“How could you see my listing if you’re blind?”

“I’d have to check with our insurance company to see if we’re covered to host guests with disabilities.”

“Does the dog drive?”

Those are three typical responses we got from Airbnb hosts while posing as guests with disabilities for a study we conducted on the home-sharing service. Some hosts were willing to accommodate us. Some were uneasy. Some were insensitive. We effectively became Airbnb’s secret shopper – even secret to Airbnb – to determine if its credo to “belong anywhere” implied that this service was designed with disability access and civil rights in mind.

Our results revealed that Airbnb’s platform perpetuates the social exclusion of people with disabilities. We don’t believe this is done intentionally, but the unregulated nature of rental listings end up subverting the goals of the Americans with Disabilities Act, which turns 30 in July.

What the ADA changed

Before this landmark law, people with disabilities had a very hard time engaging in American life. Structural barriers like buildings without elevators made it difficult to use public transit and limited where people with disabilities could work. They were essentially rendered second-class citizens.

For example, people using wheelchairs had to abandon them if they needed to ride trains or buses. Grocery stores and other buildings were usually not accessible to people with disabilities, and restaurants even refused to serve them. Public schools excluded an estimated 1 million children with disabilities. And employers could legally avoid hiring someone because of his or her disability – and paid them less than their able-bodied peers with similar qualifications.

This began to change in 1990 with the passage of the ADA, which prohibits disability-based discrimination in all areas of public life. The law makes sure that people with disabilities have the same rights and opportunities as those without disabilities. Particularly, Title III of the ADA covers public accommodations – including private hotels – and requires them to make appropriate changes in policies, practices and procedures, unless this would radically change how businesses operate.

The law, however, was never designed to cover private citizens, such as Airbnb hosts. The recent rise of the largely unregulated sharing economy thus complicates whether the ADA applies to these new types of largely person-to-person transactions. It’s a world in which workers are rarely “employees,” and businesses are often just regular people sharing their apartment with a stranger.

Traditionally, the ADA prohibits businesses like hotels from discriminating against people with disabilities. However, Airbnb is not a normal hospitality company that manages and franchises a collection of hotels and resorts. Instead, it’s a broker between hosts who temporarily sublet their homes and guests who seek affordable and unconventional places to stay.

To that end, listings are not hotels either. And the ADA specifically applies only to places with more than five rooms to rent and are not occupied by the homeowner as a place of residence. These conditions are seldom met on Airbnb. Therefore, many hosts are not legally prohibited from discriminating, which means that guests with disabilities are subject to pre-ADA conditions.

Documenting discrimination

We began our study in June 2016, investigating access to nearly 4,000 Airbnb listings by requesting lodgings with the use of simulated people who had disabilities or were able-bodied. Inconspicuously, across the U.S., we solicited hosts with fake guests who experience blindness, cerebral palsy, spinal cord injury, dwarfism or no disability. The study concluded in November 2016, and we analyzed host responses through early 2017. Following a peer review, we published our findings in February 2019.

Discrimination against our disabled guests was evident. Those without disabilities were offered preapprovals – that is, following an initial inquiry about availability, hosts may make approval automatic once a guest requests a booking – 75% of the time, whereas those with disabilities had a much harder time, depending on the disability.

Those with dwarfism were preapproved 61% of the time, while people with blindness were at 50%. Guests suffering from cerebral palsy were at 43%. And having a spinal cord injury meant a preapproval rate of just 25%.

Overall, the more extreme the disability, like using a wheelchair, the more discrimination our disabled “guests” endured.

‘How do you drive?’

Through the process of seeking preapprovals, we engaged with every host in our review, which gave us insights into how they reacted to people with disabilities.

Some hosts were extremely positive and offered assistance. For example, one eagerly explained to our “guest” with a spinal cord injury, “I can carry you and your chair up the stairs… I really want you to stay.”

Another kindly replied to a traveler with dwarfism, “We would be glad to modify anything as needed.”

But some hosts weren’t as accommodating. For example, one expressed concerns over cleaning costs specific to a traveler with blindness who uses a guide dog, “[I]f you’re willing to pay $100 for animal cleaning, I would be OK with you staying.” This was on top of the location’s typical cleaning fee assessed to all guests.

A second was especially disrespectful toward a traveler with blindness by replying, “Um. That’s a new one. How do you drive?”

And while a third was a bit more empathetic, the host was still dismissive toward a traveler with cerebral palsy, blaming an architectural constraint. “Our place has a very narrow and circular stairway, so it would be too difficult for you,” the host said.

Airbnb’s efforts to fix the problem

To Airbnb’s credit, it had recognized the problem of disability access in its listing before we began our investigation and began implementing changes during the study. This allowed us to see in real time what kind of impact it had.

The company announced a nondiscrimination policy in November 2016 that explicitly banned using a guest’s disabilities in turning down a requested stay, in addition to other characteristics like race, national origin, religion, sexual orientation and marital status.

However, our study found that compliance didn’t vary after the policy was implemented.

At the time, we felt that this was due to how nascent the policy was. However, as recently as 2019, there have been anecdotes detailing its weak enforcement. Haben Girma, a disability rights lawyer who is deaf and blind, claims she was denied lodging after disclosing that her service animal would accompany her.

The host defended himself by indicating that his rental was under construction and would be hazardous for any stay, yet the listing was successfully booked by an able-bodied colleague of Girma for the same dates.

Soon after our study, in late 2017, Airbnb purchased Accomable, a travel website focused on those with disabilities. This acquisition helped Airbnb improve its search filter by allowing hosts to list accessibility features more precisely.

It remains to be seen how effective this has been, though some guests have found the accessibility filters to be inaccurate.

Time for an ADA update

The ADA emerged from a long history of frustration of how people with disabilities were marginalized in the areas of employment and more.

Now, as the sharing economy continues to expand, we believe the law should be amended to specifically protect people from disability-based discrimination on these online platforms. While it would not be realistic or sensible to force every mom-and-pop listing on Airbnb to become ADA-compliant, there must be ways to ensure that when guests without disabilities have access to listings, those with disabilities can have access to similar ones at the same cost.

As the ADA approaches its 30th anniversary, this is an opportunity for legislators, disability rights activists, Airbnb, as well as its users to collectively and proactively ensure equal access for everyone.

[Insight, in your inbox each day. You can get it with The Conversation’s email newsletter.]

The authors do not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and have disclosed no relevant affiliations beyond their academic appointment.