Insurers Fund Their Cyber Reinsurance Recoveries Before Paying a Single Claim
Reinsurance recoveries get booked as assets before a single claim is paid. That timing is the quiet engine behind reported loss ratios in cyber, D&O and professional lines. Follow the premium and the picture changes: what looks like underwriting profit at the top of the income statement is often a ceded cash flow, and the recoverable sitting on the balance sheet is a promise, not a payment.
Cyber Recoveries Funded Before Claims
An insurer writes a policy, collects premium, and immediately cedes a share to a reinsurer. The ceded premium leaves as cash. The recoverable that comes back is an asset, recognised on the balance sheet as a receivable. It does not require a claim to exist. It requires a treaty and an accounting convention.
That convention matters because loss ratios are calculated on earned premium, not on cash. A carrier that cedes 60% of a book reports a net loss ratio on the retained 40%. If the ceded layer absorbs the early attritional claims, the net ratio can look stable while gross claims climb. The recoverable grows in the meantime, funded by premium rather than by claim settlements.
The funding gap is the point. Cash goes out the door as ceded premium on day one. Cash comes back as recoveries months or years later, after claims are notified, adjusted and agreed. In the interim the carrier is financing the reinsurer. That is a working-capital position, and it shows up in the cash flow statement, not the loss ratio. For a growing cyber book, the drag can be material: a carrier doubling premium year on year is also doubling the cash it advances to reinsurers before any recovery lands.
Corgi and the New Carrier Model
Corgi is an American insurtech that became a licensed admitted carrier, writing directors and officers liability, cyber, commercial general liability and AI liability cover for technology startups. As of mid-2026 the company was valued at $2.6 billion following a Series B1 round, according to press reports. The valuation rests on distribution and data, not on retained risk.
That distinction is the whole model. A new carrier without decades of loss history cannot retain much of what it writes. It cedes the majority of premium to reinsurers upfront, keeps a fee-like margin, and books recoverables against the ceded share. The admitted licence lets it issue paper; the reinsurance treaty lets it survive the first bad year.
AI liability cover is the interesting line here. It is new, it has almost no credible loss experience, and reinsurers price it accordingly. Capacity is thin and terms are tight. A carrier writing it is effectively renting the reinsurer's balance sheet, and the rent is the cession rate.
Where the Premium Actually Goes
Gross written premium is the headline. Net retained premium is the number that pays claims. The gap between them is the cession rate, and it varies enormously by line. Cyber and D&O books at newer carriers can cede well over half of premium. Mature casualty books often cede far less, because the carrier has the capital and the loss data to hold the risk.
The mechanism is straightforward. Reinsurance protects the primary insurer from large or aggregate losses, and it frees underwriting capacity: ceded liabilities stop counting against the carrier's own capital base. That is why a thinly capitalised insurtech can write a large book quickly. It is also why the reinsurer, not the carrier, sets the practical ceiling on growth.
Loss ratios follow the ceded share. A 70% gross loss ratio and a 55% net loss ratio can describe the same book. The difference is where the losses land. This site has argued before that cyber liability carriers recover more from reinsurers than they pay out in claims, and the same arithmetic shows up in D&O, where carriers cede premium to sidecar reinsurers to manage capacity rather than to shed risk alone.
Parametric Triggers and Basis Risk
Parametric cover replaces claims adjustment with an index. A payout triggers when a defined threshold is breached, and settlement is fast because nobody argues about the size of the loss. For cyber, the index might reference an outage duration, a service-level breach, or a modelled event severity. The trigger definition is the contract.
The trade-off is basis risk, and it sits with the insured. A company can suffer a severe, well-documented incident and receive nothing because the index did not trip. Conversely, a company can receive a payout for an event that barely touched it. Parametric cover pays on the index, not on the loss.
Reinsurers price trigger reliability directly. A trigger that fires too often is expensive; one that fires too rarely is cheap and nearly useless. The actuarial work is in the correlation between the index and actual losses, and that correlation is estimated from thin data in most cyber lines. Hedging the estimate is wise.
AI Underwriting and the Hype Cycle
Automation genuinely speeds up submission triage. It reads applications, flags inconsistencies, and routes risks to the right underwriter. That reduces expense ratios and shortens quote turnaround. It does not change the loss ratio, because the loss ratio depends on what happens after the policy is bound.
AI liability cover remains a niche product. The exposure is real but the loss history is nearly absent, so pricing leans on expert judgement and reinsurance terms. Capacity, not demand, is the binding constraint. A carrier can only write what its reinsurers will accept, and they will accept a limited amount of an untested line.
The hype cycle confuses distribution gains with underwriting gains. Faster submission flow is a real improvement in expense. It is not evidence that the underlying risk is better understood. Those are separate questions, and only one of them shows up in the combined ratio.
What to Watch in Reinsurance Treaties
Track cession rates quarterly. A rising cession rate on a stable book means the carrier is retaining less risk and paying more for capacity, which shows up in net premium before it shows up in losses. Compare that trend against the reinsurer's own commentary on capacity; when reinsurers signal a pullback, cession rates tend to rise further, and the cost of that capacity lands on the primary carrier's net margin.
Read recoverable disclosures closely. Ask how much of the recoverable is collateralised, how old it is, and whether any counterparty dispute is pending. An uncollateralised recoverable from a thinly rated reinsurer is a different asset from cash.
Compare gross and net loss ratios side by side. If net looks materially better than gross, the improvement is coming from the treaty, not from underwriting skill. That is not necessarily a problem, but it should be understood as a structural feature, not a sign of superior risk selection.
Ask about trigger definitions on any parametric layer. Request the historical index values and check how often the trigger would have fired against actual losses. A trigger that has never fired in a back-test is either well-designed or untested; the difference matters when the next event arrives.
Model cash timing, not just totals. A recoverable that arrives eighteen months after the ceded premium leaves is a financing cost, and it belongs in the analysis next to the loss ratio. For fast-growing books, that cost compounds.
Cash Timing and the Cost of Capacity
The gap between ceded premium out and recoveries in is a financing cost that rarely gets a line item. For a carrier ceding half its premium, the cash advanced to reinsurers can run into the tens of millions for a mid-sized book, and that cash is unavailable for claims, investments, or growth. The cost is not the cession rate itself but the time value of money tied up in the interim. In a rising rate environment, that opportunity cost is higher than it was a few years ago, and it deserves a place in any net margin calculation.
Not all ceded premium is equal. Proportional treaties, where the reinsurer shares a fixed percentage of premium and losses, create a steady cash drag. Excess-of-loss treaties, which only respond above a retention, involve less upfront cash but leave the carrier exposed to frequency. The mix matters: a book heavy in proportional reinsurance will show a larger recoverable asset and a longer cash conversion cycle than one leaning on excess-of-loss. Neither is inherently better, but they finance differently, and the balance sheet reflects that.
Collateral arrangements change the picture again. Some reinsurers post letters of credit or hold funds in trust, which reduces the carrier's credit risk but does not eliminate the timing gap. Others operate on an uncollateralised basis, which improves cash flow but adds counterparty exposure. The trade-off between security and liquidity is real, and it shows up in the quality of the recoverable, not just its size.
For analysts, the practical takeaway is to treat the recoverable balance as a working-capital item, not a pure asset. Track its growth relative to premium, compare it to the cession rate, and ask how much of it is collateralised. A recoverable that grows faster than premium is a signal that cash is being advanced more aggressively, and that has implications for liquidity and for the true cost of the reinsurance programme.
This article is informational and does not constitute personalised financial, legal or insurance advice. Readers should consult a qualified professional about their own circumstances.
