As part of the Biden Administration’s stated focus on narrowing the racial gap in wealth and homeownership, federal agencies launched an Interagency Task Force on Property Appraisal and Valuation Equity (PAVE), with the goal of “addressing the persistent misvaluation and undervaluation of properties experienced by families and communities of color.” More recently, on February 14, 2022, the U.S. Department of Justice (DOJ) issued a statement that “combatting housing discrimination, including bias in appraisals, is a high priority across the federal government.” And on February 22, 2022, Representative Maxine Waters (D-CA), chairwoman of the House Financial Services Committee, announced that she would be “introducing legislation to address systemic appraisal discrimination” and asked that the U.S. Department of Housing and Urban Development (HUD) “initiate a systemic fair housing investigation into housing valuation discrimination.”
While these federal efforts are commendable, the concept of appraisal bias is far from novel. Indeed, as early as the 1930s, property valuation – a service performed by the federal government pursuant to its Home Owners’ Loan Corporation (HOLC) program – deliberately and effectively “institutionalized in a rational and bureaucratic framework a racially discriminatory practice that all but eliminated Black access to the suburbs and to government mortgage money.”
Under the HOLC program, “government agents methodically included in their procedures the evaluation of the racial composition or potential racial composition of the community,” deeming undesirable and relegating to the lowest category those communities that “were changing racially or were already black.” Even as recently as 2016, HUD explicitly required its rostered appraisers to evaluate the “desirability” of a neighborhood by giving “special consideration” to the “positive and negative effect on value of gentrification.” Thus, while many describe the appraisal process as an “art, not a science,” racial and ethnic minorities may feel they are playing a rigged game.
Playing the Real Estate Market
The real estate industry has been described much like betting, with a rising housing market commonly referred to as “hot” (as opposed to the “bust” associated with low housing prices and decreased demand) and an unsupported housing market likened to a “house of cards.” Indeed, appreciating housing values resulting from subprime lending and inflated appraisals began to slow in 2006 and came crashing down shortly thereafter.
By the time the Board of Governors of the Federal Reserve System proposed amendments to Regulation Z (then codified at 12 CFR Part 203) to protect borrowers of subprime (i.e., higher-priced) mortgage loans, it had become apparent that “[a]verage loss amounts today are rising because the damage caused by collateral value inflation at origination is being compounded by severely depreciating real estate market values.” The Federal Reserve Board reasoned that “[a]n inflated appraisal also may lead consumers to believe that they have more home equity than in fact they do, and to borrow or make other financial decisions based on this incorrect information.” For example, “a consumer also may take out more cash with a refinance or home equity loan than he or she would have had an appraisal not been inflated.” Such a consumer, if saddled with “substantial unsecured consumer debt,” may be “especially vulnerable” because “they may find a cash-out refinance offer attractive.”
Ultimately, leading up to the 2008 financial crisis, “borrowers in black neighborhoods [were] five times as likely to refinance in the subprime market than borrowers in white neighborhoods,” even when controlling for income. These statistics indicate that racial and ethnic minorities were acutely affected by inflated appraisals and the resulting housing market crash. To that end, “[w]hile the foreclosure crisis has had vast consequences throughout the United States, it has had a disproportionate impact on persons of color,” even when considering that homeownership rates for Black / African American and Hispanic/Latino borrowers had been at their lowest levels in over a decade and would drop even further.
Notably, neighborhoods with populations of color of less than 10% in 2000 saw an increase in foreclosures of 215%, while neighborhoods with 90% or greater minority populations experienced an increase of 544%. Overall, the bursting of the housing market bubble took a far greater toll on the wealth of minorities than whites.
Tightening the Appraisal Rules
In 2008 and in response to the increase in foreclosure rates, which threatened significant credit loss to government investors, Freddie Mac, Fannie Mae, and the Federal Housing Finance Agency (FHFA) entered into a settlement with the New York State Office of the Attorney General, resulting in a Home Valuation Code of Conduct (HVCC) to protect against inflated appraisals caused in part by lender influence. At around the same time, the Federal Reserve Board exercised its rulemaking authority under the Home Ownership and Equity Protection Act by amending Regulation Z to provide for several new protections for consumer mortgages, including appraiser independence requirements designed to ensure the integrity of real estate appraisals.
Subsequently, Congress enacted the Dodd–Frank Wall Street Reform and Consumer Protection Act, which sunset the HVCC and codified the Regulation Z appraiser independence provisions via a new Section 129E of the Truth in Lending Act (implemented by the valuation independence requirements in current 12 CFR § 1026.42).
From here, the government-sponsored entities (GSEs) followed suit, issuing their Appraiser Independence Requirements, which imposed heightened obligations on lenders to maintain separate and independent functions for their loan production and appraisal-related services. Further, various federal government agencies, including the Office of the Comptroller of the Currency, Federal Reserve System, and Federal Deposit Insurance Corporation, updated their existing Interagency Appraisal and Evaluation Guidelines to “reflect developments concerning appraisals and evaluations, as well as changes in appraisal standards and advancements in regulated institutions’ collateral valuation methods,” while reemphasizing that appraisals must “support an institution’s decision to engage in the credit transaction.” And, effective January 1, 2010, HUD imposed heightened requirements on lenders that order appraisals in connection with loans insured by the Federal Housing Administration. In addition, and to further protect against overvaluation bias, the GSEs and various private vendors developed automated valuation models (AVMs) for lenders to use in evaluating borrower collateral.
