From reviewing Schedule P triangles across the top 20 commercial lines writers over the past three years, the deterioration pattern in the 2021-2024 accident year vintages is more consistent than any single carrier's earnings commentary suggests. The industry narrative through 2023 held that adverse reserve development was a legacy problem, confined to soft-market accident years 2015 through 2019 where pricing discipline broke down. Rate increases of 15% to 20% annually during the 2020-2024 hard market were supposed to have closed the adequacy gap for newer vintages. That assumption is now failing.

Milliman's analysis of 2024 statutory filings reveals $15.8 billion in adverse prior-year development across casualty lines, the highest figure on record and the first net adverse result for the industry since 2017. Assured Research pegs the other liability (occurrence) line alone at $12.5 billion deficient as of year-end 2025, with $10.5 billion of that shortfall concentrated in accident years 2021 through 2024. AM Best estimates commercial auto remains under-reserved by $4 billion to $5 billion industrywide. These are not isolated signals. They represent a structural underestimation of casualty loss costs that extends well beyond the soft-market vintages the industry has already acknowledged.

This analysis maps the reserve strengthening trend using Schedule P data, carrier 10-K disclosures, reinsurer reserve actions, and AM Best commentary. It examines why backward-looking actuarial benchmarks are producing systematically low estimates, how the workers' compensation favorable development buffer is declining, and what the downstream implications look like for carrier combined ratios and reinsurance treaty negotiations.

The Record: $15.8 Billion in Casualty Adverse PYD

Calendar year 2024 marked a turning point in the industry's reserve development story. According to Milliman's U.S. Casualty Insurance 2024 Financial Results study, total adverse prior-year development across all liability lines reached $15.8 billion, more than double the $3.7 billion recorded in 2023. This was the first year since 2017 that the industry posted net adverse PYD after seven consecutive years in which favorable development from workers' compensation and short-tail lines masked the casualty deterioration underneath.

The aggregate numbers are striking, but the line-level breakdown reveals where the stress is concentrated:

Line of Business 2024 Adverse PYD % of Prior Reserves
Other Liability (Occurrence) $10.0 billion 6.6%
Commercial Auto Liability $3.8 billion 5.7%
Non-Proportional Reinsurance Liability $1.7 billion 4.6%
Product Liability (Occurrence) $0.4 billion 2.8%
Workers' Compensation (Favorable) ($6.4 billion) (4.3%)

The pattern here matters as much as the magnitude. Workers' compensation released $6.4 billion in favorable development during 2024, and short-tail commercial lines contributed additional favorable releases. For years, these offsets papered over the casualty deterioration. In 2024, the casualty shortfall finally overwhelmed the favorable lines, producing the $7.8 billion net adverse industry figure that Milliman calculated across all P&C lines combined. The total adverse figure of $15.8 billion for casualty lines alone, representing 1.5% of prior reserves, captures the full scope of the problem before offsets.

Insurance Thought Leadership's cohort analysis adds granularity. Commercial lines writers with limited workers' compensation exposure posted 2.6% adverse development for accident years 2022 and prior through Q3 2025. More diversified carriers, those with substantial workers' comp or personal lines books, showed only 0.3% adverse development for the same period. The divergence confirms that the masking effect is real: carriers whose portfolios include profitable, favorably developing lines appear healthier in aggregate, even when their casualty reserves are deteriorating at the same rate as their pure-play peers.

Commercial Auto: 14 Consecutive Years of Underwriting Losses

Commercial auto liability has been the most visible casualty reserving problem for over a decade. AM Best's 2026 analysis confirms the line posted its 14th consecutive year of underwriting losses in 2024, with the annual shortfall reaching $4.9 billion. The two-year total for 2023 and 2024 exceeded $10 billion. The 11-year average annual loss stands at more than $2.9 billion. Fourteen of the top 20 commercial auto insurers recorded combined ratios above 100 in 2024.

Milliman's statutory financial results for commercial auto liability in 2024 show a weighted average calendar year loss and DCCE ratio of approximately 86%, the highest in five years. The initial accident year loss ratio estimate for 2024 came in at 80.3%, but one-year adverse development added 8.0 percentage points, underscoring how quickly initial picks are proving inadequate. Direct written premium grew 12.3% to just over $43 billion, yet the rate need continues to exceed 10% annually based on Milliman's analysis, a threshold the industry has failed to sustain consistently since 2017.

