Your Bad Debt Reserve Now Depends on How Slow You Close

A CFO at a $16M specialty contractor asked me in June whether the new receivables rule would shrink his allowance. I told him it depends on when he issues his financials. He thought I was dodging the question.

One election, made once, in a footnote

ASU 2025-05 is effective for annual periods beginning after December 15, 2025, and for the interim periods inside them. If you're on a calendar year, you're in year one right now, and it already touched your Q1 and Q2 numbers whether or not your accountant mentioned it.

Two pieces to it. Everyone gets a practical expedient: assume conditions at the balance sheet date hold for the remaining life of the receivable. No macroeconomic forecast, no reasonable-and-supportable narrative about where unemployment goes next. Then there's a second option available only to private companies that took the expedient. You can count cash collected after the balance sheet date, up through the day the statements are issued.

Read that again. Cash that arrives in February reduces the reserve you book for December.

Same receivables, different number

Two companies, identical AR ledgers on December 31. Company A closes in twelve days and issues by March 15. Company B's audit finishes in late June. B gets to count three and a half more months of collections against the same opening balance, so B books the smaller allowance. Not because B underwrites better. Because B is slower.

You also have to say so. Electing the policy comes with a disclosure of the date through which you evaluated subsequent collections. That date sits in the footnotes next to the reserve, and your lender reads footnotes. You're publishing your close calendar as an input to an accounting estimate.

An expedient that rewards a slow close is a strange thing to hand a finance team. Take the relief, but don't let the reserve become the reason nobody fixes the calendar.

Current conditions assume a portfolio you might not have

"Conditions remain unchanged" is a statement about a pool. It holds up when AR is 300 accounts and last year's 0.7% loss rate describes something real. Put four customers at 45% of the balance and it falls apart, because concentrated credit loss isn't a rate. It's an event. Events don't appear in a historical loss rate until after they've happened, and by then you're writing off, not reserving.

The expedient relieves you of forecasting the economy. It doesn't relieve you of looking at what's in front of you. A customer whose average days-to-pay went from 38 in January to 71 by June is a current condition as of your balance sheet date, and current conditions are precisely what you're still on the hook to reflect. That contractor had one: a general contractor on a stalled municipal project, $340K outstanding, paying on the schedule of a job that isn't moving. No collections through issuance date will help him, and no portfolio rate will size it.

Do this before December

Pick the policy now rather than during the audit. Three decisions: whether you take the expedient, whether you take the private-company collections election on top of it, and what date you'll evaluate collections through. Transition is prospective, so there's no restatement to hide behind if you change your mind later.

Then pull your top ten receivable balances and look at them one at a time, outside whatever rate you apply to everything else. That's the work the standard didn't simplify, and it's the only part that was ever going to matter.