Did Airbnb purge potentially illegal NYC listings ahead of its own data dump?
After tracking Airbnb listing data for almost all of 2015, Murray Cox and Tom Slee, independent researchers, found that in an early December Airbnb disclosure of NYC listing information, Airbnb removed about 1,000 listings of “entire unit” rentals from its platform.
Airbnb quietly removed 1,000 listings in New York City — potentially illegal rentals of entire units — from its platform in early November, just before a significant, well-publicized release of its own data, according to a pair of independent researchers who collected data over a one year period.
Murray Cox and Tom Slee, independent researchers who published the data to InsideAirbnb.com, tracked Airbnb listing data for almost all of 2015. They took monthly snapshots of all the Airbnb listings in New York and tracked the data over time.
As New York City rules currently read, it is illegal to rent out an entire unit in the city for 30 days or less — exactly the kind of rentals that are a key part of Airbnb’s business. Airbnb wants to draw up a formal agreement with the city to “legalize” short-term home rentals, but New York City lawmakers have resisted, saying that the company is far too lax on Airbnb hosts who have multiple listings on the service.
In today’s release, Cox and Slee say that in an early December Airbnb disclosure of NYC listing information, the company “extrapolated from an artificial and unrepresentative one-time event.” The research suggests Airbnb removed about 1,000 listings of “entire unit” rentals from its platform right before it released data to the public.
“Airbnb removed listings in New York City belonging to [hosts with multiple Airbnb listings] so that the November 17 data set would fit the message the company wanted to deliver,” the pair wrote. “The percentage of traffic going to [hosts with multiple Airbnb listings] has been a constant 30 percent of the total during all 2015 until the days before the November purge, when it dropped to only 20 percent.”
For New York City authorities, the concern is that an oversupply of entire-unit Airbnb listings would effectively take available housing off the market by turning those apartments into full-time short-term rentals.
Officials say Airbnb doesn’t do enough to remove these listings from the platform, and Airbnb says that it is working to be more transparent with cities about how it regulates the service. In November, it published a “Community Compact” outlining broad-stroke promises about transparency with local governments, and on Dec. 1 it released a trove of data about NYC listings.
The crux of Cox and Slee’s charge of misdirection against Airbnb is how the purge affects the data released in December.
When asked if Airbnb intentionally removed 1,000 listings prior to releasing the NYC data on Dec. 1 (and whether such an action would be unusual for the service), spokesperson Chris Nulty said, “We routinely review our listings to ensure guests are having the quality, local experience they expect and deserve.”
Airbnb’s early December data dump included figures on Airbnb listings from November 2014 through Nov. 1, 2015 alongside a more detailed snapshot of listing data on the day of Nov. 17, 2015. The disclosures were a move to demonstrate Airbnb’s transparency to long-skeptical New York lawmakers, attempting to “make clear that illegal hotels are not welcome on Airbnb.”
The larger, year-long data set was made available for viewing to journalists and interested parties by appointment at Airbnb’s New York City office. The Nov. 17 snapshot has much more useful information than the other data, providing raw numbers and percentages of Airbnb listings, and whether their hosts operated multiple entire units on the service. The single-day snapshot reflects data that happened post-purge, and Cox and Slee contend that it’s an “artificial and unrepresentative sample” of how many entire-unit listings were listed on Airbnb for most of 2015.
Airbnb: We’re Playing by the Rules
In response to Cox and Slee’s allegations, an Airbnb spokesperson told Re/code that listing numbers frequently drop because of the constant “churn” of people posting and removing listings on the platform. In a statement, Airbnb affirmed that the vast majority of hosts on the platforms are hosts who play by Airbnb’s rules, renting out entire units when they’re not home as supplemental cash. They aren’t, Airbnb says, de facto landlords who rent out multiple listings on the service:
The facts are clear for all to see — the vast majority of our hosts are everyday people who have just one listing and share their space a few nights a month to help make ends meet. Airbnb is an open people-to-people platform where listings come on and go off throughout the year. We’ve also done significant work to educate our community about what is in the best interest of their city and we routinely review our listings to ensure guests are having the quality, local experience they expect and deserve.
