The Federal Reserve Board has concluded adjustments designed to increase the “transparency and public accountability” of its assessment framework and curb “volatility” in associated capital mandates.
These supervisory evaluations are conducted to verify that major lenders maintain sufficient capital reserves to continue extending credit to commercial and consumer borrowers during deep economic downturns.
The regulator had previously signalled intentions to modify the framework in December 2024 to strengthen its resilience.
The pair of confirmed measures closely aligns with drafts published in 2025.
Under the first rule, the central bank will solicit yearly public feedback regarding “hypothetical scenarios” and substantial methodology revisions.
This measure updates the scenario design framework, establishes the quantitative models for the 2027 cycle, and alters the administrative timetable.
It also revises the assessment applied to institutions with significant trading operations.
These lenders will now be evaluated against two distinct “global market shock components” each year, with the scenario producing the steepest loss dictating the firm’s final result.
The second rule instructs the regulator to average performance data across the two latest annual cycles when establishing stress capital buffers for institutions evaluated in consecutive years.
This averaging process will commence in 2028 to confirm that only methodologies subject to public review are factored into the calculation.
Alongside these measures, the regulator is seeking feedback on a separate draft to alter its noninterest income methodology, aiming to better reflect varied commercial models.
If implemented, the revision will supersede the current model utilised to project fee earnings under distress. The consultation window will run for 60 days following publication in the Federal Register.
Collectively, these revisions are estimated to cut year-on-year fluctuations in capital mandates by roughly 50%, whilst leaving overall sector capital levels broadly unchanged.
