Global commercial insurance rates fell 5% in Q1 2026, a seventh consecutive quarterly decline after seven years of increases (Marsh). US property led at -10% while US casualty went the other way at +9%. Casualty adverse prior-year development hit $15.8 billion in 2024, a record for those segments. A carrier holding both sides is running two different cycles through one set of reviews.

Key Takeaways

  • US property down 10% while US casualty rose 9% in the same quarter, with US casualty the only major line in the only major region still hardening.
  • Commercial auto posted its 59th consecutive quarterly increase at +5.8%, a run spanning nearly 15 years, and still closed 2025 at a 103.5% net combined ratio.
  • $15.8 billion of adverse prior-year development in 2024 across the four primary casualty segments, with other liability occurrence alone at $10.3 billion against $4.7 billion in 2023.
  • The adverse development sits in accident years 2022 through 2024, the nominally hard years that reserving convention expects to develop favorably.
  • Personal auto closed 2025 at 91.8% and homeowners at 88.1%, so for only the second time in 13 years personal lines are projected to outperform commercial.

The Split, in the Two Datasets That Track It

Marsh's index covers roughly 2,000 large commercial accounts a quarter. Its Q1 2026 edition shows property down 9% globally with double-digit declines in five regions, cyber extending to a twelfth consecutive quarterly decline, and D&O and financial lines compressing, against US casualty up 9%.

The CIAB's broker-level index puts the same market in domestic terms. Average commercial premiums fell 1.2% in Q1 2026, ending 33 consecutive quarters of increases running since 2017: property down 5.5%, workers compensation down 3.7%, cyber down 3.5%. Commercial auto rose 5.8%, its 59th consecutive quarterly increase, a streak of nearly 15 years that predates the nuclear verdict acceleration, the expansion of litigation funding, and attorney representation in commercial auto liability claims passing 50%. It was still not enough: the line closed 2025 at a 103.5% net combined ratio.

Line of Business Q1 2026 Rate Change 2025 Net Combined Ratio Cycle Direction
Commercial Property -5.5% (CIAB) / -10% (Marsh U.S.) ~86% (profitable) Softening
Workers Compensation -3.7% (CIAB) ~90% (profitable) Softening
Cyber Liability -3.5% (12th consecutive decline) ~84% (profitable) Softening
D&O / Financial Lines -3% to -5% ~85% (profitable) Softening
Commercial Auto +5.8% (59th consecutive increase) 103.5% (unprofitable) Hardening
Other / General Liability +5% to +12% (U.S.) 108% (unprofitable) Hardening
Medical Professional Liability +3% to +8% 106% (unprofitable) Hardening
Personal Auto Normalizing (-1% to +3%) 91.8% (improved 3.5 pts YoY) Stabilizing
Homeowners Normalizing (-2% to +4%) 88.1% (lowest in a decade) Normalizing

AM Best projects industry net premium growth slowing to 4.0% in 2026 with the aggregate combined ratio rising to about 96.9 from 95.0. That average sits on top of lines behaving like separate businesses at separate points of the cycle.

Hard-Market Years That Developed the Wrong Way

Adverse prior-year development across other liability occurrence, commercial auto liability, non-proportional reinsurance liability and products liability occurrence reached $15.8 billion in calendar year 2024, the highest on record for those lines (S&P Global Market Intelligence). Other liability occurrence alone contributed $10.3 billion, more than double its $4.7 billion in 2023.

The source is what makes it structural. That development concentrates in accident years 2022 through 2024, when rate increases were running at the top of the ranges Marsh and CIAB now report. Reserving convention expects hard-market years to develop favorably, because tighter underwriting, higher rates and conservative initial picks should compound. The reverse happened, which leaves two readings: the 2022-2024 increases did not keep pace with severity trend in those years, or the development factors used to set the initial reserves were anchored to pre-social-inflation baselines.

Moody's put general liability deficiency at more than 4% of carried reserves as of year-end 2023 and commercial auto liability at roughly 3%. On a national book where GL reserves run to hundreds of billions, 4% is a directional signal rather than a calibration error, and it excludes accident years that have not yet reached the ages where prior deficiencies surfaced.

The chain ladder is the mechanism carrying the bias forward. Age-to-age factors built from historical diagonals will understate recent accident years if post-2020 years develop faster than 2015-2019 did, and adding more post-2020 data does not fix it, because that experience is itself developing adversely. There is no clean baseline left for a factor selection anchored purely to link ratios.

What the severity data implies instead is a shorter weighting window on the most recent accident year columns plus an explicit trend load on the tail, rather than a tail derived from curve-fitting history. The tail is where the effect concentrates: cases that once resolved in three to five years now run seven to ten with higher ultimate values, which shows up as a non-stationary shift in the 60-month and 84-month-to-ultimate factors across cohorts rather than as volatility that averages out. Litigation funding does the same thing structurally, removing the pressure to settle early and pushing higher-severity resolutions into later development periods.

The Mix Moves Without an Underwriting Decision

Personal lines inverted at the same time. Personal auto closed 2025 at a 91.8% net combined ratio, 3.5 points better than 2024, and homeowners at 88.1%, the lowest in more than a decade (Fitch). For only the second time in 13 years, personal lines are projected to outperform commercial.

The second-order effect lands on portfolio composition, and it is invisible in the aggregate. Take a carrier at 40% commercial property and 30% commercial casualty by premium in 2023. After two years of property rates falling 5% to 10% while casualty premiums rise 5% to 12%, the casualty share expands as a proportion of total premium even with casualty new business volume flat.

A heavier casualty weighting means longer reserve tails, higher reserve uncertainty per dollar of premium, and capital requirements that do not compress in step with the property softening. Extended tail duration also raises sensitivity to discount rate movements. None of that appears in the aggregate combined ratio, and a capital model not updated for the shifting mix overstates risk-adjusted return on equity, which then feeds target returns that accept casualty business at inadequate margins while overstating competitive capacity on property.

The review calendar compounds it. Property and casualty LDF and pricing reviews are conventionally produced on separate schedules by separate teams with independent sign-off and little cross-line coordination. That worked when both sides sat in similar phases of the cycle. It does not describe a market where one line is softening at 10% and the other hardening at 9%, and where the thing that changed most in the portfolio is the ratio between them.

Further Reading

Sources