July 25, 2024
How iBuyers are changing real estate racial disparities and individual homeownership rates in one major city
天美影视传媒 researchers investigated how iBuyers 鈥� companies that use automated algorithms to quickly buy and sell homes 鈥� have affected the well-documented racial bias against Black home sellers. In Mecklenburg County, North Carolina, they found that on average iBuyers paid more equal prices to Black and white home sellers than individual buyers, largely because iBuyers paid white sellers significantly less on average than an individual buyer.Blake Wheeler/Unsplash
Instant buyers, also , rapidly buy and sell homes using automated models to set prices. These companies, such as Opendoor and Offerpad, can turn around cash offers in a matter of hours, and they鈥檝e captured more than 5% of the real estate market in some U.S. cities.
Since new tech often replicates or exacerbates existing societal biases, 天美影视传媒 researchers wanted to investigate how iBuyers have affected the well-documented 鈥� particularly .
The team homed in on Charlotte, North Carolina, where an estimated 35% of the population is Black, and where in 2021 iBuyers held more than 8% market share. Based on an analysis of five years of property transactions in Mecklenburg County (which contains Charlotte), researchers found that on average, compared to individual buyers, iBuyers paid more equal prices to Black and white home sellers. That鈥檚 largely because iBuyers paid white sellers significantly less on average than an individual buyer would.
The team also discovered that iBuyers were then significantly less likely to sell homes to individual buyers. Instead, these companies were more likely to sell homes to institutions, such as large rental companies that鈥檝e been tied to high eviction rates and rent gouging.
The team in June at the ACM Conference on Fairness, Accountability, and Transparency, held in Rio de Janeiro.
鈥淚t’s easy for bias to seep into automated models if they鈥檙e trained on data that is itself biased,鈥� said lead author , a UW doctoral student in the Information School. 鈥淭he models that iBuyers use are essentially proprietary black boxes. Given the long history of housing discrimination in the United States, we were concerned that historical biases might be influencing these models behind the scenes, without the public being aware.鈥�
The researchers pulled 50,000 publicly available property transfer records from 2018 to 2023 for Mecklenburg County, population 1.1 million in the last census. The team cross-referenced these transfer records with North Carolina voter rolls, which list each person鈥檚 race. Controlling for 50 factors, including home size and neighborhood crime rate, the team found that on average white-owned homes sold to private buyers for $36,051 more than Black-owned homes. But when homes sold to iBuyers, that difference shrank to $4,436, because iBuyers paid Black homeowners $4,376 more on average, while paying white homeowners $27,239 less.
鈥淭here鈥檚 very little reason for us to believe that there鈥檚 some purposeful intervention going on here,鈥� said senior author , a UW associate professor in the iSchool. 鈥渋Buyers are paying Black homeowners a little bit more, but not significantly more. Rather, iBuyers don’t seem to be willing to pay white homeowners what they might be able to earn if they sold through a traditional broker.鈥�
In going through the data, the team also found aberrations in who purchased homes from iBuyers. When iBuyers sold homes in Mecklenburg, institutions 鈥� frequently real estate investment trusts 鈥� bought 25% of the homes. Yet when an individual (not an iBuyer) sold the home, institutions bought just 15%.
The team also found racial differences in this shift. When iBuyers bought and resold homes, both originally white-owned and Black-owned homes were bought up at greater rates by institutions. But the increase in institutional ownership for white-owned homes (from 9% for individuals to 17% for iBuyers) was greater than the increase for Black-owned homes (from 33% to 36%).
Conversion to institutionally owned real estate is associated with negative outcomes, including and .
鈥淭hese real estate investment trusts tend to look for cheap homes that they can buy and convert to rentals so that they can profit over decades,鈥� Weber said. 鈥淪o this change in conversion rate from people to institutions is troubling because in the U.S. one of the substantial ways that people gain wealth and transfer it between generations is through homeownership.鈥�
The researchers plan to take the method and apply it to other areas 鈥斅爏uch as Maricopa County, Arizona, and Orange County, Florida 鈥� with large amounts of iBuyers, available data on home sales and race, and demographic diversity. They also plan to interview people who鈥檝e sold homes to iBuyers to learn what the experience is like.
鈥渋Buyers are offering a service. They鈥檙e making the home sale process faster and simpler,鈥� Slaughter said. 鈥淲hile our analysis in Mecklenburg suggests iBuyers are extending some disadvantages that Black home sellers tend to face to white home sellers as well, we don鈥檛 know that people are experiencing these sales as generally harmful or whether they鈥檙e aware of the tradeoffs that are involved.鈥�
, a doctoral student in the iSchool, is also a co-author on this paper. This research was partially funded by New America鈥檚 program on Public Interest Technology.
For more information, contact Slaughter at is28@uw.edu and Weber at nmweber@uw.edu.