The severity picture explains why rate increases have not solved the problem. AM Best reports claim severity in commercial auto is increasing at approximately 8% annually, well above the 3% rate of general economic inflation. The liability coverage segment alone posted a $6.4 billion loss in 2024, the largest on record. Physical damage coverage generated a $1.5 billion profit, also a record, but the 24.6-point gap between liability and physical damage results highlights how concentrated the deterioration is in the bodily injury and liability components where litigation dynamics and settlement inflation dominate.

AM Best projects the industry remains under-reserved in commercial auto by $4 billion to $5 billion, even after the adverse development already recognized. The adverse losses from accident year 2021 and later now exceed $2.7 billion, directly contradicting the assumption that hard-market pricing corrected the reserving shortfall. As AM Best noted in its market segment report, "adverse loss development has been a constant drain on commercial auto results and is getting worse."

Other Liability: The $12.5 Billion Deficiency

Other liability (occurrence), which includes general liability, excess and umbrella lines, captures the broadest exposure to social inflation, litigation funding, and nuclear verdicts. Carrier Management's reporting on Assured Research's year-end 2025 analysis quantifies the deficiency at $12.5 billion, improved from $15.0 billion at year-end 2024 but still representing the largest single-line reserve gap in the industry.

The accident year composition of the shortfall is the critical finding. Of the $12.5 billion deficiency, $10.5 billion is concentrated in accident years 2021 through 2024. Assured Research projects ultimate loss ratios approximately 5 percentage points higher than published industry figures for accident years 2021 through 2023, with the gap narrowing to roughly 2.5 points for accident year 2024 and minimal deficiency for accident year 2025. This progression suggests the industry is beginning to reflect higher severity in its current-year picks, but the accumulated shortfall in the 2021-2023 vintages remains largely unaddressed in carried reserves.

Calendar year 2025 produced $7.3 billion in adverse development in the other liability (occurrence) line, the largest adverse figure across all P&C lines. Of that amount, $3.9 billion came from one-year adverse development on accident years 2021 through 2023, with nearly $3 billion concentrated in accident years 2022 and 2023 alone. A significant shift is visible in where the adverse development originates: only $1.0 billion of 2025's adverse development related to accident years prior to 2016, compared to 2021 when 77.9% of adverse development came from the oldest accident years and only 14.1% from recent vintages.

That migration of adverse development from legacy years to recent years is the structural change that matters most. When adverse development was dominated by pre-2016 accident years, the reserving problem was finite and declining. Actuaries could reasonably project that the legacy tail would run off and stop generating surprises. Now that the 2021-2024 vintages are the primary source of adverse development, the problem is self-renewing: each successive accident year enters the reserve base carrying the same systematic underestimation that plagued its predecessors.

The Hard-Market Vintage Surprise: Why 2021-2024 Is Different

The core assumption that sustained the market's confidence through 2023 was straightforward: the 2015-2019 soft-market cycle produced inadequate pricing, so those accident years naturally developed adversely. The hard-market cycle beginning in 2020, with double-digit rate increases across commercial casualty lines, would produce adequately priced vintages that developed favorably or at worst held flat. This assumption appeared reasonable when initial loss picks for accident years 2020 through 2022 came in at approximately 67%, consistent with historical hard-market profitability. But Milliman's data shows those initial picks crept upward toward 70% during the 2020-2022 period, and subsequent development has pushed actual results higher still.

Several factors explain why hard-market pricing did not deliver reserve adequacy:

Severity trend acceleration outpaced rate. Amwins' 2026 State of the Market outlook reports casualty loss trends holding steady at 12% to 15% annually. Cumulative rate increases during the hard market were substantial, often 50 to 100 points over several years in excess casualty segments. But the severity trend was accelerating simultaneously, driven by litigation funding, plaintiff firm capitalization, and settlement value inflation. The rate increases were chasing a moving target, and the target was moving faster than actuarial trend assumptions captured.

