AVMs should "reduce costs and broaden access to homeownership" - MBA

MBA encourages Fed agencies to prioritize reducing appraisal bias during the rulemaking process

AVMs should "reduce costs and broaden access to homeownership" - MBA

Alleviating appraiser shortages and combatting appraisal bias should remain top of mind during the rulemaking process for alternative valuation methods (AVMs), the Mortgage Bankers Association said Tuesday.

In response to a proposed rule issued by the Federal Housing Finance Agency and five other agencies, MBA wrote a letter cautioning the agencies of the “potential unintended consequences” of AVMs.

The proposed rule would implement quality control standards for AVMs used by mortgage originators and secondary market issuers to address racial bias in the real estate valuation process. 

While greater industry adoption of AVMs holds the potential to modernize and improve the valuation process, the MBA and CBA (Consumer Bankers Association) said the agencies should be mindful of the practicalities of model risk management and potential consequences, such as the impact on the availability of GSE waivers and alternative valuation methods.

Heightened compliance costs and homeownership affordability should also be considered, according to MBA.

“Mortgage lenders evaluate borrower eligibility based on an assessment of both credit risk and collateral risk – the extent to which value of the real property supports the mortgage loan,” the letter wrote. “Understanding the market value of the collateral is essential to pricing, originating, and servicing the loan. AVMs are used by the mortgage industry not only to assist in making credit decisions but to assess servicing portfolios and screen for outlier appraisals.

Fannie Mae and Freddie Mac (the GSEs) are two of the most influential AVM stakeholders, as they use their access to the FHFA appraisal dataset to inform quality control programs such as Collateral Underwriter and Loan Collateral Advisor.”

Read more: Appraisal technology and its battle with bias

Additionally, the lack of established methodologies in examining systemic bias should also be taken into account.

“While increased reliance on the models’ underlying datasets could flag “one-off” undervaluations, federal policymakers have expressed concern that the model algorithms themselves could replicate patterns of discrimination embedded in historical datasets,” the trade associations said.

MBA and CBA request that the CFPB and other regulators allow for an implementation period of at least 12 months.

Want to keep up with the latest mortgage news? Get exclusive interviews, breaking news, and industry events in your inbox, and always be the first to know by subscribing to our FREE daily newsletter.