Two federal bank regulators on Thursday adopted the first formal definition of an “unsafe or unsound” banking practice, narrowing one of Washington’s broadest supervisory tools to conduct that has caused—or is likely to cause—material financial harm. The Office of the Comptroller of the Currency and Federal Deposit Insurance Corporation said the standard will direct examiners toward threats to capital, asset quality, earnings and liquidity rather than weaknesses in policies, paperwork or controls that have no clear financial consequence.

The new rule also raises the threshold for issuing a “Matter Requiring Attention,” the confidential supervisory finding commonly known as an MRA. Those findings can force bank boards and executives to correct deficiencies before regulators escalate to a public enforcement action. Under the new framework, an MRA generally must concern an actual violation of banking law or a practice that could reasonably be expected to materially harm the institution or the Deposit Insurance Fund.

The change is a consequential recalibration of bank oversight, not a withdrawal of it. It gives banks a clearer basis for challenging examiner demands and preserves regulators’ authority over serious financial risks, but it may reduce supervisors’ ability to compel early action on governance, compliance or operational weaknesses whose financial effects are difficult to quantify. The rule takes effect 60 days after publication in the Federal Register and applies prospectively to institutions supervised by the OCC and FDIC.

A standard built around financial harm

For decades, federal law has allowed regulators to act against an “unsafe or unsound practice” without defining the term in statute. Courts and agencies relied on a standard developed during the 1966 expansion of federal cease-and-desist authority: conduct contrary to prudent operation whose continuation could produce abnormal risk, loss or damage. That flexibility supported measures ranging from corrective orders to civil penalties and, in extreme cases, termination of federal deposit insurance.

The 73-page final rule replaces that open-ended approach with a two-part test. Conduct must be contrary to generally accepted standards of prudent operation and must have materially harmed a bank’s financial condition, be likely to do so if continued, or pose a material risk of loss to the Deposit Insurance Fund. “Likely” is deliberately stronger than merely possible, and material financial condition is tied to capital, asset quality, earnings, liquidity and sensitivity to market risk.

Reputation risk that is unrelated to financial condition falls outside the definition. Policies, procedures and internal controls can still support an unsafe-practice finding, but only when examiners can connect their weakness to the required level of financial harm. The agencies said that connection should rest on objective facts and sound reasoning, which examiners must share with the bank.

What changes inside an examination

The rule creates a separate, somewhat more preventive standard for MRAs. Examiners may issue one when deficient conduct could reasonably be expected, under current or foreseeable conditions, to cause material financial harm or loss to the insurance fund, or when the conduct violates a banking or banking-related law. The agencies say that formulation lets supervisors address emerging problems before losses occur while preventing routine observations from becoming mandatory board-level remediation projects.

Issues below that threshold may be communicated as “supervisory observations.” A bank can choose whether to implement an examiner’s suggested enhancement, and the observation is not supposed to create a supervisory expectation or be treated like an MRA. Actual legal violations that do not warrant an MRA may still have to be corrected, but regulators generally cannot require additional steps unless another law authorizes them.

The practical distinction matters because MRAs consume management attention, technology budgets and board oversight. Banks have long complained that informal examination judgments can function like binding rules without public notice or consistent appeal rights. Reuters reported that the final action is part of a broader Trump administration effort to focus supervision on core financial risks rather than minor deficiencies.

Coverage will remain divided

The framework covers national banks, federal savings associations and federal branches overseen by the OCC, along with state nonmember banks, insured state branches and other institutions supervised by the FDIC. The agencies explicitly limited the rule to their own supervised institutions and removed affiliated individuals from its scope. Enforcement against executives, directors and other institution-affiliated parties will continue under earlier standards and controlling case law.

The Federal Reserve did not join the rule, leaving a potentially important divide in federal supervision. The Fed oversees bank holding companies, state member banks and many of the largest, most complex financial firms. As FT reported, institutions with different charters could therefore face different supervisory thresholds until the Fed acts, particularly where agencies share responsibility for related entities in the same banking group.

The case for greater clarity

Banking groups argue that the existing system diverts limited resources toward technical problems while obscuring the risks that can actually topple an institution. The American Bankers Association supported a material-harm threshold and urged regulators to require demonstrable evidence when examiners say a loss is likely, according to its comment letter. The Bank Policy Institute said the final standard should make supervision more objective and allow banks to innovate without treating every procedural weakness as a safety-and-soundness failure.

Regulators point to the 2023 failures of Silicon Valley Bank and other large institutions as evidence that examination resources should concentrate on balance-sheet vulnerabilities. The final rule says those failures followed material weaknesses and rapid withdrawals of uninsured deposits, reinforcing the need to prioritize liquidity, capital and interest-rate exposure. The agencies also require more intensive tailoring as a bank’s size, complexity and risk increase, lowering the practical materiality threshold for more consequential institutions.

The warning about early signals

Critics respond that financial harm is often the end of a chain that begins with weak governance, incomplete controls or ignored compliance problems. Federal Reserve Governor Michael Barr warned in a November speech that the proposal could tie examiners’ hands by making it harder to require correction before damage becomes measurable. He cited deficient controls involving money laundering or discrimination as examples that might not satisfy a financial-harm test until a violation or loss is established.

State regulators raised a related concern because they jointly supervise nearly 2,800 state-chartered institutions with the FDIC. The Conference of State Bank Supervisors said in its formal comments that material consumer and operational harm should remain relevant even when it is not immediately financial. It also warned that differing state, FDIC and Federal Reserve standards could produce inconsistent examination outcomes.

The agencies rejected the premise that a sharper threshold prevents preventive work. Examiners can still document observations, enforce actual law violations and issue MRAs when foreseeable conditions support a reasonable expectation of material harm. The unresolved question is behavioral: whether supervisors will use those channels assertively, or whether the evidentiary demands and appeal risk will make them reluctant to elevate concerns before losses become obvious.

Implementation will determine the impact

Bank boards should expect fewer findings labeled unsafe or unsound, but not necessarily fewer examination conversations. The rule shifts the burden toward explaining how a deficiency connects to financial condition or the insurance fund, making documentation of materiality central for both examiners and bank risk teams. Cybersecurity, contingency funding and third-party failures can still qualify when a severe operational disruption is likely to produce a material financial consequence.

The first examinations under the standard will show whether the change produces disciplined prioritization or a blind spot around developing risks. Agency training, written findings, supervisory appeals and coordination with state regulators will be more important than the definition alone. The rule gives banks clearer boundaries; its success will depend on whether those boundaries help supervisors act earlier on the right problems without overlooking the controls that reveal them.