Backward-looking benchmarks produced biased selections. Actuarial reserving methods rely heavily on historical development patterns. Link ratios, Bornhuetter-Ferguson expected loss ratios, and frequency-severity models all calibrate to observed historical data. When that historical data itself is developing adversely, the benchmarks carry the underestimation forward. An actuary selecting development factors from a triangle where the 2016-2019 diagonals are still deteriorating will produce development factor selections that are too low for the 2021-2024 diagonals. The problem compounds because each annual re-estimation appears to bring the older years closer to ultimate, validating the selected pattern, while the true ultimate continues to drift higher.

Extended settlement cycles delayed recognition. Moody's analysis highlights that social inflation tends to hit personal auto first because those claims settle more quickly. Commercial casualty claims, particularly in general liability and excess lines, have extended settlement cycles that can span five to eight years or longer. Accident years 2021 and 2022 are now entering the development window where settlements accelerate and actual payments begin to diverge from IBNR estimates. The adverse development observed in 2024 and 2025 reflects settlement activity on claims that were incurred during the hard-market period but are resolving at severity levels that the original reserves did not anticipate.

Management optimism anchored loss picks too low. Insurance Thought Leadership's analysis identifies management optimism as a contributing factor, noting that even within cohorts posting aggregate favorable development, 41% of individual companies showed adverse results. The pressure to report favorable prior-year development, which directly improves calendar-year combined ratios and earnings, creates an incentive to maintain loss picks that subsequent development ultimately corrects upward.

The Workers' Compensation Offset Is Running Dry

For over a decade, workers' compensation has been the industry's most reliable source of favorable reserve development. NCCI's 2024 State of the Line report confirms the line posted an 86% combined ratio, its 11th consecutive year of underwriting profitability. The accident year 2024 combined ratio came in at 99%, and NCCI estimates the industry's redundant reserve position at $16 billion. Calendar year 2024 produced $6.4 billion in favorable development, up slightly from $6.0 billion in 2023.

These numbers look stable on the surface. But the trajectory underneath is shifting. Patterns we've tracked across carrier filings show lost-time claim frequency declining 5% in 2024 versus 2023, continuing a long-term trend that has driven much of the favorable development. Average claim severity, however, rose 6% in 2024, and medical cost trend is accelerating. NCCI's data shows medical severity jumping meaningfully, with pharmaceutical and device cost inflation entering the claim stream. Net written premium declined 3% to $41.6 billion as competitive pressure pushed rates lower.

The offset mechanism is weakening for structural reasons. Years of substantial reserve releases have drawn down the redundancy cushion. Rate softening in workers' comp reduces the margin embedded in newer accident years, which in turn reduces the favorable development those years will generate as they mature. Moody's observed that in calendar year 2023, workers' comp and short-tail lines produced $12.5 billion in favorable development that offset $10 billion in adverse development from general liability, commercial auto, and personal auto. By 2024, casualty adverse PYD had grown to $15.8 billion while workers' comp favorable development held at $6.4 billion. The offset ratio is deteriorating, and the gap will widen if casualty severity trends persist at current levels while workers' comp margins compress.

Carrier-Specific Reserve Actions Tell the Story

Aggregate industry data establishes the trend. Individual carrier reserve actions confirm its breadth and severity. A cross-section of major reserve strengthening disclosures from 2024 and early 2026 illustrates the pattern:

Carrier Reserve Action Lines Affected Period
Swiss Re $2.6 billion net strengthening P&C Reinsurance Full year 2024
Everest Re $1.7 billion total strengthening U.S. casualty (primary + reinsurance) Full year 2024
Liberty Mutual $1.3 billion adverse development Other liability Calendar year 2025
Chubb $0.7 billion adverse development Other liability Calendar year 2025
CNA 4.0 combined ratio points Excess casualty, professional E&O Q1 2026
Selective Insurance ~$190 million strengthening Commercial auto Full year 2025

Swiss Re's Q3 2024 reserve addition of $2.4 billion in P&C Re, bringing the full-year net strengthening to $2.6 billion, was among the most consequential. CEO Andreas Berger described the action as positioning reserves "at the higher end of the best-estimate range," adding 13.3 percentage points to the segment's combined ratio. Everest Re's $1.7 billion total included $684 million in reinsurance segment U.S. casualty reserves and $1.1 billion in the insurance segment, with CEO Jim Williamson citing "aggressive underwriting action in certain classes exposed to social inflation."

