Kowhai Retail Group

A seasonal facility funds a loop: cash becomes stock in September, stock becomes cash by Christmas, and the line returns to zero by the end of February. That last clause is the whole covenant — the clean-down is what makes seasonal lending different from term lending in disguise. So before reading a single number in this year's request, ask the only question the structure cares about: did last year's loop close? It did not. NZ$1.1m of last season's buy is still on the shelves, the line needed the overdraft's help at peak, and the margin line paid a five-hundred-basis-point clearance tax on the way through. Which means part of the NZ$1.5m "seasonal uplift" being requested is not seasonal at all. It is the unsold past, asking to be refinanced under the season's name.

Now the buy itself. Up 22% on a year that didn't sell, with an explanation — wrong mix, not wrong demand — that is plausible, specific, and completely unverifiable from the file. The MD may be right. Merchants are sometimes right about exactly this. But notice what kind of claim it is: it is a conviction about consumer behaviour four months from now, and the file cannot distinguish a merchant who has learned from a merchant doubling down. Operating cash flow down 60% says the business enters this bet with less cushion than last year, not more. The bank is being asked to fund an increased position on the strength of an explanation for why the last position lost.

The reader's answer is B — full approval with clearance milestones and weekly sell-through reporting — and it is worth being precise about why it fails. Weekly reporting through peak does not prevent a bad Christmas; it documents one in high resolution. If the mix is wrong again, the milestones will be missed in November, when the stock is already landed, paid for, and unreturnable — the bank will hold a beautifully evidenced file about a loss it funded anyway. Conditions that cannot change the outcome are not risk management. They are the bank purchasing the feeling of rigour while becoming the buyer's risk partner with none of the buyer's upside.

The senior banker's read is C. Fund the season — fully, promptly, shaped like the season the business has actually demonstrated: NZ$2.6m, last year's true peak, with the clean-down covenant intact and the aged-stock proceeds reducing the core as they clear. The increment above that is not working capital; it is a merchandising conviction, and convictions are funded by the people who hold them — through shareholder funds, or through evidence. That second path is the constructive one to offer: if September's confirmed pre-orders from the trade customers convert part of the +22% from belief into demand, the file changes, and the facility can change with it. The MD will call this conservative. It is — precisely as conservative as his own balance sheet is being, and that is the conversation: a bank that funds the cycle is a lender; a bank that funds the bet has become a partner in a gifting retailer, on a lender's margin.