Skip to content
Hugin
A steel beam resting on supports beneath the frame of a heavy industrial test press in a dim, empty concrete hall.

Hugin News

Federal Reserve final rules open stress-test scenarios and material model changes to public input and average two years of results in capital buffers from 2029; 2027 model comments due December 1

3 min read

Original editorial artwork generated for Hugin.

The Federal Reserve Board published two final rules and a request for comment on October 2. The rules set a yearly calendar for putting stress-test scenarios and material model changes out for public input, and a bank tested in consecutive years gets a capital buffer built on two years of results, starting January 1, 2029. Comments on proposed 2027 model changes must be received by December 1, 2026.

federal-reservestress-testbank-capitalstress-capital-bufferfederal-registerfinal-rule
3source receipts1source hosts3 minread timelinkedprimary source

The Federal Reserve Board published two final rules, and a request for comment on 2027 model changes, for its bank stress test in the Federal Register on October 2, 2026. Comments on that request must be received by December 1, 2026.

What is published in advance

The stress test projects how large banks' capital would fare in a hypothetical recession. A final rule effective November 2, 2026 sets a yearly calendar:

  • Models: published by May 15 of the test year.
  • Material model changes: put out for public input by August 31 of the prior year, for at least 30 days, starting with the 2028 test (by this desk's count, August 31, 2027). A change is material if the Board estimates it could move any one firm's projected post-stress capital ratio by 20 basis points (0.2 percentage point) or more, or the average of the firms' absolute changes by 10 basis points or more.
  • Scenarios: proposed by January 10 of the test year, at least 30 days for input, final by February 28 (trading and counterparty parts, and any additional scenarios, by March 1).

What changes in the buffer

The stress capital buffer is extra capital the Board requires of the large firms it covers. It equals a firm's projected drop in its common equity tier 1 ratio (common equity capital to risk-weighted assets) plus four quarters of planned dividends, with a floor of 2.5 percent of risk-weighted assets.

A second final rule, effective December 1, 2026, averages the drops from the two most recent annual tests for a firm tested in consecutive years. Dividends are not averaged, and firms tested every other year are generally not averaged. Buffers will take effect January 1, not October 1.

Averaging starts with buffers effective January 1, 2029. The buffer effective January 1, 2028, is not averaged. For 2020 through 2025, the Board estimates, year-to-year changes in a firm's buffer would have averaged 47 basis points, not 64. Some commenters said averaging would lower capital levels. The Board says the aim is steadier requirements, not lower ones.

The 2027 comment request

The notice proposes a change to the 2027 noninterest income model, to better capture differences across firms, and revisions to the stress-test data reports. Comment under Docket No. OP-1882 on the Board's online proposals page (preferred), or by mail, courier or email. The Board says comments are generally posted without change.

What the documents do not say

  • How many banks are covered. None counts them. The nearest figure is a paperwork estimate of 35 respondents to the data reports, from holding companies with $100 billion or more in assets.
  • Whether the January 10 step applies to the 2027 test. Only the model-change step names 2028.
  • A comment deadline on the final rules. They give effective dates only.

Source links