In debate over appraisal bias, rival researchers clash over key data

By: ameer@trustedteam.com

homes-031221-topten.jpeg

Key findings about discrimination in home valuations are under dispute. That hasn’t stopped them from informing policy decisions.

During the past two years, regulators and lawmakers have introduced and adopted new rules and guidelines aimed at curbing the impacts of racial bias on home valuations. But some appraisers and researchers insist these efforts have been based on faulty data.

Conflicting findings from a pair of non-profit research groups call into question whether or not recent actions will improve financial outcomes for minority homeowners without leading to banks and other mortgage lenders taking on undue risks.

The debate centers on a 2018 report from the Brookings Institution, which found that homes in majority-Black neighborhoods are routinely discounted relative to equivalent properties in areas with little or no Black population, a trend that has exacerbated the country’s racial wealth gap. The study, which adjusts for various home and neighborhood characteristics, found that homes in Black neighborhoods were valued 23% less than homes in other areas.

“We believe anti-Black bias is the reason this undervaluation happens,” the report concludes, “and we hope to better understand the precise beliefs and behaviors that drive this process in future research.”

The study, titled “Devaluation of assets in Black neighborhoods,” has been cited by subsequent reports published by Fannie Mae and Freddie Mac, academics and White House’s Property Appraisal and Valuation Equity, or PAVE, task force, which used the data to inform its March 2022 action plan to address racial bias in home appraisal.

Meanwhile, as the Brookings’ findings proliferated, another set of research — based on the same models and data — has largely gone untouched by policymakers. In 2021, the American Enterprise Institute replicated the Brookings study but applied additional proxies for the socioeconomic status of borrowers. 

By simply adding a control for the Equifax credit risk score for borrowers, the AEI research asserts, the average property devaluation for properties in Black neighborhoods falls to 0.3%. The researchers also examined valuation differences between low socioeconomic borrowers and high socioeconomic borrowers in areas that were effectively all white and found that the level of devaluation was equal to and, in some cases, greater than that observed between Black-majority and Black-minority neighborhoods.

“That, to us, really suggests that it cannot be race but it has to be due to other factors — socioeconomic status, in particular — that is driving these differences in home valuation,” said Tobias Peter, one of the two researchers at the AEI Housing Center who critiqued the Brookings study. 

Contrasting conclusions

Peter and his co-author, Edward Pinto, who leads the AEI Housing Center, acknowledge that there could be bad actors in the appraisal space who, either intentionally or through negligence, improperly undervalue homes in Black neighborhoods. But, they argue, the issue is not systemic and therefore does not call for the time of sweeping changes that the PAVE task force has requested. 

Brookings researchers have refuted the AEI findings, arguing that, among other things, their controls sufficiently rule out socioeconomic differences between borrowers as the cause of valuation differences. They also attribute the different outcomes in the AEI tests to the omission of the very richest and very poorest neighborhoods. 

Jonathan Rothwell, one of the three Brookings researchers along with Andre Perry and David Harshbarger, said the conclusion reached by AEI’s researchers ignored the well documented history of racial bias in housing. 

“No matter how nuanced and compelling the research is, no one can publish anything about racial bias in housing markets, without our friends Peter and Pinto insisting there is no racial bias in housing markets,” Rothwell said. “Everyone agrees that there used to be racial bias in housing markets. I don’t know when it expired.”

Mark A. Willis, a senior policy fellow at New York University’s Furman Center for Real Estate and Urban Policy, said the source of the two sets of findings might have contributed to the response each has seen. While both organizations are non-partisan, AEI, which leans more conservative, is seen as having a defined agenda, while the centrist Brookings enjoys a more neutral reputation.

Still, Willis — who is familiar with both studies but has not tested their findings — said while the Brookings report notes legitimate disparities between communities, the AEI findings demonstrate that such differences cannot solely be attributed to racial discrimination.

“The real issue here is there are differences across neighborhoods in the value of buildings that visibly look alike, maybe even technically the neighborhood characteristics look alike, but aren’t valued the same way in the market,” Willis said. “Whatever that variable is, Brookings hasn’t necessarily found that there’s bias in addition to all of the other real differences between neighborhoods.”

Setting the course or getting off track?

The two sets of findings have become endemic to the competing views of home appraisers that have emerged in recent years. On one side, those in favor of reforming the home buying process — including fair housing and racial justice advocates, along with emerging disruptors from the tech world — point to the Brookings report as a seminal moment in the current push to root out discriminatory practices on a broad scale.

“It’s been really helpful in driving the conversation forward, to help us better define what is bias and be specific about how we communicate about it, because there’s a number of different types of bias potentially in the housing process,” Kenon Chen, founder of the tech-focused appraisal management company Clear Capital, said. “That report really … did a good job of highlighting systemic concerns and how, as an industry, we can start to take a look at some of the things that are historical.”

Appraisers, meanwhile, say the Brookings findings made them a scapegoat for issues that extend beyond their remit and set them on course for enhanced regulatory scrutiny.  

“What’s causing the racial wealth gap is not 80,000 rogue appraisers who are a bunch of racists and are going out and undervaluing homes based on the race of the homeowner or the buyer, but rather it’s a deeply rooted socioeconomic issue and it has everything to do with buying power and and socioeconomic status,” Jeremy Bagott, a California-based appraiser, said. “It’s not a problem that appraisers are responsible for; we’re just providing the message about the reality in the market.”

Responses to the Brookings study and other related findings include supervisory guidelines around the handling of algorithmic appraisal tools, efforts to reduce barriers to entry into the appraisal profession and greater data transparency around home valuation across census tracts. 

But appraisers say other initiatives — including what some see as a lowering of the threshold for challenging an appraisal — will make it harder for them to perform their key duty of ensuring banks do not overextend themselves based on inflated asset prices.

Even those who favor reform within the profession have taken issue with the Brookings’ findings. Jonathan Miller, a New York-based appraiser who has deep concerns about the lack of diversity with the field — which is more than 90% white, mostly male and aging rapidly — said using the study as a basis for policy change put the government on the wrong track. 

“There’s something wrong in the appraisal profession, and it’s that minorities are not even close to being fairly represented, but the Brookings study doesn’t connect to the appraisal industry at all,” Miller said. “Yet, that is the linchpin that began this movement. … I’m in favor of more diversity, but the Brookings’ findings are extremely misleading.”

Willis, who previously led JPMorgan Chase’s community development program, said appraisers are justified in their concerns over new policies, noting this is not the first time the profession has shouldered a heavy blame for systemic failures. The government rolled out new reforms for appraisers following both the savings and loan crisis of the 1980s and the subprime lending crisis of 2007 and 2008. 

But, ultimately, Willis added, appraisers have left themselves open to such attacks by allowing bad — either malicious or incompetent — actors to enter their field and failing to diversify their ranks. 

“It seems clear that the burden is on the industry to ensure that everybody is up to the same quality level,” he said. “Unless the industry polices itself better and is more diverse, it is going to remain very vulnerable to criticism.”

Related post