CNA's Q1 2026 results showed the pattern continuing into 2026. The company's combined ratio rose to 102.2% from 98.4%, with unfavorable prior-year development adding 4.0 points, driven by excess casualty and professional E&O. CNA described the action as adding "prudence to recent-year reserves in excess casualty." Selective Insurance's $190 million in commercial auto reserve strengthening during 2025, concentrated in accident years 2024 and 2025, illustrates how the problem is manifesting in mid-market writers, not just large commercial and reinsurance carriers.

The breadth of these actions across primary carriers, specialty writers, and reinsurers indicates a systemic pattern rather than company-specific underwriting failures. When Swiss Re, Everest, Liberty Mutual, Chubb, CNA, and Selective are all strengthening casualty reserves within the same 18-month window, the signal is industry-wide.

Litigation Cost Drivers: Why the Trend Is Not Mean-Reverting

The casualty reserve deterioration is ultimately a claims severity story, and the litigation cost drivers behind that severity show no signs of reverting to historical norms. Three structural factors are compounding simultaneously.

Nuclear verdicts continue to escalate. Data compiled across multiple industry sources shows 89 nuclear verdicts (those exceeding $10 million) totaling $14.5 billion in 2023, a 15-year high. Thermonuclear verdicts exceeding $100 million are increasing in frequency. Two-thirds of nuclear verdicts concentrate in three lines: product liability (24%), auto accidents (23%), and medical liability (21%). These are precisely the lines showing the most adverse reserve development in Schedule P data.

Third-party litigation funding has become a structural capital market. The U.S. litigation funding market reached an estimated $17 billion by 2021 and is projected to exceed $30 billion, according to multiple industry analyses. The market grew 44% between 2019 and 2022. Average internal rates of return for litigation funders ranged from 20% to 35% in 2019 and 2020. This level of return attracts permanent capital, not speculative money. Litigation funding extends case duration, finances more extensive discovery and expert testimony, and increases the settlement floor because funders need to recover their investment plus return before the plaintiff sees proceeds. Nine states had pending legislation to regulate third-party litigation funding as of early 2026, but no comprehensive federal framework exists.

Plaintiff firm capitalization has transformed the litigation market. Large plaintiff firms now operate with balance sheets that allow them to hold cases longer, invest more in trial preparation, and reject early settlement offers that would have resolved claims at lower severity in previous cycles. The combination of litigation funding and plaintiff firm capitalization creates a feedback loop: higher verdicts establish new anchoring points for settlement negotiations, which in turn increase severity on subsequent claims, which attracts more capital into litigation funding.

For reserving actuaries, the critical implication is that backward-looking trend assumptions systematically understate the severity trajectory. A trend selection based on 10-year historical severity data includes years when litigation funding was smaller, nuclear verdicts were less frequent, and settlement anchoring points were lower. Projecting that historical trend forward will underestimate the severity of claims currently in the pipeline. This is not a temporary deviation that will self-correct; it is a structural change in the litigation cost function.

Reinsurer Exposure Through Casualty Excess-of-Loss Treaties

The primary market's reserve deterioration transmits directly into the reinsurance capital cycle through casualty excess-of-loss treaties. This continues a pattern we first documented in our reinsurance market analysis. When primary carriers' loss estimates for individual claims cross attachment points that were originally expected to contain them, the reinsurer's exposure activates retroactively. The $1.7 billion in adverse PYD on non-proportional reinsurance liability that Milliman recorded in 2024 (4.6% of prior reserves) quantifies this transmission mechanism.

Swiss Re's $2.6 billion net reserve strengthening and Everest Re's $1.7 billion total strengthening both reflect the reinsurance sector's recognition that casualty severity on underlying claims has exceeded the attachment points embedded in treaties written during 2019 through 2023. Fitch Ratings' 2026 outlook identifies U.S. casualty reserve adequacy as a key risk for the reinsurance sector, noting that reserves set during the pandemic years of 2020 and 2021 may prove insufficient for 2026 settlement values.

Reinsurers have responded by raising attachment points, tightening contract language, demanding greater transparency into cedents' claims practices, and requiring multi-year collateral commitments and higher retentions. These structural changes in treaty terms will eventually force primary carriers to retain more casualty risk, which will accelerate the pressure on primary company reserve adequacy.

