A sectoral decree nationalising all of a country's mining concessions strikes every investor in that sector on the same day. Each holds its own policy and believes its file isolated; they are not, because several of them are often insured with the same carrier, and because that carrier does not reason by insured but by event. This module looks at the loss from that side, the insurer's, and it is useful to the insured precisely for that reason: it explains behaviour that nothing in its own policy allows it to foresee.
The asymmetry producing everything else fits in one sentence. An insured's policy limit applies per insured, whereas the insurer's own reinsurance protection most often applies per event, across all insureds. A carrier that has written four hundred million euro policies in one country must therefore pay four hundred million if the same decree strikes them together, while its own cover reimburses only a single tranche above a single retention. It carries the gap net, and that gap appears on no document its insureds can read.
That configuration explains the question that decides considerable sums and looks scholastic until one sees it operate: do the measures constitute one event or several? A single decree of 14 March, applied the same day to four concessions, looks like one event. Four separate prefectoral orders, taken over three months, applying an announced policy, look like four. The classification changes nothing in what each insured receives, since each has its own limit; it changes everything in what the insurer recovers in reinsurance, and it is therefore the insurer, not the insured, that has an interest in settling it one way.
The practical effect must then be named, because an insured can suffer it without ever understanding where it comes from. An insurer seeking to gather several losses into one event for its own reinsurance needs a common date and a common cause. It will therefore tend to adopt, for all its insureds in the country, the same date of triggering event, that of the decree, whereas one investor had lost control six months earlier through a series of particular measures and another lost it six months later. The date that suits the reinsurance attachment is not necessarily the accurate one for each file, and an insured that accepts without argument the date it is offered can lose a policy year by it.
Wordings organise that grouping through a related claims clause, deeming to constitute a single loss all facts arising from one originating cause. That clause operates in both directions and one must see which way it runs for the insured. Between several insureds it does not concern it: it has its own limit. But it operates within its own policy, where a series of staggered measures is reduced to one loss, therefore to one limit and one deductible. Depending on the size of the loss, that same clause is favourable to it, by making it bear only one deductible, or unfavourable, by capping at one limit what several would have covered.
A timing consequence deserves stating because it is discovered too late. If a series of measures is reduced to a single originating cause, the loss is deemed to have occurred at the date of that cause. An investor struck by the last measure of a series begun before its policy incepted may therefore be told that the loss predates its cover, when it suffered nothing before. It is the same mechanism as that of creeping expropriation, applied no longer to a file but to a public policy, and it is harder to argue because the originating cause there is public and dated.
Finally one must say what an insured can actually do with all this, without which the module would remain a lesson in insurance economics. Three things. Work out its own date of triggering event, with its own documents, and not adopt for convenience the date the insurer offers everyone. Ask, where the policy allows, how the related claims clause operates on its own series, and do the arithmetic on both hypotheses before taking a position. And know that an insurer simultaneously exposed on several insureds in one country investigates more slowly and settles less readily, not from ill will but because its own cover is under discussion while it handles the file.
This module closes on an observation reaching beyond its subject. An insured reads its policy and believes it has read its risk; it has read the half that concerns it. The other half is its insurer's position, which decides the speed, the rigour and the generosity with which its file will be handled, and which appears in no document it holds. Inferring it from what can be observed is a skill, and it begins with knowing that limits are not counted the same way on the two sides.
A decree of 14 March 2026 transfers all of an Oceanian country's nickel concessions to a state company. Four investors are affected, all insured with the same European carrier, for limits of 120, 95, 60 and 210 million euros. The carrier holds excess of loss protection of 300 million per event, above a 50 million retention. One of the four, operating a concession in the east of the country, had suffered from June 2025 the refusal to renew its export permit, the requisition of its stocks and the appointment of an administrator by the mining authority; it had then considered notifying, without doing so, the insurer having judged the measure premature. That carrier now proposes to its four insureds that 14 March 2026 be adopted as the single date of the triggering event. What should the fourth do?
One must first see who benefits from the proposed date, and the answer is not the insured. A single date of 14 March allows the insurer to present the four files as one event to its own reinsurers, and so to recover a 300 million tranche above a single 50 million retention, rather than bearing four retentions and an argument about attachment. On total commitments of 485 million, what is at stake for it runs into tens of millions. None of that is improper and none of it appears in the insured's policy, which it nonetheless needs to know in order to understand the insistence it meets. For the fourth insured, 14 March is probably the wrong date and a costly one. The facts of June 2025, refusal to renew the export permit, requisition of stocks, appointment of an administrator, are the classic sequence of a creeping expropriation, and effective control was very likely lost by then. Adopting March 2026 moves it into a later policy year, with that year's terms, limit and deductible, which are not necessarily those of 2025, and above all it weakens its own chronology by starting it nine months too late. It must therefore work out its date for itself, with its own documents, and not adopt for convenience a date built for another purpose. One point deserves attention and runs against it: the insurer judged the measure premature in 2025 and the insured did not notify. That absence of notification will be raised, and the answer is to produce the exchanges of the time, showing that the insurer itself was then discouraging notification, which bars it from complaining of that today. Finally the related claims clause of its own policy must be checked and the arithmetic done on both hypotheses before taking a position: if its 2025 series and the 2026 decree are reduced to one originating cause, it bears only one deductible but it is capped at one limit, and depending on the size of its loss that clause helps or hurts it.
- 01A policy limit applies per insured, the insurer's reinsurance per event across all insureds: it carries the gap net and that appears on no readable document.
- 02One event or several changes nothing in what each insured receives, and everything in what the insurer recovers: it is therefore the insurer that has an interest in settling it one way.
- 03A single date offered to all a country's insureds serves the reinsurance attachment, not the accuracy of each file: work out your own date.
- 04The related claims clause operates within your own policy: one deductible helps, one limit does not, and the arithmetic depends on the size of the loss.
- 05Reducing a series to one originating cause dates it to that cause: an investor struck last may be told the loss predates its cover.