Veros Economists Find No Evidence Of Racial Bias In Its AVM

Veros Real Estate Solutions (VEROS), released research that finds no evidence of racial bias in its VeroVALUE Automated Valuation Model (AVM). This validation provides an important quality control check for lenders seeking tools to identify potential racial bias in home valuations.

VeroVALUE is an AVM derived from high-quality data, industry leading models, and predictive technologies, all resulting in first-rate accuracy. An in-depth analysis of the VeroVALUE AVM was conducted across five housing markets, including Chicago, Atlanta, Houston, Los Angeles and Philadelphia, which were selected because of their diverse populations. By studying a wide range of cities, Veros reinforced the conclusion from its first report that its AVM did not display racial bias.

The study utilized neighborhood racial compositions to determine if a correlation existed between the percentage of Black, Hispanic, Asian and White populations and the number of properties in the area that were undervalued or overvalued by more than 15%.

The findings were consistent in all five cities, which revealed no evidence of racial basis in the VeroVALUE AVM. The study showed that the proportion of properties undervalued by more than 15% had no correlation with the proportion of Black or Hispanic populations in a neighborhood.

Through an extensive analysis of the VeroVALUE AVM, the findings suggest that bias tested AVMs can be leveraged to provide a quality control check to identify potential undervaluation related to bias. A properly tested professional-grade AVM, such as VeroVALUE, can be an innovative solution in the housing industry to appropriately flag any potential instances of bias and ensure fair home valuations.

“It is crucial to utilize professional-grade AVMs that can help identify potential racial bias in home valuations and to flag concerns of potential undervaluation of minority-owned properties,” said Eric Fox, Chief Economist at Veros. “Proven solutions like VeroVALUE give the housing market a precise, accurate and non-biased solution that is built on quality data.”