A few years ago research revealed that Black hosts on Airbnb were earning less than their white peers.  It was a finding that prompted the platform to change its algorithm, with the revenue gap narrowed as a result.

Research from the Rotman School at the University of Toronto highlights, however, that the voluntary algorithm was only successful in narrowing the gap when Black hosts adopted it.

Smart pricing

The Smart Pricing tool was introduced by Airbnb in 2015 to help hosts in optimizing the prices of their properties by better account for guest demand.  The tool, when switched on by the host, will automatically adjust rates within parameters set by the host.

The research found that hosts who turned the tool on saw monthly occupancy rates rise, with nightly rates falling on average.  Together, this resulted in a boost to revenue of around 9% on average.  What’s more, this was a feature that benefited Black hosts the most, with this largely because their demand was around 20% lower than it should be before using the tool.

The tool managed to reduce the revenue gap between Black and white hosts by 71%, but was unfortunately not removing it entirely.  What’s more, as Black hosts appeared to be 41% less likely to use the tool than white hosts, the income gap for them was even greater than before the tool was introduced.

Race blindness

The researchers suggest that the rules preventing AI-based algorithms from making racial distinctions in their algorithms may actually help to create this scenario whereby marginalized races are further behind than they were before.

“Because the algorithm is race-blind, it produces prices that are closer to white hosts’ optimal price than Black hosts’ optimal price. It’s blind but it’s not fair,” they say.

They argue that developers could bypass these limitations in the law by including socio-economic information into the algorithms that clearly correlate with race.  Of course, the platform could also do more to encourage Black hosts to use the Smart Pricing tool as well.