During our cold file reviews, we come across a couple of scenarios that are more common than most practices would like to admit and one is worth sharing, because the reasoning behind the ‘wrong’ answer was genuinely good. It just was not answering the right question.

The scenario

A company had a material loan due to its parent, sitting on the balance sheet as a long-term liability. The loan agreement, however, contained a standard clause: the parent could demand repayment at any time.

The reviewer raised it as a review point. The balance, as drafted, would fail quality control.

The pushback from client and it was a good one

The team came back with a well-researched response, backed by tangible evidence:

• The loan had been outstanding for number of years with no demand ever made.
• Part of the funding had, over time, been recapitalised into equity supporting the idea that this was long-term capital, not short-term financing.
• The funds had financed a long-life infrastructure asset operating under a multi decade lease consistent with long term institutional investment rather than short term financing.
• The parent’s own audited accounts classified the corresponding receivable as non-current.
• The parent had even provided written confirmation that it had no intention of demanding repayment within the next twelve months.

Taken together, it is a compelling commercial story. The funding clearly behaves like long-term capital. So, was the review point wrong?

No. And here is why.

IAS 1 does not ask whether a lender is likely to call a loan. It asks a narrower, sharper question: does the borrower have an unconditional right, at the reporting date, to defer settlement for at least twelve months? If a contractual on-demand clause exists, that right does not exist full stop.

This is not a judgement call open to weighing probabilities. The 2020 amendments to IAS 1 (effective for periods beginning on or after 1 January 2023) closed this exact gap by adding paragraph 72A, which states plainly that classification is based on the rights in place at the reporting date, and is not affected by management’s expectations about whether those rights will be exercised.

Run each piece of evidence back through that lens, and none of it survives:

Evidence offered	Why it does not change the answer
No demand made historically	An expectation, not a right explicitly excluded by IAS 1.72A
Parent’s own non-current classification	Reflects the lender’s view of recoverability, not the borrower’s contractual right
Written confirmation of no near-term demand	A statement of intent does not amend the loan agreement itself
Partial recapitalisation into equity	Only changes the terms of the portion converted the remainder is still subject to the on-demand clause
Funding a long-life asset under a long lease	Speaks to how the funds were used, not to the legal right to defer settlement

The one thing that would have changed the outcome

Had the loan agreement itself been formally amended before the year end removing the on-demand clause the classification would follow the revised terms. Comfort letters, patterns of behaviour and strategic intent do not do that job. Only the contract does.

The takeaway

This is a useful reminder for anyone reviewing files (or preparing accounts) with shareholder or intercompany funding: substance-over-form arguments feel intuitive, and they are often right in the boardroom. But IAS 1’s current/non-current test is deliberately mechanical, precisely so that users of accounts see the balance sheet as it legally stands at the reporting date not as management believes it will play out.

Good evidence, wrong test. It happens more often than people expect, and it is exactly what a robust cold file review is there to catch.