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. An insurer writes a pharmaceutical plant with an insured value of 800 million euros. Its fire treaty only covers risks up to 150 million. How does it place the remaining 650 million?
Facultative reinsurance, each risk being offered to the market individually
Facultative reinsurance, commonly called fac, is a placing method where each risk is offered individually: the cedant submits the risk, the reinsurer examines it and is free to accept or decline, hence the name. It stands against the obligatory treaty, where the reinsurer is contractually bound to accept everything falling inside the defined scope, without looking at files one by one. Three situations call for facultative, and the case here is the first: a high value risk exceeding the capacity of the existing treaty. The other two are the atypical risk, whose profile falls outside the treaty's criteria, and the risk on which the cedant wants an outside opinion before committing. In practice the placement runs through a slip shown to several reinsurers who each take a share of the capacity, which makes facultative slower and more expensive in handling costs than a treaty, but that is the price of a file by file examination.
Glossary entry →2. A cat XL treaty on a household book is written on a subject premium of 30 million euros, at a rate of 1.5 per mille. What is the treaty premium?
45,000 euros
The subject premium is the cedant's gross premium base used to compute the premium of an excess of loss or catastrophe treaty: treaty premium equals rate times subject premium, here thirty million times 0.0015, so 45,000 euros. It represents the total underlying exposure the layer covers, and it is what lets burning cost and rate on line be computed afterwards, two indicators that mean nothing without a stable base. What is actually negotiated is not the rate but the definition of the base: is the subject premium net of brokerage, does it include taxes, does it span every line or just one? A badly defined subject premium creates ambiguity about what the treaty really covers, and it surfaces at the moment someone asks what was covered, which is too late. As a benchmark, a burning cost reads off the same base: 450,000 euros of losses to the layer over thirty million is fifteen per mille.
Glossary entry →3. On a quota share treaty: ceded premiums 3 million euros, loadings 600,000 euros, ceded claims 1,200,000 euros. Profit commission is set at 25 percent. How much goes back to the cedant?
300,000 euros
Profit commission lets the cedant receive, after the fact, a share of the profit the reinsurer made on the treaty when experience comes in below an agreed threshold. The formula is constant: commission equals the rate times the difference between ceded premiums, loadings and ceded claims, where that difference is positive. Here, three million less six hundred thousand less one million two hundred thousand leaves a profit of 1,200,000 euros, of which twenty-five percent is 300,000 euros. The natural mistake is to apply the rate to ceded premiums, which would give 750,000: the rate applies to the profit, never to the base. Loadings comprise the reinsurer's expenses and the ceding commission already paid, which is why they come off before any sharing. The mechanism exists so the cedant keeps an interest in managing its own experience even though it has transferred the risk, and it is commoner in proportional treaties than in excess of loss. A sliding scale variant moves the rate with actual experience.
Glossary entry →4. A specialist insurer buys a cover triggering as soon as its annual loss ratio passes 115 percent, whatever caused the deterioration. Which form of reinsurance is this?
A stop-loss, or aggregate excess of loss
Stop-loss is a non-proportional form that protects not against an individual loss but against the deterioration of a whole book's experience across a period. The reinsurer responds once total incurred losses pass an agreed threshold, almost always expressed as a percentage of premium, and up to a limit. The distinction from the two excess of loss forms is the one that matters: per-risk excess of loss looks at each policy, per-event excess of loss looks at each catastrophe, stop-loss looks at neither, it looks at the year's technical result. That is why it answers equally to one large isolated loss and to a multitude of moderate ones, where the other two forms let the second pass through. It follows that stop-loss is the broadest protection a cedant can buy, and therefore the dearest: the reinsurer is taking on the volatility of the result, not merely the severity of an event.
Glossary entry →5. A liability cover is priced in 2020 on 2020 verdict levels, and the claim settles in 2030 against 2030 verdicts. What is the liability insurer really selling?
A long option on future litigiousness
A long tail line is one where the claim is reported and settled long after the premium is banked, sometimes a decade later, unlike property where settlement follows the event closely. That lag looks harmless and is dangerous, because it turns the liability insurer into the seller of a long option on future litigiousness: it takes a price today built on today's verdict norms, and will pay tomorrow on the norms, generally heavier, prevailing at settlement. The smallest annual gap becomes a chasm through compounding over ten years, which is arithmetic, not pessimism. Long tail is therefore the multiplier of social inflation, and it explains something that disorients newcomers: a reserving shortfall does not show when it forms, it shows years later, as the old years finish running off and nothing can any longer be corrected in the rates that produced them.
