Comment by abeppu | Hacker News Reader

Comment by abeppu | Hacker News Reader

Parent Full thread

It's not a binary.

SF gets described as 'housing first', but it certainly does not act like it. In SF, we actually have unspent funds from our 2018 "Prop C" ballot initiative which dedicated funding to specific categories, the largest of which was for permanent housing. I.e. there has been money available for 5 years for permanent housing, but the city is not willing to buy available units -- it will only consider developing new projects, which get stuck in planning hell, and have extremely high per-unit costs. The city is not willing to rent vacant market-rate units, even when large numbers are available, including in rent-stabilized buildings, where it could have long-term predictable rents. When rents dropped sharply during the pandemic, we did not put people into empty, cheap apartments. Instead, we payed $5k/month per tent in parking lot safe sleeping sites.

Mayor Breed is also trying to take funds which are set aside for permanent housing and use them for shelter beds and prevention services.

Funding exists to put more people into actual housing basically immediately, but the city instead pursues more shelter beds and ever larger contracts with Urban Alchemy.

Most of the $5k/tent was in the services required for a high-needs population. As someone who has been impacted by a building set on fire by a meth addicted neighbor, simply placing high needs homeless people who often have serious mental health or drug abuse issues, in apartments isn't the end of the story. In many cases, it makes the lives of other residents of the building worse, as evidenced by the issues (including high rate of nuisance evictions) plaguing SROs in SF.

Santa Clara County tried a random assignment housing first experiment in 2021 where homeless people given apartments died at roughly the same rates from drug overdose, etc. than homeless people who didn't receive housing.

For these reasons and more, it makes a lot of sense to not acquire one-off units in existing buildings for PSH, and Prop C set asides for PSH instead of shelter or whatever the city deemed most useful for homeless people is kind of downstream of nonprofit politics and not a statement about voter mandates or what's best for homeless people.

You seem to be conflating drug use with homelessness. Lots of unhoused people are not on drugs and lots (the vast, vast majority) of drug users have homes.

I think conflating the two is not productive.

70% of San Francisco’s homeless population self reports as struggling with drug abuse or mental illness in the Point in Time count. It’s not a conflation; it’s reality. That’s also why the tent camps cost $5k a month to staff security, medical workers, etc.

If they found that out using a street poll, then they probably missed a lot of homeless who don’t appear homeless, are higher functioning, maybe they are crashing in their car and taking showers at a gym.

We know little about this problem, and most of our information depends on unreliable self reporting.

Maybe you should read the methodology of the point in time count, which does try to find people living in marginal housing.

You provided no link to a methodology. But if it’s like other point in time surveys, it is only looking at people on the street or using services, and are well known for selection bias.

You have provided no evidence either and your claims are vague and not derived from any named study. In any case, PIT as operated by SF HSH can be found here, which certainly you could have reached with a simple Google search of the terms I used: https://hsh.sfgov.org/about/research-and-reports/pit-hic/.

In addition, people who are not legible to PIT methodology are certainly not rolled up into state level statistics upon which this claim of failure and the article rest, so in many ways their existence is not relevant to my point or this discussion.

There is plenty of literature describing why this methodology isn’t great:

Here you go from my region: https://snohomishcountywa.gov/2857/Point-In-Time

https://www.bloomberg.com/news/articles/2019-03-04/the-probl...

> In addition, people who are not legible to PIT methodology are certainly not rolled up into state level statistics upon which this claim of failure and the article rest, so in many ways their existence is not relevant to my point or this discussion.

OK, I agree that PIT just counts what it counts. But then that means any homeless that aren’t counted by PIT aren’t relevant. It’s just bad data that we can’t really work with.

I’m going to be honest: I can’t find a single way the criticisms of PIT you’ve linked relate to the conversation at all. We are discussing why the $5k/mo SF spent on tent villages is incommensurable to studio rents. Even if it were the case that PIT overrepresented people with mental health or drug abuse issues, it’s still true that the 10k or so homeless individuals counted by it had a 70% rate of reporting those problems. Which means housing them is not a matter of simply paying rent.

I'm glad they aren't just buying up existing housing stock. We have a housing shortage, and subsidizing demand will just further send prices up for everyone else. Net new housing, even if more expensive, will do far more good overall.

I'm all for net new housing. And I think we should green-light new residential projects faster. But telling currently homeless people to _wait_ for new projects to be planned and built, when there's already a glut of vacant condos struggling to sell seems cruel. I'm not saying the city needs to bid aggressively -- but when the market slows down, why not get a deal?

Reply on news.ycombinator.com