The WireCaptives & Alt Risk

Markel's Enstar Cover Prices a Runoff Tail at 10% of Limit

Markel will pay Enstar a $120 million risk premium for $1.2 billion of adverse development cover above $4.8 billion of runoff reserves. The new argument: the headline price only works because Markel also hands over $2.3 billion of reserves at face value, so captives should read 10% of limit as a floor, not a quote.

On October 27, 2026, five Markel insurance companies signed an adverse development cover agreement with Cavello Bay Reinsurance Limited, a wholly owned subsidiary of Enstar Group. Markel’s Form 8-K and the October 28 announcement give the terms. The portfolio is $4.8 billion of net reserves on 2025 and prior underwriting years: the Global Reinsurance division now in runoff, plus exited risk-managed professional liability, architects and engineers, and other runoff lines. Markel retains the first $2.6 billion. Enstar takes the next $2.3 billion, then provides $1.2 billion of cover above the $4.8 billion. Markel pays a $120 million risk premium for that top layer. Closing is expected by the end of the fourth quarter of 2026, subject to regulatory approval.

Legacy transactions of this size rarely publish this many terms. For a captive board weighing how to close out retained risk, this one is a usable price point, provided it is read correctly.

The structure in four numbers

All arithmetic below is ours, on the filed figures.

MetricArithmeticResult
Premium as a share of limit$120 million / $1.2 billion10.0%
Premium as a share of subject reserves$120 million / $4.8 billion2.5%
Point where Enstar starts paying$2.6 billion / $4.8 billion54% of carried reserves
Point where the cover exhausts($4.8 billion + $1.2 billion) / $4.8 billion125% of carried reserves

Two details matter. First, the retention and the ceded layer do not sum cleanly: $2.6 billion plus $2.3 billion is $4.9 billion against $4.8 billion of subject reserves. On the third-quarter earnings call, Brian Costanzo put the transferred amount at $2.25 billion, so the gap is rounding. Second, the cover applies to covered losses paid on or after April 1, 2026, per the 8-K, so the $2.6 billion retention has been eroding since spring.

The pure adverse development layer attaches at 100% of carried reserves. Our explainer on loss portfolio transfers and adverse development covers notes that attachment points are sometimes set at 105% or 110% of the central estimate. Markel bought protection from the first dollar of deterioration, with no corridor.

Who it affects

Captive owners and risk retention groups with closed long-tail programs; self-insured groups considering a buyout of old accident years; and any board that has been quoted, or expects to request, an adverse development cover. Markel’s book is reinsurance and professional lines at carrier scale. The loss development does not transfer to a captive. The deal economics do.

The reserve mechanism: why 10% is not the whole price

The lever is the tail above carried reserves, and the price of removing it. At 10% of limit, the premium implies Enstar views a full $1.2 billion loss as remote. But the $120 million is not Enstar’s only compensation.

The 8-K sets the reinsurance premium at $2.4 billion, satisfied by transferring assets at closing. Costanzo described the pieces on the call: $2.25 billion of reserves transferred, plus the $120 million risk premium. That sums to $2.37 billion, in line with the filed $2.4 billion. So the last-to-pay $2.25 billion moves at roughly face value, with no credit for the time value of money. Costanzo said a good proxy for the forgone return is about a 4% pretax yield on that $2.25 billion. Four percent of $2.25 billion is about $90 million a year at the outset, declining as claims pay. Because Enstar’s layer is last to pay, behind a $2.6 billion retention, it holds those assets for a long time.

That is the mechanism our explainer on captive reserve discounting describes: the assuming party’s margin sits largely in the discount. A captive that asks for the top layer alone, keeping its assets, removes that income from the reinsurer’s side of the ledger. The risk premium has to rise to replace it.

The second support for the price is on Markel’s side. Simon Wilson told analysts the premium “will be funded by management margin in the reserves,” and Costanzo said the premium and the margin release should be “relatively neutral” in the fourth quarter. The announcement says the cover reduces the prudency management needs in its best estimate. Markel was carrying margin above its central view and is exchanging that margin for contractual protection. A captive booked at its actuarial point estimate has no margin to release. For that captive, the same premium is a new cost, not a swap.

Third, capital. Costanzo’s rough estimate was that the transaction frees “a little under $0.5 billion” of capital. A single-parent captive capitalized to a domicile minimum will not see relief on that scale.

Counterparty terms are part of the price

Per the 8-K, $500 million of the transferred assets goes to the reinsurer outright and the balance goes into a trust account for the ceding companies’ benefit. Markel keeps claims control, subject to claims protocols, and may commute the agreement upon certain recapture trigger events. Markel describes Cavello Bay as an S&P “A” rated platform.

Those terms deserve attention from the group that booked a $205 million bad debt charge at State National in the second quarter, on paper issued for a reinsurer that Insurance Journal describes as now bankrupt. We covered that in Markel’s fronting charge, and the wider trend in the fronting collateral squeeze. Costanzo confirmed that the cover does not include State National, as Insurance Journal also reported. A captive buying a cover creates a recoverable, and should negotiate trust funding with the same care a fronting carrier now applies to the captive. See the gross, ceded, and net bridge.

Commutation rights have value too. Two days before Markel’s filing, The Hartford reported a $497 million pre-tax net gain from commuting its asbestos and environmental adverse development cover, receiving $1.12 billion. Covers are not always held to expiry. actuary.info’s analysis of the Markel deal sets the two transactions side by side.

What this means for your next review

Ask your actuary for three things before requesting quotes. One: the distribution above the central estimate, with the probability that ultimate losses exceed 100%, 110%, and 125% of carried reserves. Markel’s cover spans 100% to 125%; you need to know how much of your own risk sits in that band. Two: carried reserves in undiscounted and discounted form, so you can see what ceding assets at face value would hand to the reinsurer. Three: a statement of whether carried reserves include margin above the central estimate. If not, model the premium as a direct charge to surplus.

Markel’s motive is also worth noting. Tom Gayner said the purpose was “to reduce the uncertainty.” Costanzo said the Global Reinsurance book had shown an adverse trend in transaction liability claims over the prior three to six months, and Gayner said the runoff division cost 3 points on a 91.9% combined ratio in the quarter. Buyers come to this market when the tail is already moving, and the reinsurer prices with that knowledge.

Our read

We read 10% of limit, or 2.5% of carried reserves for a 100% to 125% band, as a floor for captive buyers. The price depends on three supports a typical captive lacks: $2.25 billion of assets ceded at face value, reserve margin to fund the premium, and a diversified multi-billion-dollar portfolio. A captive seeking a standalone cover on a single casualty program should expect a quote well above 10% of limit, an attachment above carried reserves instead of at them, or both. Firm casualty reinsurance terms, which we tracked at mid-year renewals, point the same way.

The first test arrives with Markel’s year-end results: whether the fourth quarter is in fact neutral, and how much margin release it took to make it so. If the release is close to $120 million, the benchmark holds. If Global Reinsurance reserves needed strengthening before closing, the true cost of this cover was higher than the headline.

Sources