The reinsurance market dynamics amplify the primary signal in both directions. During the reserving deterioration phase, adverse development on ceded claims flows through to reinsurer results with a lag, creating a second wave of reserve strengthening that pressures reinsurance capital and pricing. During the correction phase, higher reinsurance pricing and tighter terms flow back to primary carriers through increased ceded costs, compressing underwriting margins and potentially triggering additional primary reserve strengthening as actuaries reassess the net retained position.

Industry-Wide Context: Strong Results Mask the Underlying Vulnerability

The reserve deterioration is unfolding against a backdrop of headline industry profitability that may be creating complacency. AM Best reports the industry posted a 92.6% combined ratio for 2025, generating $61 billion in underwriting profit, nearly triple the $22 billion recorded in 2024. Policyholder surplus grew 11.4% to $1.2 trillion. Pre-tax operating earnings reached $153.1 billion, 43% higher than the prior year. Personal auto alone contributed $14 billion in net underwriting profit in 2024, a dramatic reversal from the $17 billion loss in 2023 and $33 billion loss in 2022.

Fitch projects the industry combined ratio to rise to 96% to 97% in 2026, with return on surplus declining from 10.1% to 9.1%. These projections assume stable reserve development. If casualty adverse PYD continues at 2024 levels or accelerates, the combined ratio impact could be 2 to 3 points worse than forecast. A $15 billion adverse PYD swing on an industry writing approximately $900 billion in net premiums earned translates to roughly 1.7 combined ratio points, enough to move the 2026 result from adequate to marginal.

The risk is that the industry's strong surplus position and recent profitability encourage carriers to compete on price precisely when reserve adequacy is deteriorating. Amwins reports that double-digit rate increases from 2024 have slowed to flat-to-single-digit renewals for most commercial accounts. If the rate deceleration continues while casualty severity trends persist at 12% to 15% annually, the 2025 and 2026 accident years will enter the reserve base with the same structural under-pricing that is now producing adverse development on the 2021-2024 vintages.

Why This Matters for Reserving Actuaries

The data presented here has direct implications for actuarial practice in the current reserving cycle.

Loss development factor selections need contemporaneous adjustment. Traditional link ratio selections based on industry aggregate triangles will understate development for carriers concentrated in commercial auto and general liability. Actuaries should consider whether their selected development patterns are calibrated to the experience of pure commercial lines writers (2.6% adverse development) or diversified carriers (0.3% adverse), and which profile more closely matches their company's book composition.

Expected loss ratio selections should reflect forward severity trends, not historical averages. With casualty severity trending at 12% to 15% annually per Amwins and claim severity in commercial auto running at 8% per AM Best, Bornhuetter-Ferguson expected loss ratios selected from the most recent three to five accident years are already biased low. The initial picks for those years are themselves proving inadequate, which means the a priori assumption carries the underestimation into current-year estimates.

The workers' comp offset cannot be assumed stable. Actuaries who rely on aggregate company-level development that blends casualty and workers' comp results should decompose their analysis to isolate casualty development from the favorable workers' comp trend. As the offset diminishes, aggregate metrics will deteriorate faster than line-level indicators might suggest.

Documentation under ASOP No. 43 requires explicit treatment of social inflation assumptions. The Actuarial Standards Board's guidance on unpaid claim estimates expects actuaries to disclose material assumptions and their basis. Given the industry-wide evidence of systematic underestimation in casualty reserves, an actuary who selects development patterns and severity trends without explicitly addressing social inflation, litigation funding growth, and nuclear verdict frequency may face questions about whether the actuarial opinion adequately reflects known conditions.

Stress testing against the 2015-2019 development pattern provides a relevant baseline. The soft-market vintages' development path is now well-documented. Applying that development speed and magnitude to the 2021-2024 accident years provides a plausible adverse scenario. If the hard-market vintages ultimately develop at even 60% to 70% of the soft-market pace, the reserve shortfall will be material for most commercial casualty writers.

The next 12 to 18 months of development on accident years 2022 and 2023 will be particularly revealing. These years are now entering the settlement acceleration window where actual paid losses diverge most sharply from IBNR estimates. The degree to which Q3 and Q4 2026 reserve reviews confirm or reverse the adverse trend will determine whether the 2021-2024 vintage becomes a reserving cycle on par with 2015-2019, or something substantially worse.

Further Reading

Sources