A cedant cedes because its treaty obliges it to, and the double automatism of the scope has already been established: inside the definition, the cedant must cede and the reinsurer must accept. That obligation is contractual, it is negotiated at renewal, and it stops where the scope stops. There is a second obligation to cede that comes from no contract and is not negotiated: the one the state imposes. It names the reinsurer, it fixes the share, and the cedant chooses neither its counterparty nor the amount. An outwards underwriter working in an emerging market meets both, and the order in which they are taken decides what is left to place.
Compulsory cession, also called legal cession, requires the insurers of a country to reinsure a fixed fraction of their business with a designated reinsurer, usually owned by the state or by a group of states. Three stated objectives are pursued and they are consistent with one another: retaining premium on the territory rather than watching it leave, building local reinsurance capacity that would not exist without a guaranteed flow, and reducing foreign currency outflows in economies where these are constrained. None of the three is a technical objective. They belong to economic policy, which is why they withstand the technical arguments put against them.
The limitation of the arrangement is structural, and it deserves to be stated precisely because it has nothing to do with the quality of the designated reinsurer. Reinsurance exists to move a risk out of the economy that produces it: a cedant placing a drought layer with a carrier on another continent swaps a counterparty correlated with its own exposure for one that is not. A compulsory cession to a national reinsurer does the reverse: it retains the risk inside the very economy that produces it, and so rebuilds the correlation the cession existed to break. The effect stays small on an idiosyncratic peril. It becomes decisive as soon as the peril is regional, which is precisely when reinsurance matters.
The figures show the movement continuing rather than receding. Kenya raised in 2025 from 20 to 25 percent the share of non-life treaties that must compulsorily go to its national reinsurer. Uganda stacks three superimposed cessions, and that stacking is the decisive point: the obligations do not replace one another, they add up, each taking its share of what is left after the previous one. A cedant can therefore see more than a third of its portfolio leave before it has placed a single euro of chosen reinsurance. The deduction happens at the top of the program, on gross business and not on the remainder: it cuts into the very material on which the rest of the plan is built.
What this changes for the outwards underwriter fits in one sentence: it does not build its program on its portfolio, it builds it on what the legal cessions leave behind. Three practical consequences follow. The capacity it then negotiates rests on a reduced base, so its treaties are smaller and their fixed costs spread over less premium. The diversification it presents to its reinsurers is already eroded by a share it did not choose. And it carries credit risk on an imposed signature that it can neither select to its liking nor replace.
The obligation still has to be performed, and this is where the cession bordereau becomes the central instrument. A bordereau is the line by line list of ceded risks: for each policy, the sum insured, the premium, the ceded share and the characteristics needed for exposure analysis. It is indispensable on a surplus treaty, where the cession is computed policy by policy, and on any account whose geographical accumulation the reinsurer wants to measure. Its quality is the chronic weak point of the chain: incomplete addresses, missing zone codes, absent values. Recent treaties expressly condition certain covers on the delivery of compliant bordereaux, and the errors and omissions clause covers only good faith failures, never the systematic absence of data.
The second instrument is the subject premium base, and it decides what a rate means. The base is the premium volume to which the contractual rate applies, and its definition is one of the most contested points of a treaty: premiums gross or net of acquisition commissions, net or not of cessions to prior treaties, taxes and management expenses in or out, inwards business from other carriers counted or not. A simple case shows what is at stake. Take a base presented at 240 million euros for a rate of 2.8 percent, whose definition now excludes a 31 million subsidiary brought in the previous year. On a restated base of 271 million, the equivalent rate falls to 2.48 percent, and the 4 percent price reduction on display becomes a reduction of 15.
On January 1, 2026, a non-life insurer in an East African market reports 180 million euros of gross written premium in commercial property. Three legal obligations bear on it, in this order: a 25 percent cession to its national reinsurer, a 10 percent cession to a regional reinsurer computed on what remains, then a 5 percent cession to a continental facility computed on the balance. It then wants to place a quota share at 40 percent on what it retains. How much is left for its own account, and what is to be said about the diversification of the resulting program?
The computation cascades rather than adds, and that is the first thing to see. The 180 million fall to 135 after the 25 percent national cession, to 121.5 after the 10 percent regional cession taken on that remainder, and to 115.4 after the 5 percent continental facility taken on the balance. The legal cessions have therefore carried away 64.6 million, or 35.9 percent of gross, not the 40 percent that adding the three rates would have produced. That four point gap is no accounting subtlety: it covers 7.4 million of premium, and an underwriter who adds the rates understates its own available capacity. The quota share is then placed on 115.4 million, cedes 46.2 and leaves 69.3 million net, or 38.5 percent of the opening gross portfolio. Then comes the second question, which is the one that counts. The resulting program looks heavily reinsured, since the cedant keeps only a little over a third of its premium. But the 64.6 million ceded by obligation sit with three carriers whose exposure is backed by the same regional economy as the cedant itself: a continental drought, a currency crisis or a sovereign shock reaches them together, and at the very moment the cedant calls on its recoveries. The diversification implied by the cession rate is therefore greater than the diversification actually achieved, and the gap between the two shows up in none of the usual ratios. Three consequences follow: the quota share is negotiated on a base cut by a third, the portion the cedant can present as freely placed is small, and the credit risk it carries on the three imposed signatures cannot be replaced.
- 01Two obligations to cede coexist: one contractual and negotiated, which stops where the treaty scope stops, the other legal and imposed, which names the reinsurer and fixes the share.
- 02Legal cessions stack in cascade rather than by addition: each takes its share of the previous balance, and 25 plus 10 plus 5 do not make 40 percent of gross.
- 03Retaining the risk inside the economy that produces it rebuilds the correlation the cession existed to break, and the effect becomes decisive as soon as the peril is regional.
- 04A legal cession is taken at the top of the program: the capacity negotiated afterwards rests on a reduced base, and the diversification presented to reinsurers is already eroded.
- 05The cession bordereau performs the obligation line by line, and its quality is the chronic weak point: the errors and omissions clause does not cover the systematic absence of data.
- 06The subject premium base decides what a rate means: a base whose perimeter moves makes any comparison of rates illusory from one year to the next.