In January 2020, weeks into her tenure as a non-executive director of St James’s Place, Dame Helena Morrissey told a national newspaper she could not find a single person who could explain the firm’s charges in one sentence. She did not think the fees were clear enough, and she said so in public.
A sitting director questioning her own firm’s charges is unusual, and she was early as well as candid. Four years later, in February 2024, St James’s Place set aside £426m to refund clients who had paid for ongoing advice they never received. The charging model a director had named as impenetrable in 2020 had become a £426m provision.
The uncomfortable part is not that no one saw the problem. Someone did, and said so on the record, and the firm’s course changed only when the problem arrived as a number. That gap, between a credible challenge and any consequence, is the subject of this piece.
Two recent cases from UK financial services show it from opposite sides. In one, the challenge was raised and went unheeded until it became a loss. In the other, it was never made at all. Both point to the same conclusion. A challenge function that depends on the goodwill of the people it exists to challenge is not a control.
The warning raised in public
Morrissey did the job an independent director is there to do, and did it in the open. Weeks into the role she told the Telegraph she did not think St James’s Place’s charges were clear enough, and that she had not met anyone who could explain how they worked in a single sentence. That was not a stray remark. It named, in 2020, the exact area that would later cost the firm £426m.
She left the board a little over a year later, amicably, to chair AJ Bell, and there is no suggestion she was pushed out or silenced. That is what makes the case instructive rather than scandalous. The challenge was raised, clearly and publicly, by a director with the standing to be taken seriously, and the firm’s charging model still ran on until 2024, when the ongoing-advice provision made it unavoidable.
This is the quieter failure mode, and the more common one. A concern is raised. It is noted. The business is performing, the consensus is comfortable, and nothing in the machinery obliges the board to act on an early warning before it hardens into a number. Being listened to and being acted on are different things, and only the second one changes an outcome.
The challenge that was never made
Metro Bank shows the other half of the pattern. In January 2019 the bank disclosed that around £900m of commercial property loans had been risk-weighted at 50% when the requirement was 100%. The misclassification had overstated the bank’s capital strength. When it surfaced, the shares fell almost 40% in a single day. The Prudential Regulation Authority fined the bank for the weak reporting controls behind the error, and the Financial Conduct Authority fined it for publishing a risk-weighting figure it already knew was wrong.
Most accounts file this as a technical slip. That reading is too kind. A risk weighting is precisely the number that hides in plain sight. It looks settled, technical, somebody else’s responsibility. Everyone at the table assumes the classification was checked by whoever owns the model, and the model owner assumes the policy was approved above them. The assumption travels around the room unbroken. Catching it takes one person with the standing to ask who last tested that a given exposure was a 50 and not a 100.
At Metro, no one did, and the reason was structural. The bank was built around a dominant founder, and the board assembled around him deferred by habit. When the chair is also the personality the whole business is organised around, open disagreement carries a social cost that quiet agreement never does. So the question that would have caught the error went unasked. The market recognised the problem before the boardroom admitted it. Proxy advisers turned against the founder’s re-election, and he had left within the year.
Consensus is not agreement
The two cases look different and share one root. In neither did the board have a mechanism that turned a challenge into a consequence. At Metro the question was never forced onto the table; at St James’s Place it was raised and left to sit. In both, whether the board engaged with a risk came down to mood, personality and timing, and nothing in the structure required otherwise.
This is the distinction boards most often miss. Consensus dependency is not agreement. Agreement is what a board reaches after the awkward question has been asked and answered on its merits. Consensus dependency is what a board settles for when no one is required to ask. From the inside the two feel identical, which is why boards mistake one for the other, right up until a £900m line item or a client compensation provision proves the difference.
Why better directors and stronger culture do not fix it
The instinctive remedies are more independence and a healthier culture. Both matter, and neither is enough on its own.
Morrissey was already the high-calibre independent director those remedies are designed to produce, and it was not enough. Metro already had a board, a chair and committees; what it lacked was any challenge independent enough to use them. Adding more directors of the same kind, or running another programme on speaking up, leaves the underlying incentive untouched. As long as challenge can be overruled or ignored without consequence, there will never be enough of it at the moment it is needed, because that is the moment it is least welcome.
Challenge has to be independent by construction
The fix is to stop treating challenge as a behaviour and start treating it as a structure. Strategic Governance as a Service is built on that principle. It is a continuous, adversarial governance function retained at board level, whose whole mandate is to surface the failure mode before it becomes a loss.
Three features separate it from the arrangements that failed at both firms. It is a standing mandate rather than an optional act, so a challenge has to be answered on its merits and on the record. It carries enforceable independence, through a contractual notice period and a board-minuted challenge log, so it cannot be quietly switched off and any attempt to sideline it leaves a record. And it is held by a principal whose position does not depend on the decisions under review, unlike an internal function compromised by the reporting line it sits in. This is what the Red Team Protocol™ delivers as a board-level process, and it is what adversarial by design means in practice.
The argument rests on having sat on both sides of it. Owen Vallis chaired risk committees at Credit Suisse, overseeing fiduciary risk across more than £50bn in assets, and built the risk function from scratch as Group CRO at SICO Bank Group, a $6bn balance sheet operating across three regulatory jurisdictions. He has been the risk voice whose objection was heard with courtesy and then set aside, and he has designed the structures that stop that happening.
Every board believes it would have caught these failures. St James’s Place and Metro Bank believed the same, and both had people who could have. The question is not whether your board has challengers. It is whether your challenge function holds any power that does not depend on the goodwill of the people it exists to challenge.
A Diagnostic SGaaS engagement maps where challenge in your governance architecture is contingent rather than structural, and what to do about it. That is the conversation worth having while the next £900m line item is still hypothetical.
From the SGaaS White Paper
How challenge becomes a control
The white paper develops the structural argument in full, showing how the Red Team Protocol™ turns board challenge from an act of individual courage into a continuous, board-minuted mandate carrying the enforceable independence the arrangements at St James's Place and Metro Bank both lacked.
Owen Vallis is the founder of Marentis Labs, the firm that originated Strategic Governance as a Service. He spent ten years as UK Head of Fiduciary Risk Management at Credit Suisse and holds active board roles in the charity and education sectors. Schedule a confidential discussion.