An unpaid invoice from a state-owned energy company looks exactly like a commercial default. The same invoice, the same delay, the same silence, and often the same embarrassed contact on the phone. Two policies can respond, trade credit and political risk, and they do not respond in the same way. What decides is never the nature of the debtor, it is the cause of the default. A badly run state-owned company that does not pay is an ordinary credit risk. The same company, forbidden by its supervising ministry from paying a foreign supplier, is a political risk. The debtor is identical in both cases, and the cover that responds is not.
Commercial default has a definition that can be established. It takes two forms, insolvency established by court proceedings, and de facto failure, that is, a payment delay extending beyond a constitutive period set in the contract. In both cases the debtor cannot pay, and the credit insurer has been tracking that capacity continuously: it analyzes the solvency of the buyers in the insured's portfolio and grants guaranteed limits that it raises or lowers as it goes. The loss is proved by documents that exist, a judgment, a payment schedule, a reminder that went unanswered, and recovery follows the ordinary law of receivables.
Sovereign non-payment is the reverse: the debtor can pay and does not, or is not allowed to. Three distinct situations are regularly confused here, and confusing them is expensive. The first is the public entity whose supervising authority forbids it to settle, by internal instruction or circular, which is an administrative act rather than a financial difficulty. The second is the state that suspends payments generally, by moratorium, which strikes every creditor at once. The third is not a non-payment at all: the buyer has paid, in local currency, and the funds cannot leave the country. That one belongs to non-transfer, with its own cover, its own triggering event and its own waiting period.
This distinction does not decorate the file, it decides it. The two covers share neither trigger, nor constitutive period, nor evidence. Trade credit asks you to establish an inability to pay, which is a documented and provable fact. Political risk asks you to establish a cause, and the cause lives in a decision the state has no interest in publishing. A claim notified under the wrong policy is not simply redirected: it is late under the right one, because the clock ran while the file was being worked elsewhere.
Evidence is therefore the real craft of this module, and it is built while the facts are happening. The debtor will almost never say that it is forbidden to pay: writing it would tell all its suppliers that it will pay none of them, which would cost it its supplies immediately. It will cite a cash difficulty, or it will say nothing. The evidence is therefore built as a bundle: the date of a ministerial circular, the fate of the same buyer's other foreign suppliers over the same period, central bank currency allocation notices, the entity's published accounts. One element is worth more than all the others, and it must be sought early: what the debtor kept paying, to whom, and on what date. A buyer that settles its local suppliers and stops settling its foreign ones is not short of money, it is following an instruction.
The structure of the market follows the same dividing line. Public export credit agencies act as a market of last resort for what the private sector will not take alone, high-risk countries and tenors beyond seven years; private insurers cover short-term risk on solvent markets. The history of the line says as much: the first public agency, in Britain, was founded in 1919, in the aftermath of a war that had shown the private market withdrawing exactly when it was needed, and Coface was created in France in 1946. The tension between public export support and international subsidy rules produced, in 1978, an OECD arrangement setting minimum premiums and common conditions. The same unpaid invoice can therefore be placed with an agency or on the market, and the two do not recognize the same loss nor require the same proof.
The practical consequence fits in one sentence, and it reverses the order in which people believe they work. Classification does not happen at the claim, it happens at underwriting, when choosing which cover to buy and over what scope. A file that discovers its classification on the day of notification has already lost what matters, because the documents that would have proved the cause were contemporaneous with the facts, and nobody will produce them two years later.
A French turbine manufacturer has delivered to a state-owned energy company. Two instalments are unpaid, four months apart. On the first, the buyer wrote that it was facing cash difficulties. On the second, it wrote nothing, and the exporter learns from a peer that the supervising ministry has instructed state-owned companies to suspend payments to foreign suppliers. The exporter is covered by trade credit insurance for commercial risk, and by political risk cover for sovereign non-payment. What does it notify, and under what?
Two unpaid instalments, one debtor, and very probably two different losses. The first carries the marks of commercial default: the debtor writes that it cannot pay, which is exactly what trade credit covers, and its letter is the evidence. The second is the opposite: nothing is written, and what is known comes from a third party. The temptation is to notify everything under one policy for convenience, and that is the most expensive mistake in the file, because each cover has its own trigger and its own constitutive period, and a claim notified under the wrong one is not redirected, it is late under the right one. The work to do now, while the facts are fresh, concerns the second instalment: date the ministerial instruction, establish whether other state-owned companies stopped paying their foreign suppliers on the same date, and check whether this same buyer kept settling its local suppliers. That last point is what separates cannot pay from is not allowed to pay, and nobody will produce it two years from now.
- 01The nature of the debtor decides nothing: the cause of the default decides which cover responds.
- 02Three situations that get confused: the public entity forbidden to pay, the state that suspends by moratorium, and the blocked transfer, which is not a non-payment but a non-transfer.
- 03Evidence of the cause is built while the facts unfold, never afterward: the debtor has no interest in writing down why it is not paying.
- 04The fact that separates cannot pay from is not allowed to pay is almost always the same: what the debtor kept settling with others, on the same date.