Bowman, The Final Chapter on Modernizing Bank Regulatory Stress Testing

Bowman's Mansion House speech is a supervisory document, not a monetary policy signal: the Fed's Vice Chair for Supervision lays out the near-final package to overhaul bank stress testing. The two headline rules (model and scenario transparency, and averaging the SCB over two years) are already public and broadly priced; the new material is a third 2027 model proposal, a preview of supervisor-only forward-looking stress tests, and an explicit end-of-year timeline for risk-based capital and G-SIB surcharge finalization. Direction of travel is looser and more predictable bank capital. Nothing here moves the front end.

What’s new: The two final rules were telegraphed by the Board's December 2024 commitment and the subsequent proposals, so the market knew the shape: disclosure plus a two-year SCB average. Genuinely new: a third proposal to replace the 2027 noninterest income model; the plan for forward-looking supervisor-only stress tests; and the hard signal that risk-based capital and G-SIB surcharge reforms will be finalized before year-end. The SCB change is explicitly capital-neutral, so there is no aggregate capital delta to price.

KEY FINDINGS

• The Board will consider final revisions to the stress testing framework "in the coming weeks," and Bowman expects risk-based capital reforms plus G-SIB surcharge changes to be finalized before the end of this year. A concrete near-term deregulatory calendar is the main catalyst for bank capital and bank equity positioning.
• "Together, these rules reduce SCB volatility by half without materially changing aggregate levels of required capital." Capital-neutral reduction in volatility lowers planning uncertainty but should not change aggregate bank capital demand or funding.
• A third, newly flagged proposal would revise the noninterest income model for the 2027 stress test to better capture business diversity. Model-level change, modest read-through, but it shifts how fee and trading revenue is treated in loss projections.
• Forward-looking stress exercises would be supervisory only: "the results of these forward-looking tests would not affect capital requirements," and would not be made public. No capital consequence, so limited market impact, but it institutionalizes supervisory scenario analysis.
• Bowman floats two comment-driven ideas she will consider in the final rule: freezing firm balance sheets before scenario release, and running two global market shock scenarios on the same as-of date using the larger loss for the SCB. Both would further damp SCB volatility and take a downward tilt off loss capture.

FROM THE DOCUMENT

In the coming weeks, the Federal Reserve Board will consider final revisions to the Board's stress testing framework.

In addition, and before the end of this year, I expect the Board to finalize reforms to risk-based capital requirements for large and small banks as well as improvements to the global systemically important bank (GISB) surcharge.

Together, these rules reduce SCB volatility by half without materially changing aggregate levels of required capital.

Importantly, the results of these forward-looking tests would not affect capital requirements.