Two structures finance the same export, and they are often thought to differ only in banking mechanics. They differ in the identity of the insured, and that governs everything else. Under supplier credit, the exporter itself grants the payment terms: the receivable arises on its own balance sheet, it carries it, and it is the one that insures. Under buyer credit, a bank lends to the foreign buyer or to the buyer's bank, pays the exporter as shipments proceed, and the receivable that remains is a bank loan. The insured is no longer the exporter, it is the lender. One and the same default therefore does not produce one loss; it produces two, or none, depending on the structure, and the first question to ask on an unpaid item is not what happened but who was carrying the receivable.
The immediate consequence is that the exporter under buyer credit is out of the payment risk, and it knows this. What it knows less well is what it is not out of. The bank's loan is almost always tied to the commercial contract: drawdowns are conditioned on performance certificates the exporter signs, and the credit agreement opens a recourse against it if the buyer resists repayment on grounds of the supplier's non-performance. The exporter has therefore swapped a payment risk for a performance risk, attached to a receivable it no longer controls. That is not a smaller risk, it is a different one, and it wakes up precisely when the bank files its claim, because that is when the cause of default gets argued.
Nor is the loss quantified the same way, and this point fixes dates that cannot be recovered later. The bank's loss is a loan: it bears on outstanding principal and interest, it follows an amortization schedule, and it crystallizes on acceleration, a dated act that makes the whole balance due. The exporter's loss under supplier credit is an invoice: it bears on an amount due at a commercial maturity date, and it builds up across successive unpaid installments. The first structure produces a single large loss on a date the lender chooses; the second produces a run of small losses on dates the invoicing calendar imposes. Waiting periods, notification duties and limitation periods all run from these different starting points.
Proof, next, is where the two structures part most harshly, and it is the difficulty nobody anticipates. What is covered here is a cause of default, not a default. Yet the bank lending to a foreign buyer has no visibility into what happens there: it observes an unpaid installment and nothing more. The material establishing the political cause, an administrative circular, a refused transfer authorization, correspondence from the supervising ministry, circulates within the commercial relationship, that is, at the exporter, which is not the insured and is under no contractual duty to hand it over. A file can therefore fail not because the cause was missing but because the party holding it was not the party that had to prove it.
On top of this sits a conflict of interest on the same default, and it is structural rather than accidental. The bank wants to accelerate, because that is what crystallizes its loss and starts its waiting period. The exporter wants instead to preserve a commercial relationship on which its order book and its spare parts depend, and a rescheduling suits it better than early maturity. Worse, acceleration opens the argument about the causes of default, and a buyer in difficulty has every interest in invoking supplier non-performance rather than an administrative block, which shifts the burden onto the exporter. The two parties to the same contract therefore want opposite outcomes, with different insurers and policies that do not speak to each other.
Finally, one must distrust the binary presentation, because real contracts are mixed. A forty million contract commonly breaks down into a down payment settled directly, a tranche financed by buyer credit, and a retention released after acceptance, which remains pure supplier credit. The same exporter is then insured on one part of the price, paid out on another, and unknowingly exposed on a third, the retention being the portion that survives longest and is lost most often. The reading rule is short and better than the nomenclature: insurance follows the money, not the sale contract. Look at who disbursed, what remains owing, and to whom.
What follows from all this is not a preference for one structure. Buyer credit relieves the exporter's cash position and takes the receivable off its balance sheet; supplier credit leaves it in control of the file and of the evidence. What follows is that the structure is chosen knowing which risk one keeps, and that documentary coordination between exporter and lender is organized at signing, not at the loss. An agreement providing from the outset that the exporter will hand the lender the material establishing the cause of a default costs one clause and saves a file; the same request made eighteen months later, while both parties argue over attribution, can no longer be obtained.
A French equipment maker signs a 40 million euro contract. A 15 percent down payment is settled directly by the buyer. The remaining 85 percent is financed by a buyer credit granted by a European bank to the foreign buyer, drawn against shipping certificates; the maker is therefore paid cash on that tranche. The commercial contract also provides for a 4 million retention, releasable twelve months after acceptance. Eighteen months after commissioning, a circular from the buyer country's finance ministry prohibits currency transfers in respect of external financial debt. The buyer stops repaying the loan and does not release the retention. The bank is considering acceleration; the buyer writes that performance reservations remain outstanding. Who files what, and what must the equipment maker do?
The costliest mistake would be to reason in terms of one default, one claim. There are two receivables here, two insureds and two policies. On the 85 percent, the maker holds no receivable: it has been paid, and the loss belongs to the bank, which will crystallize it by acceleration and whose waiting period will run from that act. On the 4 million retention, by contrast, the maker holds an unpaid commercial receivable, under a separate policy, with its own maturities and its own waiting period: that loss is its own, and it alone can notify it. Two points then decide the file. The first is proof: the ministerial circular establishes the political cause and it is dated, but the maker holds it through its commercial relationship while the bank is the one that must produce it. It has every interest in organizing that transfer at once and in writing, since its own notice on the retention rests on the same document. The second is the performance challenge, which must be treated as a manoeuvre rather than a technical dispute: if the reservations prosper, the bank will exercise its contractual recourse against the maker, which will then have lost twice. It must therefore document acceptance without delay, and refuse to let acceleration be framed as a commercial disagreement.
- 01The structure decides who the insured is: under supplier credit the exporter, under buyer credit the lending bank.
- 02An exporter paid cash under buyer credit is not clear of everything: it keeps a performance risk attached to the credit agreement's recourse.
- 03A loan crystallizes on acceleration, an invoice accumulates maturity by maturity: waiting and limitation periods do not share a starting point.
- 04The party that must prove the political cause is not the party holding the proof, and a file can fail on that mismatch alone.
- 05Insurance follows the money, not the sale contract: in a mixed deal the retention is the longest-surviving portion and the one most often lost.