The Set of Three

Start with what the bank did not say, because it is the part that decides this. It did not say a line was wrong. It did not dispute the stocktake, the rebate policy or the capitalisation criteria, and it did not need to: it read the three judgements as a set and did one subtraction. Each judgement was tested by the audit on its own merits, and on its own merits each passed. The gap is not in any of the three. It is between them.

That is why A, tempting as it is, answers a question nobody asked. Of course the opinion stands; of course the accounts comply; of course a lender reads for the bad year and an auditor reads for fair presentation. But the audit committee chair did not ask whether the accounts comply. She asked whether the audit missed something, and the honest answer is that it tested three judgements and never the picture they make together. The standards have a name for judgements that are each reasonable and all lean the same way: an indicator of possible management bias. The audit looked for it inside each estimate and not across them.

B overcorrects into the wrong instrument. A restatement is for accounts that are wrong, and nothing here has been shown to be wrong; withdrawing a signed opinion because a lender disagrees with the dividend would misstate what an audit is far more than the three notes misstate the year. C borrows a decision that belongs to the board. The auditor reports on the accounts, not on whether the company should pay out 60% or 75% of them, and an audit partner who starts advising on distributions has stepped out of the only chair that gives the signature its weight.

The senior read is D, and it has two halves that must both be said. The first: each judgement complied, with evidence, and the opinion is not withdrawn — say it plainly, because the committee will otherwise hear hedging as doubt. The second: the three were never tested as a set, and that is a finding about the audit, not about the client. What changes is the next engagement, not the last one. Materiality judged on the aggregate of the year's accounting judgements, not estimate by estimate. Cash conversion in the analytical review alongside profit, so that a 48-point fall sits on the same page as a 9% rise. The three policies revisited at planning with the question the bank asked: what do they do together. And one sentence for the chair to carry to the board — the opinion says the accounts present fairly; it does not say the dividend is wise, and nobody should read it that way.

The bank added up what the audit had checked one at a time. The signature is right. The next plan is what answers the phone call.