Coutts was fined £8.75 million because the wealthy files went stale
The FCA’s penalty turns on source-of-wealth and enhanced checks for politically exposed and other high-risk clients that sat open for years, not on a single missed wire.
A backlog became a finding
On 3 November 2025 the Financial Conduct Authority fined Coutts £8.75 million. The number is what the evening papers carried. The case underneath it is duller and more damaging. A multi-year look at private-bank files found source-of-wealth work and enhanced due diligence that had not kept up with politically exposed people and other high-risk clients. Periodic review, the control every private bank claims to run, had become a queue. Examiners read queues as governance when the board pack never shows how old the queue is or who owns it. Faircroft’s interest is not the brand. Coutts is a useful exhibit because the clients are exactly the ones a correspondent assumes are “already looked at.” Wealth, politics, and a famous name are not a substitute for a date on the file. If the last real refresh was years ago, the relationship is a story the bank is telling itself.
What “source of wealth” means when it fails
Source of wealth is not a job title and it is not a PDF of a passport. It is a coherent account of where the money came from, checked against something other than the client’s own sentence, and revisited when the client’s life changes. Enhanced due diligence is the extra work you do when the person is politically exposed or otherwise high risk: more documents, more media, a senior sign-off, and a reason for keeping the account that a stranger could read. The FCA’s criticism, in the public telling of this case, is that those steps did not happen on the timetable the bank’s own standards implied. That is a different failure from a rogue payment. No one needs a cinematic transfer. They need a file that still says what it said at onboarding while the client has changed countries, companies, or political status. Private banks are good at the first meeting. They are bad at the seventh year. This fine is about the seventh year.
The three questions a correspondent should add
Any bank that clears for a private bank, or that relies on one, should put three questions on the next form and refuse to accept “see our policy” as the answer. How often are politically exposed files actually refreshed, in months, not in adjectives. When is adverse media run again, and in which languages. Who owns an escalation if the papers do not come back, and what is the date on which the account moves toward exit rather than another polite chase. Those questions are ordinary. They are also the ones relationship managers dislike, because the clients who will not re-paper are often the quiet ones who never complain and never trade. Quiet is not the same as clean. A dormant high-net-worth profile is where a review date goes to die. If your dashboard cannot show aged reviews with an executive name next to the oldest bucket, you are running the control Coutts was fined for running badly.
Adverse media is a rerun, not a souvenir
A news search done at onboarding and stapled in the file is a souvenir. Families with assets in more than one country do not confine their public record to English, and they do not confine it to the month they opened the account. A 2026 refresh that still cites a 2019 Google printout is not a refresh. The operational test is simple. Take ten high-risk names. Show the last search, the language, the person who read it, and what changed in the risk rating afterwards. If nothing ever changes, either the clients are saints or nobody is reading. Politically exposed status is not permanent in a useful way and it is not frozen either. A person becomes exposed in the middle of a relationship. The work required at that moment is the same urgency as a new onboarding, not a note to “pick it up at annual review.” Annual review is how a mid-year change becomes a two-year gap. The Coutts exhibit is a warning against that gap, not a licence to gossip about any one client.
What the fine does not say
It does not say every Coutts client is suspect. It does not say private banking is a prohibited business. It does not publish the underlying files, and Faircroft will not invent them. It does not tell you to exit every politically exposed person in your own book. It says a well-known private bank left high-risk maintenance undone for long enough that the conduct regulator priced the failure at £8.75 million. The price is a footnote. The duration is the holding. Correspondents sometimes treat a fine as a brand event and tighten nothing in their own wealth channel. That is the wrong lesson. The right lesson is to benchmark your own alert ageing and your own review ageing against the failures named in public. If your oldest PEP review is older than the period the FCA was willing to tolerate at Coutts, you do not have a talking point. You have a finding waiting for a visitor.
Proof, not a policy release
Counterparties will ask, in 2026, whether the gaps were closed or merely described. The honest pack is a before-and-after sample: a redacted file as it stood, the same file after the chase, the date, and the name of the person who accepted it. A sentence in the annual report about “enhancing our framework” does not survive that request. Neither does a slide that shows review completion as a single percentage with no ageing. Relationship managers will argue for one more courtesy letter. Courtesy is fine as a first step. It is not a control. Pair it with a hard date. If the documents are not in by that date, the account leaves or the limits collapse. Write the date down before the meeting, not after the client has charmed the room. Private-bank culture is built to avoid that sentence. Enforcement cases are built from the meetings where nobody said it.
How Faircroft would file it
In a committee pack we would keep the FCA notice, the £8.75 million figure, and a one-page comparison with our own wealth-channel ageing. We would not speculate about unnamed clients. We would ask the business for quarterly completion rates on periodic review, not for the policy manual. And we would treat any correspondent that cannot produce those rates as a reliance problem, however old the brand on the letterhead. A famous lobby is not a file.