Who Gets to Share in the "Sharing Economy": Understanding the Patterns of Participation and Exchange in Airbnb

Recently, Airbnb has released a number of studies that claim that the company is having a positive impact in the regions it operates in by driving tourists and revenue to areas underserved by traditional hotels, providing... [ view full abstract ]

Recently, Airbnb has released a number of studies that claim that the company is having a positive impact in the regions it operates in by driving tourists and revenue to areas underserved by traditional hotels, providing lower-income residents with a revenue stream and building up interpersonal trust (Airbnb, 2015). However, these claims have yet to be validated by independent research and they are further complicated by recent findings showing that racial discrimination against guests is a real concern in the platform (Edelman, Luca, & Dan, 2015). Issues of inequality in access and use go beyond the Airbnb platform and have been a consistent concern in both popular and academic literature on the “sharing economy.” Yet we still do not fully understand how these new institutions interact with the existing socio-economic cleavages, at least partly because large scale data on the “sharing economy” has not been easily available. In this paper, we overcome this lack of data using a unique dataset containing information about roughly 125,000 Airbnb listings from 104 metropolitan statistical areas in the US. We match the information found on the Airbnb platform to the US Census data on the census-tracts in which listings are located. This allows us to study how income, race and education impact the economic activity on the platform. We show that while the platform appears to favor lower income and more racially diverse areas at first glance, the true patterns of participation and exchange are much more complicated. Despite the fact that census tracts with higher income and a greater proportion of white people have less listings and fewer reviews, these still have better ratings and demand higher prices. More importantly, we show that education is a heavily influential factor in all of these patterns, suggesting that these new institutions might be benefitting individuals with higher cultural capital disproportionately. While our data has some inherent limitations, because we only study the hosts on the platform and do not have complete information on the exchanges, we nonetheless believe that our findings are a significant contribution to the debate on the sharing economy and its future.