What Airbnb has maintained all along, and maintained loudly in its December NYC data release, is that the overwhelming number of hosts who rent out units on the platform do so in a quasi-kosher way: They only rent out one unit. Though it’s assumed that if Airbnb is formally legalized in New York that renting out a single whole unit will be permissible, the law currently on the books says that it’s illegal do that for less than 30 days at a time. And to be clear, lawmakers don’t seem to care much about targeting people who are renting out their entire primary residences to make some money on the side.
Some New York City council members — notably Helen Rosenthal and Jumaane Williams — and other dogged critics of Airbnb say that the company’s “most of our hosts only rent out one unit” defense doesn’t hold water. They argue that it doesn’t take into account the big portion of revenue generated by the small portion of hosts who operate multiple entire-unit listings.
A few days after the Dec. 1 data dump, The Verge found that hosts with three or more listings (less than two percent of the total host population) accounted for 24 percent of the revenue generated between November 2014 and Nov. 1, 2015. A week later, Quartz reported that Airbnb expects “by this time next year, entire-space hosts with one or two listings will be earning 93 percent of revenue.”
So What About Those 1,000 Missing Listings?
Between Nov. 1 and Nov. 15, 2015, Cox and Slee’s data show that Airbnb culled 1,438 listings belonging to hosts with two or more units from its platform, from a total of 3,331 such listings the two had recorded. They say that, accommodating for month-to-month churn, roughly 1,000 units were intentionally removed. There was no corresponding decrease in private room rentals, as opposed to entire-unit rentals, during that month, according to their information.
Cox and Slee charge that Airbnb purged the 1,000 listings in order to present a more favorable portrait of its NYC data in its Dec. 1 release. They argue that the only logical conclusion to draw is that the listing bloodletting was timed to substantially change the Nov. 17 snapshot that was released, and that Airbnb will “only expel [multiple listing operators] when their presence is a nuisance to the company.”
The two also say that Airbnb’s pledge to reduce the revenue generated by multiple-listing hosts from 25 percent to 7 percent is unrealistic, as “levels of multiple-listing entire homes have already jumped back to 13 percent [from an immediately-post-purge 10.3 percent] of total listings, only two months after the purge.” Airbnb provided Re/code with data showing that the number of single-listing hosts stayed at around 94 percent of the total population between Nov. 17, 2015 and Feb. 8, 2016, which Airbnb spokesperson Christopher Nulty says is proof that there isn’t a glut of multiple-listing operators:
Our community in New York has evolved to a point where 94 percent of hosts have just one listing and where there is no material presence of illegal hotels, which is why accusations from the same elected officials who called for there to be no illegal hotels on the platform and now want to fine middle class families $50,000 is akin to asking someone to walk on water and then, when they do, fining them for not swimming.”
Nulty also pointed out that Airbnb doesn’t ever disclose when listings get pulled from the platform, excepting a 2014 removal of 2,000 listings that was forced by a legal battle with New York State Attorney General Eric Schneiderman*. As for what kind of policies guide which listings Airbnb pulls when it decides to yank them, Nulty directed Re/code to the Nov. 11 “Community Compact,” which spells out broader transparency commitments to the communities in which Airbnb operates.
95 Percent Might Be Okay. But What About the Other 5?
Cox and Slee, whose report was circulated widely by NYC officials critical of Airbnb, say they believe Airbnb is “confused about its own data.” Cox, a New York-based Australian photojournalist (Slee is an ex-tech industry employee with a Ph.D. in theoretical chemistry), says that “when they talk about 95 and 94 percent of all listings being for single units, they’re trying to hide the fact that there are hosts with 3, 4 or 5 listings.
“Talking about the number of hosts avoids the number of homes that are taken off the market,” Cox said in a phone interview with Re/code. “Ninety-five percent of hosts may be okay, but how are the 5 percent doing? How many houses [listed on Airbnb] do they have?”
Read original article