Every answer and its explanation appears here once you have finished the path. Each one then links to the matching glossary entry, where the concept is set out in full with its worked example.
1. Three subsidiaries, losses of 12, 9 and 6 million, a one million deductible per claim, a primary layer from 0 to 20, a country limit of 25. The group is tempted to declare three claims to open three limits. Is it right?
No: the country limit caps everything at 25 whatever happens, while three claims call for three deductibles instead of one, two million lost for no gain
The arithmetic settles it in two lines, which beats a discussion of principle: the extra limit is useless beyond 25, and the extra deductibles cost a certain two million. The right position is a single claim, 27 million of loss, one deductible, 26 indemnifiable, brought back to 25. The answer invoking a prohibition through common cause mistakes the register: nothing forbids arguing three claims, it is simply unfavorable here. And this result does not generalize: as soon as the per-claim limit bites before the country limit, the calculation inverts, which is precisely why it is redone on every file.
Glossary entry · agregat2. The module states that the definition of the claim helps on deductibles and hurts on limits. What follows?
Compute the position both ways before notifying, because no general rule holds
The cost of splitting is immediate and known, deductibles can be counted; the gain is conditional and depends on two caps acting in different places, the per-claim limit and the country limit. Neither general rule therefore holds, and both answers proposing one are wrong for the same reason. Letting the insurer decide hands the other side a choice with a computable price, and defers to a characterization that is arguable rather than to a calculation that can be checked.
Glossary entry · franchise3. The country aggregate is shared with insureds you cannot see, in a queue nobody publishes. What follows?
Notifying early is an economic argument and not merely a duty: the aggregate is consumed in the order files settle
Nothing shares the aggregate pro rata and nothing requires its status to be published: it is consumed in the order files settle, meaning a slow file can find an exhausted aggregate for reasons entirely outside itself. It is the only situation in this subject where the insured's diligence acts on what it will receive independently of the merits of its claim. Believing the aggregate has no effect on the insured is the error that leads to unhurried handling while others are being paid.
Glossary entry · point-attachement4. The primary layer defines the claim by common cause, the upper layer applies a seventy-two hour clause. Three orders spread over five weeks. What happens?
The primary sees a single claim and pays its limit, the upper layer sees three events none reaching its attachment point, so it never attaches
The flaw is visible in neither contract read separately: the primary is coherent, the upper layer is coherent, and it is their juxtaposition that opens the hole. An hours clause borrowed from natural catastrophe describes a windstorm well and describes nothing of a state measure built over weeks: it is a drafting borrowing, not a coverage choice. Follow-the-fortunes, invoked by another answer, is a real reinsurance mechanism and does not operate here: between two layers of a direct insurance programme each contract keeps its own definition of the event.
Glossary entry · clause-horaire5. Three caps coexist in this programme. Which bites most often, and why is that counterintuitive?
The country limit, though it is the one least looked at because it does not appear in the schedule of covers the insured bought
The per-claim limit and the annual aggregate appear in the schedule and are read at renewal; the country limit lives on the insurer's side, in its exposure management, and only shows up when it blocks. It bites all the more because a political crisis strikes several insureds in the same country at the same moment, which is the very definition of the risk. That is why it decides more often than the other two while being the least present in anyone's mind, and why having it named at placement beats discovering it at the claim.
Glossary entry · couche-de-reassurance