The study found that users of the home sharing platform generally eat in the neighborhood restaurants near where they are staying. However the spillover effect does not hold true when 50 percent or more of a neighborhood’s residents are black or Hispanic.
“Airbnb has made repeated claims that it helps the local economy in black neighborhoods, especially in New York City,” said Mohammad Rahman, a professor at Purdue University’s Krannert School of Management who specializes in the digital economy and big data. “We do not find any evidence of that economic spillover effect in restaurant employment.”
Rahman and his team focused their initial research on the impact of Airbnb on restaurant employment growth in New York City, the most visited, and the most active Airbnb city, in the United States.
They analyzed neighborhood data from the Bureau of Labor Statistics, the Census and Airbnb as well as 3.5 million Yelp reviews of more than 34,000 New York City restaurants between 2005 and 2015 to measure the economic impact home sharing has on local restaurant employment. (The researchers removed neighborhoods with significant tourism activity prior to Airbnb’s 2008 launch and controlled for restaurant popularity and neighborhood characteristics.)
The researchers found that neighborhoods experiencing rapid Airbnb growth typically saw a growth in restaurant employment. Such areas also experienced a surge in their share of Yelp reviews by visitors to New York, a measure that researchers said reconfirmed their employment findings. But restaurants in predominantly black or Latino neighborhoods with high rates of Airbnb bookings did not see a corresponding increase in employment and Yelp reviews.
Researchers said Airbnb visitors may be drawn to more affordable accommodations in black and Hispanic neighborhoods – but not enough to eat there.
“Visitors may not feel comfortable wandering around and checking out restaurants in these minority areas,” Rahman said. “It’s the uncomfortable reality in our society. They may be a lot more careful and just sleep at the Airbnb, coming and going in the comfort of an Uber or a Lyft.”
There could also be a more benign reason for the disparity, he said. “The kinds of restaurants these visitors are looking for may not be present in these neighborhoods.”
The researchers subsequently expanded their study to five other cities and found similar patterns in Austin, Chicago, Los Angeles, Portland and San Francisco.
There was one exception: the impact of home sharing on restaurant employment does extend to majority Hispanic neighborhoods in Los Angeles, where nearly half of residents are Hispanic.
Nick Papas, an Airbnb spokesman, disputed the results of the study, a working paper that he called “deeply flawed.”
“Airbnb undoubtedly boosts local businesses,” Papas said in a statement. “Using a subjective and voluntary input like Yelp reviews to draw conclusions in what purports to be a rigorous analysis is wrong.”
According to the company, Airbnb guests spend 32 percent of their money in the neighborhoods where they stay, and 95 percent of New York City hosts recommend local small businesses to their guests. The company said it’s working to highlight local businesses, partnering last month with the Queens Chamber of Commerce to sponsor a tourism mixer where Airbnb hosts mingled with representatives from neighboring restaurants and shops. And in New York, Papas said Airbnb guests are growing at a faster rate in predominantly black neighborhoods than they are citywide.
“We are now creating significant economic opportunities in communities of color that have historically been ignored by hotels,” Papas said.
Benjamin Edelman, a Harvard Business School professor who studies the economics of online markets, has conducted his own research into discrimination on the Airbnb platform. He was not involved in Rahman’s study, but reviewed the research.
“Airbnb is wrong to assume its service benefits all groups equally. All available evidence is to the contrary,” said Edelman, whose own research found that black hosts earn 12 percent less than non-black hosts with comparable rentals. Another Harvard study found that users whose names are perceived to be black find it 16 percent harder than whites to book lodging on Airbnb.
“If Airbnb wants to do right by minority users, there are plenty of good options at the company’s disposal, most notably removing the information that facilitates discrimination – information that makes race of guest and host needlessly prominent and effectively invites discrimination," Edelman said.
Rahman said he relied predominantly on government employment data – not Yelp reviews – to address concerns about the voluntary nature of online restaurant reviews.
Rahman said he and his colleagues plan to dig deeper into the data and analyze reviews on Yelp and Airbnb for “racial code words,” as well as the races of Airbnb guests, that may offer more insight into the disparity.
He said it’s imperative to understand the full effects of the sharing economy as regulators struggle with ways to frame legislative discussions surrounding its impact.
“We hear a lot of statements about how the digital transformation is blind to the inequalities in society, but we do not live in a vacuum,” Rahman said.
“We cannot regulate and dictate how visitors are going to behave,” he said, “but if we can identify how race is playing a role, there are certain things that can be rectifiable when guests come into these minority neighborhoods.”