The broker’s triangle: balancing cost, capital and exposure relief

Every cedent wants three things from a reinsurance programme: it should be cheap, it should release capital, and it should take away the exposures that keep the board awake. Two of them are usually easy. The craft is in the centre of the triangle.

The three corners

Cost. Not the premium but the net cost: what you pay, less any ceding commission, less what you expect to recover.

Capital relief. The reduction in required capital, on whichever measure binds: regulatory, rating agency or internal model.

Exposure relief. The reduction in what can actually hurt you: the net 1-in-200 loss, the largest net single-name exposure, the concentration in one country or sector.

The value test

Capital costs money. If the cedent’s cost of equity is $k$, reinsurance creates value on capital grounds when the capital it releases is worth more than it costs:

The ratio $\text{NC}/\Delta K$ is the cost of capital released, and it turns very different structures into one comparable number. Exposure relief does not fit neatly into it, which is why this is a triangle and not a line: some protection is worth buying even when the capital maths alone says no, because a single loss would breach appetite, a covenant or a rating trigger.

Where the classic structures sit

An illustrative credit portfolio, with the cedent’s cost of capital at 10%:

Quota share sits near the capital and exposure corners but far from cost: on a profitable book you give away margin on every policy. It earns its place when you need growth capacity or the ceding commission is generous.

Per-buyer excess of loss is pure exposure relief. It looks expensive per unit of capital, because a granular credit book’s capital is driven by the systemic tail rather than any one name, but it is the cover that stops one insolvency becoming a headline.

Aggregate stop loss attacks what actually drives credit capital, a recession that hits many buyers at once, so it is usually the cheapest capital relief available. Combined with per-buyer cover, it gets close to the centre.

Why the deal can work for both sides

If reinsurers priced the same risk on the same capital, there would be no triangle to optimise, only a transfer of margin. The value comes from diversification. A credit tail that consumes a lot of a specialist cedent’s capital consumes much less inside a reinsurer whose portfolio is dominated by property catastrophe, because, as I argued in The quiet hedge, the two rarely fail together. The trade creates value when

so the reinsurer can price below the cedent’s own cost of holding the risk and still earn its return. A broker’s job is to find the reinsurers for whom that gap is widest.

What brokers actually optimise

Model gross and net consistently. The same simulated economy, the same loss distribution, every candidate structure.

Map the efficient frontier. For each structure, plot net cost against capital relief and mark the exposure relief. Discard anything dominated: a structure that costs more and relieves less than another is never the answer.

Tune the levers. Retentions and limits, reinstatements, sliding scale commissions, loss corridors, multi-year terms. Small changes move a structure a long way across the triangle.

Test it against the market. The theoretically best structure is useless if capacity for it is thin or expensive at this renewal.

The centre of the triangle is not one structure. It is the point where the cedent’s CFO, CRO and chief underwriter can all sign the same placement, and showing them that point, with numbers, is what a good broker brings.

Banu Munnangi is SVP, Credit Reinsurance AI at Marsh Re, where he leads data and AI for the Credit, Bond and Political Risk specialty. More writing · Expertise

Views are my own and do not represent my employer. Examples use illustrative figures, not client or employer data.