Chubb posted $575 million of favorable short-tail development in the third quarter of 2026, and buried inside the same release was $13 million of adverse development in long-tail casualty, concentrated entirely in accident years 2019 through 2022 (Q3 2026 earnings call transcript, Oct. 21, 2026). CEO Evan Greenberg told analysts Chubb is “slow to recognize favorable development but very quick to recognize adverse,” and that accident years 2023 and later are too “green” to touch in either direction: “we wouldn’t touch them.” Total reserves exceed $71 billion, and the paid-to-incurred ratio slipped to 80% for the quarter versus 86% year to date.
Chubb is the carrier that moves last and moves little. When the market’s most conservative reserver swings negative on a specific four-year window, that is not noise.
Who it affects
Self-insured employers, captives, and public entity pools that sit beneath excess casualty or umbrella towers written between 2019 and 2022 are the direct read-through. The signal is sharpest for programs with meaningful general liability, umbrella, or excess casualty exposure in that window, including hospital captives carrying professional liability limits above a retained layer and group captives writing excess GL for construction, trucking, or municipal members.
Reserve mechanism
Greenberg’s disclosure is a case-adequacy and incurred-development signal, not a frequency or severity story. Chubb did not cite a new loss trend; it cited development runoff landing worse than booked for 2019-2022 specifically, while 2023 and later remain unassessed in either direction. That asymmetry matters for anyone pricing a tail factor. If a carrier with Chubb’s reserve discipline is still finding adverse emergence four to seven years into those accident years, the implied tail factor for mid-maturity casualty years is running longer than many self-insured programs assume. The silence on 2023-and-later is its own data point: there is no external signal yet for the most recent retained layers, so actuarial opinions leaning on industry benchmarks for those years are flying without that confirmation.
This result lands alongside a run of similar disclosures this year. Everest flagged roughly $200 million in casualty reserve strengthening across most accident years in Q2, AIG cited adverse excess casualty development tied partly to a decade-old accident year, and Lockton’s market analysis found $1.6 billion of commercial auto and liability adverse development concentrated in accident years 2022-2023. Five carriers and one broker analysis independently landing on the same accident-year window since May stops looking like company-specific conservatism and starts looking like a cohort problem working through the industry’s 2019-2022 casualty book.
What this means for your next review
Ask your actuary to isolate accident years 2019 through 2022 in your own GL, umbrella, or excess liability triangle and compare actual incurred emergence against what the last reserve study projected. Separately, ask your excess or reinsurance broker whether renewal quotes for towers written in that window have begun to reflect carrier-side strengthening, or whether pricing is still lagging the signal. For fronting and reinsurance structures, confirm your captive manager understands how upstream reserve actions could affect collateral asks before the next renewal cycle.