Distributor credit is a loan book, so price it like one

Regional consumer brands in China compete with national ones on several fronts, and one of the least discussed is credit. The largest dairy and beverage companies can ask distributors to pay before goods leave the warehouse. A regional brand often cannot. It offers thirty, forty-five or sixty days instead, and distributors in smaller cities, many of them family businesses without easy access to bank finance, choose it partly for that reason.

I do not object to the credit. I object to the way most of us manage it. Distributor credit is recorded as a trade receivable, granted by sales managers, reviewed by finance at the month end and provided for by ageing. It is treated as an accounting balance. In substance it is a loan book: money lent, unsecured, to small businesses whose fortunes depend on their local markets and their other brands. A bank with a portfolio like that would have a credit policy, a committee, a price for time and an estimate of loss built from borrower quality. Most consumer companies have none of those things.

Credit that outlives the product

When I became CFO of Cuiling in April 2021, we were owed RMB 820 million by distributors, or 71 days of sales. RMB 214 million was more than ninety days overdue. Regional sales heads granted terms of forty-five to sixty days, and they granted them on chilled milk that lasts seven to fifteen days as readily as on ambient milk that lasts six months.

That mismatch is the heart of the problem. If a distributor sells fresh milk within a week and pays us seven weeks later, then for six weeks it holds cash that belongs to us. Some distributors used that float to carry slower-moving brands; a few used it to fund businesses that had nothing to do with food. When those businesses failed, our receivable failed with them. In 2020 two distributors in Jiangxi collapsed, and we wrote off RMB 58 million. Neither had been late in paying us until the month before it stopped.

The accounting did not warn us. Our provision was based on ageing, so a distributor who paid on day fifty-nine every month looked as safe as one who paid on day ten. The expected-credit-loss approach in the financial-instruments standard asks for something better: an estimate based on reasonable and supportable information about the borrower, not just how long a debt has been outstanding. We had that kind of evidence. It sat with the sales teams, who knew which owners had opened a property business and which had started carrying a rival's yoghurt.

Who decides, and on what evidence

The first change was to move the decision. Credit limits and terms are now set by a monthly committee that I chair, with the vice-president of sales and the head of internal control as members. Regional heads propose; the committee decides. That sounds bureaucratic, and in the first year it was. The sales organisation saw it as finance taking control of a commercial relationship, and they were not wrong to feel that.

We also changed what the sales teams were measured on. Regional heads had been paid on shipments, so a distributor who took more stock on longer credit helped their bonus twice. From 2022 part of their incentive depended on cash collected from their region and on its overdue balances, and within two quarters the discussion of credit limits became noticeably more realistic.

The second change was the evidence. Every distributor now has a score built from four things: payment history over two years, sell-out data from the shops it serves, the results of unannounced stock checks, and what we know of the owner's other businesses. The score sets a limit and a term, and it feeds our loss estimate directly. Distributors in the lowest band trade on prepayment; those in the highest receive up to thirty days on ambient products. Nobody receives more than fifteen days on chilled.

The third change was to stop shipping to any account more than sixty days overdue, without exception and without an appeal to the president. That rule caused more argument than the rest combined. It also did more good, because it made the conversation with a struggling distributor happen in the second month, not the sixth.

Charging for time

The change I would urge on other companies most strongly is the one we made last: putting a price on credit. A distributor who pays in advance now receives a 1.5 per cent settlement discount, and the price list distinguishes between prepaid and credit terms. Once credit has a price, distributors begin to compare it with a bank loan, and the stronger ones choose to prepay. The share of our distributor sales paid in advance rose from 21 per cent to 46 per cent. The price has to be visible to work. When the discount was buried in annual rebate terms, almost nobody changed behaviour; once the invoice showed the two prices side by side, the choice became a calculation each owner could make in a minute.

For smaller distributors who cannot prepay, we arranged supply-chain finance through a partner bank in 2023: the bank pays us on delivery and lends to the distributor against our invoice, at a rate set by its own credit view. I should have arranged it before we shortened terms. Some distributors left in 2021 and 2022 who might have stayed with that option.

The results were real, and so was the cost. By December 2023 distributor receivables were RMB 270 million, or 27 days of sales, and overdue balances beyond ninety days were RMB 29 million. About 240 distributors left or were consolidated, and sales in Jiangxi fell 9 per cent in 2022. Our vice-president of sales had predicted both. The board accepted them because the alternative was to keep lending to businesses we did not understand.

Directors can test this without a banking background. Ask how long the company's products last and how long distributors take to pay. Ask who sets the limits, and whether the loss provision changes when a distributor's behaviour changes, or only when its debt grows old. If the answers are sales managers and ageing, the company is running a loan book without knowing it.