: it costs money to build. New houses are always higher end. How high depends, if only a few houses can be built, they will be mansions. Once demand for mansions is filled builders move down market to lower profit houses until it is no longer feasible to build anything (humans don't fit in a doll house).
The paper is no only slightly better supported by data than my arguments above and really isn't useful.
For every X number of expensive apartments, builders must sell Y% low income housing.
Even if you only care about the low income stuff, upzoning still helps, and gets you what you want.