Home / Daily Dose / Automated Valuation Models: Benefits and Unexpected Challenges
Print This Post Print This Post

Automated Valuation Models: Benefits and Unexpected Challenges

As professionals from all industries search for ways to work more efficiently and safely during a national health crisis, property appraisers have turned to automated valuation models (AVMs), which allow for contact-free assessment of a home’s market value.

Researchers at the Urban Institute acknowledged that AVMs hold promise when it comes to reducing costs and increasing accuracy, and they write that the practice has helped keep the housing market moving during the COVID-19 pandemic. Still, in a late-December study, they found that AVMs in majority-Black neighborhoods produce larger errors, relative to the underlying sales price, than AVMs in majority-White neighborhoods, potentially contributing to the wide housing wealth gap between Black and White homeowners.

Researchers Michael Neal, Sarah Strochak, Linna Zhu, and Caitlin Young, who wrote the paper entitled "How Automated Valuation Models Can Disproportionately Affect Majority-Black Neighborhoods," say that historically, appraisals, including in-person appraisals, have been tainted by bias.

"AVMs have drawbacks," the authors wrote. "Historically, AVMs have not been able to take a property’s condition into account when determining a home’s value. Like in-person appraisals, the accuracy of AVMs depends on having a large-enough number of comparable sales in the area to ensure greater accuracy. And the use of comparable sales in the area may reinforce past racial bias."

The report, which can be accessed in full on urban.com, examines key drivers of the magnitude of inaccuracy and how they differ among neighborhoods.

For example, the authors found that the percentage magnitude of inaccuracy in majority-Black and majority-White neighborhoods may be similar, the lower sales prices in majority-Black neighborhoods increase the percentage magnitude of inaccuracy significantly in all three cities we examined, possibly causing greater damage to the overall home values in these neighborhoods.

The Institute's findings suggest that the expanded use of AVMs could disproportionately affect majority-Black neighborhoods and reinforce the impacts of past racial discrimination that often resulted in the undervaluation of Black-owned homes. According to the researchers, a significantly lower appraised value, even from an AVM estimate, could lead to a canceled sales contract, which can contribute to the Black-White homeownership rate gap and the Black-White wealth
gap.

"When aggregated across society, these risks also have implications for policymakers responsible for macroeconomic and financial market supervision. More research is needed to identify the full scope of policy implications stemming from AVM inaccuracy," the authors noted. "Nevertheless, the policy suggestions we present are meant to ensure that as the use of AVMs increases, its costs are better understood and more progress is made to ensure that all households experience the benefits of homeownership.

The entire report is available on urban.org.

About Author: Christina Hughes Babb

Christina Hughes Babb is a reporter for DS News and MReport. A graduate of Southern Methodist University, she has been a reporter, editor, and publisher in the Dallas area for more than 15 years. During her 10 years at Advocate Media and Dallas Magazine, she published thousands of articles covering local politics, real estate, development, crime, the arts, entertainment, and human interest, among other topics. She has won two national Mayborn School of Journalism Ten Spurs awards for nonfiction, and has penned pieces for Texas Monthly, Salon.com, Dallas Observer, Edible, and the Dallas Morning News, among others. Contact Christina at [email protected].
x

Check Also

Federal Reserve Holds Rates Steady Moving Into the New Year

The Federal Reserve’s Federal Open Market Committee again chose that no action is better than changing rates as the economy begins to stabilize.