Glossary entry →6. In January 2022 an insurer adds a cyber exclusion to all its commercial package policies. Its catastrophe XL treaty, signed for three years in 2020, carries no equivalent exclusion. What is the problem?
The cover is no longer back-to-back, but the gap works in the insurer's favor
Back-to-back cover means an exact match between the scope of a direct policy and that of the treaty reinsuring it: every loss covered directly is covered by the treaty, and every policy exclusion appears in the treaty too. The gap described here is the favorable case, which is what makes it instructive: the policy excludes cyber, the treaty does not, so if a systemic cyber loss hit several insureds, the losses above the retention would flow to a treaty that still accepts them. The contract has no defect: it carries a temporary asymmetry the reinsurer will correct at renewal. The reverse gap is the expensive one: a policy covering what the treaty excludes leaves the insurer alone with its retention and everything above it. The causes of such gaps are ordinary rather than exceptional: policy and treaty renewals do not fall on the same calendar, an endorsement to a policy may never have been passed through to the treaty, and a multi-year treaty freezes terms the direct market moves every year.
Glossary entry →7. A reinsurer finds it has accumulated 4.5 billion euros of gross exposure across a set of zones covering the French Riviera and the Rhône delta. Which framework lets it see that, and compare with its competitors?
CRESTA zones, the market's standardized geographic breakdown
The CRESTA system, for Catastrophe Risk Evaluating and Standardizing Target Accumulations, is a geographic zoning framework run by a trade body of the world's main insurers and reinsurers. It splits each country into zones that are homogeneous for natural catastrophe exposure, at a granularity matching how much the country matters for natural perils: Germany has 130 zones, Japan 68, France 57. Its value lies in uniformity more than in geographic precision: a breakdown particular to each player would make accumulations incomparable, therefore unverifiable at market level, which is exactly the level at which a concentration turns dangerous. The finding described here leads either to an underwriting decision, restricting renewals in those zones, or to a transfer, for instance buying an index-linked protection on the insurance-linked securities market. The full chain is there: a shared framework makes an accumulation visible, and a visible accumulation becomes a decision.
Glossary entry →8. In 2025 Kenya raised from 20 to 25 percent the share of non-life treaties that must go to its national reinsurer. What is the structural limit of this mechanism?
It keeps the risk inside the economy that produces it, rebuilding the correlation reinsurance exists to break
Compulsory cession, or legal cession, requires a country's insurers to reinsure a fixed share of their business with a designated reinsurer, usually state-owned. The stated aims are coherent and rarely disputed: keeping premium in the country, building local capacity, reducing hard currency outflows. The limit lies elsewhere and it is structural rather than political. Reinsurance's first function is to break correlation, that is, to have a local risk carried by a balance sheet not exposed to the same peril. A compulsory cession does exactly the reverse by keeping the risk inside the very economy that produces it, so the national reinsurer will be in difficulty at the precise moment its cedants are too. The effect is sharpest where the peril is regional, earthquake, cyclone, drought, which is to say in the countries that most need outside capacity. Uganda for its part stacks three cessions on top of each other, compounding the same effect several times.
Glossary entry →9. An adverse development cover removes a cedant's volatility on prior years while leaving it most of the profit. What does it exchange its claims risk for?
An unpriced counterparty risk, correlated with the transferred peril
Retrospective reinsurance covers liabilities already incurred, through a loss portfolio transfer or an adverse development cover, and it is the core instrument of the legacy market. Its ambiguity comes in two steps. First, it does not extinguish the cedant's obligation to its policyholder: whatever becomes of the carrier, it is still the original insurer the insured can call on. Second, the exchange itself moves the risk more than it removes it, since a claims risk, known and reservable, becomes a counterparty risk that is neither priced nor reserved. And that counterparty risk is not independent of the first: the acquirer fails precisely when the liabilities develop badly, which is the very scenario the cover was bought against. The product also tends to sit less on dead books than on the prior years of very much living ones, where stripping out volatility while keeping the profit describes an accounting effect more than a transfer of risk, a question neither auditors nor supervisors pursue with any consistency.
Glossary entry →