Another Rising Housing Market
The various enhancements to the appraisal process served to fine-tune the industry’s risk assessment and management abilities overall and succeeded in reducing expected mortgage default rates, particularly for refinance transactions. Indeed, after controlling for factors such as loan-to-value ratio, credit score, debt-to-income ratio, interest rate, and loan balance, it appears that cash-out refinance transactions during 2010–2017 were “only” 55% riskier than purchase transactions, as opposed to 96% riskier before the tightening of appraisal standards. While this tightening of standards has resulted in almost half of cash-out refinance loan applicants receiving lower-than-desired appraisal values, the refined process suggests a safer bet for consumers.
However, over the last 18 months, housing prices once again have been on the rise due to record-low mortgage rates and flexible work-from-home policies that allowed buyers to expand their home search. As for those homeowners staying put, they extracted more cash from their homes through cash-out refinances in 2020 than in any year since the financial crisis, assured by more stringent underwriting and appraisal standards than were available before the financial crisis. As the rate of cash-out refinances continues to rise, will increased housing values remain sustainable? Industry predictions suggest that housing prices will decelerate each month of 2022 and steadily temper the rapid home price acceleration seen in 2021. However, the extent of any deceleration or potential decline in the housing market remains to be seen.
Precisely because of such market volatility, the existing appraisal framework is designed not only to protect borrowers from taking on more debt than their property is worth but also to limit the investor’s risk of loss in the event of borrower default. Fannie Mae, for example, emphasizes that, to be eligible for delivery, the loan must be “secured by a property that provides acceptable collateral.” To that end, Fannie Mae cautions lenders to “pay particular attention and institute extra due diligence for those loans in which the appraised value is believed to be excessive or when the value of the property has experienced significant appreciation in a short time period since the prior sale.”
Further, when using automated valuation tools, the federal agencies expect lenders to “establish a level of acceptable core accuracy and limit exposure to a model’s systemic tendency to over value properties.” Fannie Mae’s own valuation tool issues “flags and messages that signal heightened risk of overvaluation.” Similarly, other government investors require a lender to ensure that an appraisal value adequately supports the loan to be guaranteed or insured.
But government investors acknowledge that there is no one “right” property value. For example, the Department of Veterans Affairs (VA) requires the appraisal value to be “reasonable,” Fannie Mae and Freddie Mac refer to it as an “opinion” of value, and HUD acknowledges it is an “estimated” value. Similarly, the federal banking agencies note that an appraisal is “an opinion as to the market value” of a property on a particular date. Most importantly, the Uniform Standards of Professional Appraisal Practice (USPAP) definition of an appraisal, which all appraisers must comply with, recognizes that the “opinion of value” may be numerically expressed “as a specific amount, as a range of numbers, or as a relationship” to a prior value or benchmark.
These nebulous characterizations are not surprising, given that the appraisal process relies on the professional judgment and expertise of the appraiser and permits the appraiser to employ subjectivity. For example, HUD acknowledges that factors of location, site, view, design, quality of construction, age, condition, and functional utility “are all subjective factors that require subjective adjustments.”
Prohibition on Appraisal Discrimination
Yet, while subjectivity is expected, consideration of a prohibited factor such as race or ethnicity is prohibited and constitutes unlawful discrimination. The Fair Housing Act prohibits discrimination in “making available [a residential real estate-related] transaction, or in the terms or conditions of such a transaction,” by a person or entity “whose business includes engaging in residential real estate-related transactions.” For purposes of the Act, a “residential real estate-related transaction” includes “the making or purchasing of loans or providing other financial assistance – (1) for purchasing, constructing, improving, repairing, or maintaining a dwelling; or (2) secured by residential real estate” (i.e., a mortgage loan transaction), as well as “the selling, brokering, or appraising of residential real property.” Fair Housing Act claims against lenders are often brought to HUD by community organizations that receive annual grants from HUD under the agency’s Private Enforcement Initiative and commit to finding and referring discrimination complaints for investigation and conciliation.
In addition to antidiscrimination laws, the USPAP explicitly prohibits an appraiser from relying on “unsupported conclusions relating to characteristics such as race, color, religion, national origin, sex, sexual orientation, gender, marital status, familial status, age, receipt of public assistance income, disability, or an unsupported conclusion that homogeneity of such characteristics is necessary to maximize value.” Further, for every appraisal report, the appraiser must certify that the report contains their “personal, impartial, and unbiased professional analysis” and that the appraiser “had no bias with respect to the property or the parties involved” in the appraisal assignment.
