What Housing Shortage, Indeed?

Before we get to the main content of this post, I wanted to let you all know that I recently co-hosted my California YIMBY colleague Nolan Gray’s podcast, Abundance. Our guest for this episode was none other than The Atlantic’s Jerusalem Demsas, one of the best reporters covering urban policy today.

You can watch a video of the episode below or find an audio version at your neighborhood podcast dispensary. Despite how it may appear on the video, Nolan and I were not holding Jerusalem hostage in the basement of California YIMBY global HQ when we recorded this interview. Also, I don’t know why I hunched over like that whenever I started talking; that’s apparently just something I do when I’m thinking.

Despite the quietly unhinged visuals, I thought it was a great conversation, and I hope you will too.

Anyway, on with the show.

For subscribers based in New York, I highly recommend this Mara Gay op-ed on the Hochul plan to upzone New York’s suburbs. Gay has been putting out a lot of high-quality work on housing policy in and around the city, and this piece in particular is a real humdinger:

Gay’s emphasis on the segregationist roots of “local control” and restrictive zoning is well earned, but it was bound to ruffle a few feathers. And many of those feathers belong to Kevin Williamson of The Dispatch, who published a detailed response to Gay on Friday:

You can probably guess whose side I’m inclined to pick in this dispute. The most I can say for Williamson’s foray into housing policy is that it isn’t his worst take — that honor probably goes to the time he said that receiving an abortion should be punished with a trip to the gallows. But it’s still a pretty bad take.

The fundamental issue with Williamson’s latest take is that he misunderstands the nature of housing crises. He makes a lot out of New York State’s vacancy rate, while admitting that most of the low-cost and vacant housing is concentrated in areas where people aren’t choosing to live. But the regional misallocation of housing is the crisis. The fact that Buffalo’s metropolitan area has a good amount of relatively cheap housing is simply immaterial to the question of whether New York City’s posh Long Island suburbs needs to build more.

This point is perhaps difficult to understand if you’re the national correspondent for a Substack publication and you can work from pretty much anywhere. But most people don’t have that luxury; they need to go where the jobs are.

The jobs are overwhelmingly in New York City. Based on a little back of the envelope math I did with these figures, it seems that the five boroughs claim nearly 60 percent of the state’s gross domestic product. That’s an extraordinary concentration of economic power, fueled by a variety of factors, not the least of which is the city’s preeminence as a global financial and cultural superpower. But it also reflects a worldwide trend toward urbanization.

One thing we should have learned by now, both in California and New York, is that trying to reverse that trend through an artificial cap on housing supply can only lead to disaster. Yes, California’s restrictive zoning and permitting rules have succeeded in dragging down the state’s population growth, but at the expense of triggering a mass homelessness crisis; driving up carbon emissions; strangling economic growth and innovation; and cutting off economic opportunity for countless people. This is the sort of catastrophic social engineering one might think a small government conservative like Williamson would oppose.

This brings us to Williamson’s other main claim, which is that not even the New York City metropolitan area has a housing shortage. In fact, he says, “it has been estimated that there are four abandoned housing units in New York City for every homeless person or household.”

I was curious about the methodology behind this estimate. Unfortunately, trying to track down the original source for this statistic ended up turning into a whole saga.

It started off, naturally enough, when I clicked the link Williamson cites in his essay. This led me to a Cato Institute report called “How Economists Understand the Damage from Rent Controls.” But oddly, this is not the source of the “four abandoned units for every one homeless person” stat. Instead, the Cato report pulled the estimate from a book: Thomas Sowell’s Basic Economics: A Common Sense Guide to the Economy.

Naturally enough, I decided to consult the book. I was able to find a PDF scan of it online; not a searchable one, sadly. But after some combing through the PDF, I was able to find the relevant passage. It turns out Sowell did not come up with the estimate himself; instead he cites another book, Richard W. White’s Rude Awakenings: What the Homeless Crisis Tells Us.

I was not able to find a full PDF of Rude Awakenings, unfortunately. I had almost given up on tracing Williamson’s citation to its source when I decided, just for the hell of it, to see what I could find in the free Google Books preview. Which is how I dug up this tantalizing little scrap:

Now, the Cato report that Williamson cites was first published in 2018. But when you trace the citation all the way back, you end up at a book that was published in 1992 — more than 30 years ago. God only knows where White got his numbers from (I would have to buy the book in order to figure it out), but I can tell you that his estimate doesn’t correspond to the present situation in New York.

Let’s start with how many homeless people currently reside in New York. HUD’s point-in-time data doesn’t go back to 1992, but the latest PIT counted 61,840 homeless people in the city. That may sound like it matches White’s data pretty closely, but you need to keep in mind that the PIT is, always and everywhere, a massive undercount of the actual homeless population. For example, the city of San Francisco estimates that the number of people who experience homelessness over the course of a year is more than 250% of the PIT count. We don’t have corresponding estimates for New York, but we can assume that the PIT there is also pretty far off from a real census.

As for vacant units, in 2021 — a weird year, one where you would expect vacancies in New York to be unusually high due to peak pandemic migration — the Census Bureau counted 353,400 empty units in the city. But a vanishingly small share of these units could reasonably be called “abandoned,” as you can see from the following chart.

Of the units that are vacant for one reason only, less than 10% are simply being “held as vacant.” Overwhelmingly, these vacancies appear to be either short-term or intermittent ones.

So much for the claim that New York City has a vast reservoir of “abandoned units” ready to be occupied — much less enough of these units to house everyone in the city who is currently homeless. The truth is, the city’s rental vacancy rate is below 5 percent, which is pretty much what you would expect in metropolitan area suffering from runaway housing costs. Both the high housing costs and the low vacancy rate point in the same direction: they tell us that demand for housing in the city is very high, and there isn’t enough of it to go around.

That’s why Hochul’s housing reform package is so important. New York City needs to lift the artificial cap on housing supply, but so does the rest of the metropolitan area—including Williamson’s beloved suburbs. Mara Gay is right.

P.S., if you’re curious to learn more about the methodology behind vacancy rates and point-in-time counts, please see this post I wrote for UCSF’s Benioff Homelessness and Housing Initiative, called “Vacancies are a Red Herring.”