A Lawsuit Aimed at $42 Billion of Bank Capital Relief That Hasn't Arrived Yet — While JPMorgan and Citigroup Already Spent the Money

Better Markets, the Washington nonprofit that has litigated and lobbied around post-2008 bank rules for a decade, filed suit on September 10, 2026 in the U.S. District Court for the District of Columbia against the Board of Governors and Vice Chair for Supervision Michelle Bowman. The group asks the court to throw out the rulemaking and force a restart under a different official, alleging Bowman privately coached Wall Street CEOs on how to engage the public comment process for the 2026 capital proposals.

Strip out the rhetoric and the contested package is worth about 4.8% of Category I and II CET1 requirements, or roughly $42.1 billion. Most of that comes from one piece: the GSIB surcharge recalibration, which cuts surcharge dollars by about 10%, or $23 billion, and pulls the average applicable surcharge from 2.7% toward 2.3%. The Basel III component actually raises requirements by about 1.4% on its own.

Scale matters here. That $42.1 billion is equivalent to about 46% of a single year of buybacks at the four largest banks. Even a total plaintiff victory would not reverse capital returns; it would defer a one-time step-down in requirements and, more damagingly, restart the clock on a process that already consumed three years.

This is the part the headline narrative gets backwards. The capital-return cycle running right now was financed by earnings and by regulatory certainty the lawsuit does not reach: the Board's February 2026 decision to hold stress capital buffers at current levels into 2027, confirmed in JPMorgan's own stress-test disclosure as running through September 30, 2027.

The leverage-ratio rule that did become final tells the same story from the other direction. The agencies stated that it lowers holding-company tier 1 requirements by less than 2% in aggregate, and warned explicitly that capital freed at the depository subsidiary generally would not be available for distribution to external shareholders. Leverage relief did not pay for these buybacks; profits and a frozen buffer did.

The 2026 stress test was the mildest in years: 32 banks absorbed nearly $708 billion of hypothetical losses with the aggregate CET1 ratio falling only from 12.8% to 11.2%. That 1.6-point decline is why the sector can fund payouts without the proposals — but the dispersion underneath separates banks with two points of slack from those with one.

Two readings jump out. Bank of America enters with the lowest starting ratio in the group at 11.4%, so a deferred surcharge cut bites hardest where the buffer is already thinnest. And BNY loses just 0.1 point under severe stress — proof that for custody banks the binding constraint is the systemic-risk surcharge and leverage math, exactly the pieces now in litigation.

Bank capital rules have a supply chain. Upstream sit the balance sheets that carry market plumbing; downstream sit the borrowers and the non-bank lenders who price against bank capital costs. A delayed print keeps the cost of intermediation elevated on both ends.

Reuters notes that litigation against the Fed over rulemaking is rare, and the recent precedent cuts against speed: the bank trade groups' December 2024 stress-test suit spent 2025 in repeated stays rather than rulings. A plaintiff challenging a proposal faces the threshold problem that there is no final agency action to vacate — which is why the practical outcome is more likely a slower, more defensible record than a dead rule.

Here is the inference investors should hold loosely: even the harshest outcome leaves the frozen stress buffers and the finalized leverage rule intact, because neither is named. The lawsuit can postpone the next leg of relief but cannot claw back the current one — a distinction that separates a headline risk from a capital-return risk.