Washington says it is depoliticising banking. The first exhibit is a sharply worded letter about a gun-finance case. If this is a revolution, it is starting with stationery.
The White House has declared “reputational risk” persona non grata and promised “fair banking for all Americans.”
Executive Order 14331 tells independent regulators to strip out fuzzy notions which could enable “politicised or unlawful debanking,” put clocks on internal reviews, and sketches a government-wide strategy to prevent viewpoint-based account closures.
It reads like a manifesto for neutral, risk-based supervision. It is also, for now, a manifesto. The order sets direction and deadlines but it does not rewrite the Bank Secrecy Act or the sanctions list.
If reputational risk sounds like a Beltway abstraction, consider this: the Office of the Comptroller of the Currency already told examiners in March to stop treating it as a standalone exam topic and began deleting references from its handbook.
The order simply tries to make that posture government-wide. In other words, reputation has not left the building – it has left the paperwork. Banks must still manage public blowback when it translates into liquidity or counterparty risk. They are just less likely to be second-guessed for it in an exam manual.
Enter Exhibit A of the new era: the Consumer Financial Protection Bureau ’s closure of a four-year investigation into Credova – A PublicSquare Company, a buy-now-pay-later firm that finances firearms and outdoors purchases.
In an August 19 letter to Credova’s parent, PSQ Holdings, the CFPB’s chief legal officer calls the probe “weaponization,” alleges improper influence from the New York Attorney General, and explicitly invokes the executive order in shuttering the case and lifting document-retention obligations.
The letter even notes CFPB staff once pressed Credova to stop leasing firearms as a settlement term. Stirring stuff, as letters go. But it is still a letter: an agency characterisation, not a court ruling.
PublicSquare duly celebrated; markets cheered. As the Financial Times reported, shares jumped on the news and the company positioned the closure as vindication for “Second Amendment” commerce. The politics write themselves.
Yet a single closure, however splashy, does not establish a systemic diagnosis, let alone its cure. One letter does not a doctrine make.
After years of pressure campaigns, the US government has leaned on banks and processors to cut off lawful but disfavoured customers. There is history here. “Operation Choke Point” in the mid-2010s placed certain industries under a reputational cloud, leading to well-documented fights with Congress, agency mea culpas, and an Federal Deposit Insurance Corporation (FDIC) statement urging risk-based, customer-by-customer decisions rather than categorical bans. The new order is framed as a repudiation of that era. Fair enough.
But repudiation is not regulation, and rhetoric is not evidence.
Everything is not as it seems
When you look for prevalence, the picture is murkier. Reuters combed through the CFPB’s complaint database and found that out of 8,361 detailed complaints about bank account closures since 2012, 35 explicitly cited political or religious bias.
Supporters of the order argue under-reporting is rife because banks rarely state reasons. Critics say the bigger access problem in US banking is still income, not ideology. Both can be true. The point is that Washington is building a sweeping “fair banking” policy on a thin evidentiary base while tasking agencies to find retroactive villains within 120 days.
Legally, the new posture borrows some glow from the Supreme Court’s unanimous decision in NRA v. Vullo, which revived a First Amendment claim alleging a New York regulator crossed the line from persuasion into coercion when urging insurers and banks to consider the “reputational risk” of working with the NRA.
The case is about government pressure on intermediaries as a form of viewpoint discrimination. It is a live warning shot, proving coercion and tracing it to an actual rights violation remains a high bar. The Court told lower courts to apply that standard; it did not pronounce that every “don’t you think about the optics?” nudge is unconstitutional.
In the meantime, the compliance knot for banks tightens. Off boarding controversial but lawful customers will now demand even more pristine documentation showing concrete, measurable risk. This will likely include fraud typologies, sanctions exposure, chargeback performance, financial crime red flags, counterparty stability.
‘Because our reputational risk committee felt queasy’ was poor practice before; it will be exhibit A for plaintiffs now. Yet safety and soundness obligations have not been repealed. AML, OFAC, network rules, sponsor-bank covenants and enterprise-risk appetites endure.
The likely near-term outcome is not mass onboarding of controversy, but heavier memos, narrower rationales, and more litigation about where ‘politics’ ends and ‘risk’ begins.
There is also a transatlantic echo worth noting. The UK’s Financial Conduct Authority, after a political row over high-profile account closures, said in 2023 there was “no evidence” of systematic closures for political beliefs.
Different market, different facts, but the same caution; intuition about debanking often outpaces the record. The US order, by contrast, acts first and asks agencies to quantify later. That sequence will satisfy partisans and frustrate practitioners.
So what would good look like?
Start by defining “political” with the precision regulators demand elsewhere. A workable standard should distinguish:
- customers whose activities trigger specific, bank-independent hazards (sanctions hits, fraud rates well above peer, BSA look-backs), fro
- customers whose activities are lawful but controversial, where the hazard is indirect and reputational.
Then require any offboarding in the latter bucket to clear procedural gates: internal challenge by an independent risk function, a documented causal link to quantifiable loss channels, and a customer appeals path that does not vanish into a call-centre.
Second, publish the rewrites. The order gives the Small Business Administration 60 days to notify lenders, regulators 120 days to review past policies, and 180 days to purge “reputation risk” from guidance and deliver a Treasury strategy.
When those manuals change, do it in daylight. Banks need to know what examiners will consider out of bounds, and customers need to know when their appeal is more than a suggestion box.
Finally, keep score. If politicised debanking is as rampant as the rhetoric suggests, complaint data, supervisory findings and court dockets should start to show it. If not, then the order may end up as a constitutional victory statement with limited operational bite.
