Validating Regulator Positions Before You Commit Capital

Start with the assumption, not the regulator

Most validation exercises fail at the first step: teams go looking for information about regulators before they have written down what they actually believe. You end up gathering intelligence with no benchmark to test it against.

Before any external work, force the leadership team to state, in writing, the specific claims underpinning the entry case. Not "the regulator is supportive" but "the PRA will accept a branch structure with capital held at group level, the FCA will treat our product as outside the consumer duty remit, and neither will require a Part 4A variation before Q3." Each claim needs an author, a confidence level, and the evidence currently behind it.

This is uncomfortable. It exposes how much of the entry thesis rests on one conversation, one former colleague's view, or one advisor's read from six months ago. That exposure is the point.

Separate the three things people confuse

Regulator behaviour splits into three questions that require different validation methods:

• Formal position: what the rules, guidance and published decisions say
• Current supervisory posture: how the relevant team is actually applying discretion right now
• Individual intent: what specific decision-makers will do when your file lands on their desk

Desk research answers the first. Structured practitioner interviews answer the second. Only direct, attributable engagement, or high-quality proxies for it, answers the third. Teams routinely use evidence from one category to validate assumptions in another. A published speech does not tell you how a case officer will treat your authorisation. A former regulator's view does not tell you what the current head of division thinks.

Build a triangulation matrix, not a source list

For each critical assumption, identify at least three independent sources that would need to be wrong for the assumption to be wrong. Independence matters more than seniority. Three former regulators from the same era, all now advising firms, is one source wearing three hats.

Good sources to combine:

• Recent authorisation outcomes for structurally similar firms (public and, via counsel, private)
• Practitioners with live files in the same supervisory team in the last twelve months
• Trade body policy leads who have seen the regulator's response to consultation feedback
• Counsel who has recently tested the specific point in pre-application meetings
• Where appropriate, direct pre-application engagement with the regulator itself

If your matrix collapses to two overlapping sources, you do not have validation. You have consensus among people who talk to each other.

Run the pre-application meeting as a test, not a pitch

When you do engage the regulator directly, most firms squander it. They present the plan and ask if there are concerns. The regulator gives a polite, non-committal response, and the team reads silence as support.

Design the meeting to falsify your assumptions, not confirm them. Present the two or three points you believe will be most contentious and ask specifically how the supervisor has treated similar structures. Ask what would cause them to escalate. Ask what the closest recent refusal or delay looked like and why. If you leave the meeting without at least one piece of information that contradicts your prior view, you probably did not ask hard enough questions.

Watch for the false positives

Three signals are routinely misread as regulator support:

• Procedural engagement: they answered your questions, therefore they back the plan. They did not. They responded to correspondence.
• Absence of objection: nobody said no in the meeting. Regulators rarely say no in meetings. They say no in letters, months later.
• Sympathetic individuals: one person in the supervision team likes the idea. That person is not the decision. The head of division, the policy team, and often a separate authorisations function will each have a view.

Know when you have enough

You have validated an assumption when you can name the specific evidence that would change your mind, and that evidence has not appeared despite active search. If you cannot state what would falsify the assumption, you have not tested it. You have decorated it.

Next step

Before your next investment committee, take the three assumptions your entry case depends on most heavily. For each, write down the evidence that would prove it wrong, and who on your team is accountable for looking for that evidence in the next thirty days. If nobody owns the disconfirming search, the assumption is not validated, whatever the paper says.