On July 30, 2026, Everest Group disclosed approximately $200 million in North America treaty casualty reserve strengthening in its second-quarter earnings, spread across what CFO Elias Habayeb described as “most accident years.” The charge was not confined to one bad underwriting year. It reflects a judgment that virtually the entire recent casualty book carried inadequate loss picks, and that waiting for the Q3 reserve study was not the right answer.
That breadth matters. A single-year blip can be attributed to anomalous development in one slice of the book. An adverse signal running across most of the recent accident-year stack means the development pattern itself was wrong, and any reserve opinion that drew on the same industry casualty triangles may be carrying the same understatement.
Who it affects
The immediate impact is on Everest’s ceded book, but the signal travels upstream. Self-insured employers, captives, and risk pools that layer excess or treaty reinsurance above a large-deductible or self-insured retention face two downstream risks: tail factor assumptions in their actuarial opinions that relied on the same industry triangles Everest is now revising upward, and collateral or retention increases that may follow if their fronting carrier or excess reinsurer is running a parallel strengthening exercise behind the scenes. The exposure is highest for programs with significant general liability, commercial auto, or umbrella exposure in accident years 2021 through 2024.
Reserve mechanism
When a reinsurer books strengthening “across most accident years” rather than one isolated period, the actuarial read is that the assumed development pattern was too optimistic system-wide. Tail factors for mid-maturity accident years (three to five years of development) were understated in the aggregate. If your actuarial opinion used ISO or NCCI industry triangles for GL or auto, those benchmarks incorporate the same underreported emergence that Everest is now correcting.
Everest’s simultaneous decision to cut casualty treaty gross written premiums 19% year-over-year reinforces the signal. The company is not reducing casualty because market rates softened; it is walking away because it believes the line is underpriced for the risk it carries. The June 17 launch of Annapurna Re, a $600 million casualty reinsurance sidecar with Stone Point Insurance Solutions as anchor investor and Mubadala Investment Co. as a strategic backer, extends that logic: transferring the exposure to third-party capital at current pricing is rational capital management when you view casualty as chronically undervalued. It also signals that casualty reinsurance capacity may become more selective and more expensive at the next program renewal.
The split between Everest’s charge and the divergent outcomes at Hartford and Travelers, documented in Q2 carrier reserve outcomes, shows that book composition and accident-year mix are producing wildly different results across the market. Uniform trend and pattern assumptions across programs are not defensible.
What this means for your next review
Ask your actuary whether your GL and auto tail factors draw on credible own data or industry development triangles. If the answer is industry triangles, the Everest Q2 charge is concrete evidence those tails may be understated for accident years 2021 through 2024. Ask separately whether your fronting carrier or treaty reinsurer has disclosed any casualty reserve strengthening in its own earnings: that disclosure has historically preceded collateral notices to large-deductible and captive programs by six to nine months. Lockton flagged the same development concentration as far back as May, documenting adverse emergence in AY 2021-2023 across the broader casualty market; see that analysis for the quantitative backdrop.
The Q3 reinsurer earnings cycle (October 2026) will confirm whether the Everest charge is an isolated company-specific correction or the leading edge of an industry-wide strengthening wave. For programs with meaningful casualty exposure, it is worth scheduling a diagnostic on your triangle before those disclosures arrive rather than after.