The CFPB has proposed sweeping revisions to its small-business lending data collection rule under Section 1071 of the Dodd-Frank Act, reversing much of its earlier approach and signaling a major shift for community lenders, banks and credit unions.
In a new proposal scheduled to be published Nov. 13, the CFPB said it plans to streamline the rule’s scope, reduce the number of covered lenders and data points, and delay mandatory reporting until Jan. 1, 2028. The Bureau said the changes would “reduce complexity for lenders and improve data quality” while maintaining the law’s core goals of tracking fair-lending patterns among small businesses.
The proposal would raise the threshold for covered financial institutions from 100 to 1,000 annual small-business loans over two consecutive years, effectively exempting hundreds of small credit unions from data collection requirements. It would also narrow the rule to focus on traditional business loans, lines of credit and credit cards, excluding merchant cash advances, agricultural lending and loans of $1,000 or less. The definition of “small business” would tighten from firms earning $5 million or less to those with $1 million or less in annual revenue.
The CFPB also proposes eliminating several discretionary data fields such as: Pricing information, denial reasons and the number of employees. The Bureau argued that a narrower dataset will be more reliable and less costly to produce.
For credit unions, the proposed rollback could mean fewer compliance costs and reporting obligations, especially for smaller institutions. However, the retreat could also reduce public visibility into how well credit unions serve minority- and women-owned businesses, an issue at the heart of Section 1071’s intent.
Public comments on the proposal are due 30 days after publication in the Federal Register.
© Arc, All Rights Reserved. Request academic re-use from www.copyright.com. All other uses, submit a request to TMSalesOperations@arc-network.com. For more information visit Asset & Logo Licensing.