Thread by @dwachsmuth

I'm very excited to announce the release of "The High Cost of Short-term Rentals in New York City", my team's new analysis of Airbnb's impact on NYC. The report is available here:

sharebetter.org/wp-content/upl…

Over the next few tweets I'm going to highlight the key findings.

1/

First, a word about methodology. Using some new techniques for spatial inference (which we'll be publishing in a data science journal), my team has produced what I think is the most rigorous, accurate, comprehensive spatial analysis of short-term rentals to date.

2/

We'll also shortly be releasing our tools as open-source code for other researchers to make use of, to help raise the bar for this kind of research.

3/

So what did we find?

The first question we asked is "Where is Airbnb activity located in New York, and how is it changing?" In the last year there's been a decisive shift in short-term rentals: Manhattan is stagnant or shrinking, and Brooklyn is booming.

4/

STRs in New York City are also an increasingly small share of the entire metropolitan region; in other words, Airbnb growth is happening further and further away from the center.

5/

Across NYC, entire-home listings continue to dominate Airbnb both in terms of # of listings (51%) and revenue earned (75%), but their share of the overall pie is shrinking a little bit.

6/

Here's a key finding: 2/3 of NYC revenue on Airbnb is from illegal listings. 85-89% of entire-home reservations each month are illegal under NY law. That's 45% of all reservations last year, and 66% ($435 million) of all revenue. (In Manhattan, 77% of revenue was illegal.)

7/

We identified a set of "high-revenue" neighborhoods and a set of "high-growth" neighborhoods. In general, Midtown and Lower Manhattan are high-revenue, but growing slowly, while Harlem and North-Central Brooklyn are low-revenue but growing fast.

8/

Ok, on to Q2, "Who makes money from Airbnb in New York?" This is where the major inequality becomes visible.

The top 10% of hosts made 48% of all revenue last year. The median NYC Airbnb host earned $5,200, but the median host from the top 10% earned $33,700—6 times as much.

9/

We were particularly interested in commercial operators of STRs. These are very hard to identify, but one way is to look at hosts with multiple entire homes or a large portfolio of private rooms. These "multi-listing" hosts are 12% of all hosts, but earned 28% of revenue.

10/

Here's a cool map: the different geographies of the four largest commercial operators in NYC. Very different spatial patterns to their listings, and high concentration in particular neighborhoods—STRs is a very local business.

11/

Q3 is "How much housing has Airbnb removed from the market?"

This is where the bad news comes in. We estimate Airbnb has taken between 7,000 and 13,500 units of housing off the market in NYC. These "full-time" STRs are growing much faster than the rest of the Airbnb market.

12/

Most of this housing lost comes from entire-home Airbnb listings which are rented frequently enough that they can't have a primary resident. Depending on how you define the thresholds, there are between 5,600 and 12,200 of these listings, and they're growing rapidly.

13/

But the area where we broke new ground is to identify 4,700 *private-room listings* which spatial analysis tells us are actually whole apartments converted into de facto hostels. We call these "ghost hotels" (h/t @twieditz), and there are 1200 of them in the city.

14/

These 1200 ghost hotels earned $30 million last year, took a minimum of 1400 units of housing off the market, and have flown under the radar of regulators, although they are almost certainly illegal. Here's an Airbnb guest review from one of the larger NYC ghost hotels:

15/

Combining the entire-home listings with the ghost hotels, there are significant swathes of the city where 2% or more of ALL housing has been converted to full-time Airbnb use, which is a scandal when the rental vacancy rate is as low as it is.

16/

Finally, Q4: "Is Airbnb driving gentrification in New York?"

The short answer is "yes", and I have a separate paper undergoing peer review which makes this case in more detail: researchgate.net/publication/31…

17/

What the report demonstrates is that Airbnb has become a major new source of residential rents in NYC. Across much of Manhattan and parts of Brooklyn, something like 5% of all residential rent payments now happen through Airbnb.

18/

This new rental revenue creates powerful incentives for landlords to evict tenants & convert apartments to STRs. The pressure is highest in Harlem, the Lower East Side, and Bed-Stuy—these are where high-frequency Airbnb hosts make far more money than standard landlords.

19/

One consequence of the competition from Airbnb hosts for apartments is that median rents citywide are going up thanks to Airbnb. Applying an existing econometric model to the NYC case, we get a rough estimate of an extra $380/year for the median new rent.

20/

Finally, existing research by @InsideAirbnb and some new findings from our report make it clear that Airbnb's impacts on New York housing are heavily racialized. Murray Cox already demonstrated that it's white hosts making the money in Black neighborhoods in NYC.

21/

What we found is that historically Airbnb revenue is highest in white areas of the city, but that high-growth neighborhoods (where gentrification pressures are the strongest) are disproportionately communities of color. Above all, Bed-Stuy and Harlem.

22/

There's plenty of more information in the report, but I'll wrap up there. Again, it's available for download at sharebetter.org/wp-content/upl….

In the next little while I'll retweet some of the press coverage the report has received, and answer any questions people might have.

23/23