Unfortunately, there is no bright-line rule for determining whether an appraisal is discriminatory. It is the position of numerous advocacy groups and public policy organizations that the “widespread presence of anti-black bias” and “even outright discrimination and racism” has directly contributed to the devaluation of black assets in the housing market, suggesting that the discriminatory conduct should be apparent on the face of the appraisal report.
Yet, absent explicit references to a homeowner’s race or ethnicity, or statements regarding the demographic composition of the neighborhood, how can a lender know or verify that an appraiser chose valuation factors based on professional judgment and expertise rather than a prohibited characteristic such as race or ethnicity? Of course, an appraisal value that is lower than desired by the borrower or needed to support the loan is not itself evidence of discrimination.
Not only are lenders prohibited from increasing an appraisal value or ordering another appraisal absent a determination that the appraisal is deficient, but they also must employ their own licensed or certified appraisers if they perform a substantive review of the appraisal report. And neither the law nor government agency guidelines require a lender to perform a substantive review of the appraisal before relying on it to make a loan. Rather, the expectation is that the lender will confirm that the appraisal was performed by state-certified or licensed appraiser, contains sufficient information and analysis to support the institution’s decision to engage in the transaction, and includes a certification that the appraiser complied with the USPAP. Ultimately, through the appraisal process, the lender “is responsible for ensuring that the subject property provides adequate collateral for the mortgage.”
Potential Improvements to the Appraisal Process
Given that the existing appraisal framework is designed to prevent lender influence and protect against overvaluation of collateral, lenders are in a difficult position to detect discrimination and correct it. Suggestions made by federal agencies, government investors, advocacy groups, and public policy organizations vary widely, conflict with one another, and ultimately reflect each entity’s own stake in the game. For example, a study commissioned by the Federal Financial Institutions Examination Council recommends that the USPAP be amended “to require appraisers to identify mortgage borrowers as ‘intended users’ of appraisals prepared in relation to residential mortgage transactions.”
While this suggestion would create a duty of care where none currently exists, and therefore a private right of action for negligence, for borrowers who believe that an appraiser has undervalued their home, this change would conflict with existing government investor guidelines, which emphasize that the “intended user” of the appraisal is the lender, and the “intended use” is to “evaluate the property that is the subject of this appraisal for a mortgage finance transaction.” Further, it is unclear what the intended use of the appraisal report would be for a mortgage borrower who may not own the property or ultimately purchase it.
Virtually all other suggestions for detecting and eliminating appraisal discrimination or bias are directed at the appraiser and recognize, at least implicitly, that lenders’ obligations in the appraisal process cannot easily be changed. These suggestions include “expand[ing] the participation of people of color in the appraiser profession” (given that the vast majority of appraisers are white and only 1% are black) and prohibiting appraisers from using in their appraisal report commentary any words or phrases that can be viewed as potential race- or ethnicity-related red flags.
However, even these appraiser-specific recommendations are at odds, reflecting the differing, and sometimes incompatible, priorities among industry participants. While government investors propose “modernizing” the appraisal process through “better use of data, technology, and process design,” the Consumer Financial Protection Bureau, much like advocacy groups, has expressed “concerns that AVMs may reflect bias in design and function or through the use of biased data and may introduce potential fair lending risk.”
Nevertheless, other than recommending that lenders review the language of appraisal reports for red-flag language, the industry has come up empty-handed in terms of what lenders can do to detect appraisal discrimination and correct it. While government investors have made some updates to their guidelines to require lenders to “evaluate the appraisal and ensure it complies with … the Fair Housing Act and other federal, state, or local antidiscrimination laws,” the nature and scope of this obligation remains a topic of discussion and some disagreement.
Even the nature and extent of appraisal bias – and the steps in the appraisal process that might be the problem – are subject to contention. On the one hand, Freddie Mac asserts that “the topic of a potential appraisal gap is worthy of considerable research,” while Fannie Mae, much like public policy think tanks, notes that “the frequency of ‘undervaluation’ [does] not have a notable racial pattern.”
But even assuming significant bias, there is no consensus on which legitimate alternatives could substitute for the existing valuation process. This is no surprise, given the industry’s use of credit scoring models for their predictive value, despite concerns that these models may have a disproportionate and unfavorable effect on minorities.
Indeed, the lender’s entire appraisal-related process, from the use of automated valuation tools to the scope of appraisal review, could very well be examined for potential discrimination, even when the lender is performing these actions consistently in accordance with its policies and procedures and without regard to race or other prohibited factor. This level of scrutiny into a lender’s involvement with a mortgage settlement service is unprecedented. If this were a different settlement service, such as title, inspection, or hazard insurance, there would be no expectation for a lender to scour the results for deficiencies, or implicit biases, that could signal discrimination. For example, courts acknowledge that a lender would not be expected to uncover a defect in title that a knowledgeable title company or attorney failed to detect.
Ultimately, lenders will need to understand the expectations and risks arising from their role in the appraisal process, particularly once the historic run on housing prices begins to wind down. Further, while no one has yet come up with a solution for detecting and addressing appraisal bias, the industry might benefit from working together to eliminate potential bias by considering conscientious enhancements to the appraisal process rather than burning down the house.