Building resilience and durability into mortgage rules

Since 2021, the CFPB has been reviewing a number of its rules and guidance documents, including those published by predecessor agencies like the Federal Reserve Board of Governors and the Federal Trade Commission. We have found that many of these policies were drafted in ways that are unnecessarily complex, often to accommodate the preferences of dominant industry incumbents, rather than the market as a whole. We identified a particularly serious problem in federal mortgage rules that has major implications for the market.

In 2010, Congress enacted reforms to prevent mortgage lenders from setting up borrowers to fail in an unaffordable loan. The implementing rules establish a formula that help lenders understand their compliance obligations. It involves a variable called “average prime offer rate” or “APOR.”

APOR is not a market-based concept. It is calculated and published by the CFPB. To date, these calculations have been largely based on data controlled by a separate entity. Freddie Mac was the only entity that compiled the relevant mortgage pricing data through its Primary Mortgage Market Survey that the CFPB relied on to construct APORs. However, Freddie Mac recently announced that it was changing the Primary Mortgage Market Survey to no longer include points, fees, and adjustable rates data relied on by the CFPB to calculate APORs. The CFPB has had other difficulties relying on a single entity for calculating the APOR benchmark over the last decade. For example, in 2016, Freddie Mac discontinued publishing the one-year variable rate mortgage data. The CFPB found a different data provider, but shortly thereafter that provider also discontinued providing the necessary data.

Last month, the CFPB announced a new methodology and the selection of a new data provider for calculating APOR. The need for these revisions to the APOR methodology highlighted important broader weaknesses in our system around single points of failure and the reliance on overly complicated benchmarks, which we will explore resolving moving forward.

This pattern underscores the need for more durable rules that don’t over-rely on single entities. No consumer protection rule should be designed so that its important protections are threatened by single points of failure or single sources. The need for revisions to the APOR methodology also highlighted the risks of relying on complicated reference rates that must be manually constructed rather than potentially more robust market-based measures that stand on their own.

The CFPB is fortunate to have a mindset of self-critical analysis. The CFPB’s shift to a new data provider is a patch that will allow us to continue providing required benchmarks. Moving forward, we will continue to explore long-term solutions to move away from single points of failure, ensure system resilience, and eliminate unnecessary complexity. We continue to welcome feedback on how to rethink the approach to rulemaking.