Triple-I and CAS Put Excess Auto Tort Litigation at $43.84B. Their Two Methods Split Threefold.

Two counts of the same quantity sit one row apart on page 16 of the new Triple-I and Casualty Actuarial Society white paper: 106,100 excess settled or judged state motor vehicle tort filings under the federal-court-based method, 35,800 under the new state-data method (Triple-I and CAS, Figure 24, September 2026). A 40/60 blend turns them into 64,000, and 64,000 times $660,000 to $687,000 is the $42.77 billion that carries the headline.

The paper, authored by William Nibbelin and released September 30, puts the combined federal and state excess at $43.84 billion for 2015 to 2024. The October 1 release led with the 30-year high in federal filings. The method spread, and the flat severity series behind it, got no mention.

Key Takeaways

• $43.84 billion splits into $1.07 billion from federal courts and $42.77 billion from state courts, which the paper sizes at 25.3% to 30.5% of its own $143.6 billion to $173.1 billion auto liability inflation estimate and at 3.4% of ten-year auto liability net incurred loss and DCC.
• 2.96 to 1 is the ratio between the two state methods' excess filings, 106,100 against 35,800. Run on the state method alone the excess is about $25 billion; on the federal method alone it is about $72 billion. The published figure is a weighted blend of the two.
• Plus or minus 10% is the paper's own sensitivity to the weighting: a 50/50 blend lifts the excess value about a tenth and a 30/70 blend cuts it by the same, moving the 3.4% loss share to roughly 3.1% or 3.7%.
• $873,000, $872,000 and $839,000 are the paper's average amounts in controversy for motor vehicle tort judgments across three decades, with settlements falling from $774,000 to $668,000, so every dollar of the excess comes from the count of filings and none from their size.
• 0.89 to 0.80 is the fall in ISO bodily injury liability claims per 100 earned car years from 2015 to 2024 while severity rose from $17,014 to $28,278, the insurance series a reader has to supply because the paper's "frequency" is filings per nonfarm payroll employee.

Inside the $43.84 Billion

The federal piece is small and fully observable. The Federal Judicial Center's Integrated Database records 55,025 motor vehicle tort filings from 2015 to 2024, 18,090 in no-fault states and 36,935 in tort states. Applying each group's growth over its 2008 to 2014 baseline yields 2,821 excess filings, a 57.8% and 55.6% disposition rate trims those to 1,600 settled or judged, and multiplying by the average amount in controversy gives $1.07 billion (Figure 15). Filings reached 6,809 in 2024 after a 5.0% compound rate from 2014, following a 2.8% annual decline from 1995 to 2008.

The state piece is 97.6% of the total and almost entirely extrapolated. Complete filing and disposition data exist in 11 states covering 50.6% of the population, down from 18 states and 62% in the 2025 edition, after a state data aggregator the prior paper relied on was folded into the National Center for State Courts site without the same detail (Appendix A). That shrinkage is why a second method exists at all. Last year's $42.8 billion for 2014 to 2023 came from the federal-ratio method alone.

The denominators are the pair's own prior work. The October 2025 Increasing Inflation on Liability Insurance report by Jim Lynch and Nibbelin put increasing inflation at $91.6 billion to $102.3 billion for personal auto liability and $52.0 billion to $70.8 billion for commercial auto, which sum to the $143.6 billion to $173.1 billion range. The 3.4% share implies ten-year auto liability net incurred loss and DCC of about $1.29 trillion, consistent with the $132.0 billion of private passenger liability incurred losses NAIC data show for 2024 alone (Triple-I Facts and Statistics).

Figure 24: How the Two Methods Reach 64,000

Excess value in this framework is a product of four terms: filings, the excess percentage of filings above the 2008 to 2014 trend, the disposition rate for settlements and judgments, and the average amount in controversy. The two state methods share the fourth term, which is borrowed from federal data because no state publishes award amounts. They disagree on the first three, and mostly on the second.

Term	Federal-based method	State-based method
State filings, 2015 to 2024	5,183,000	4,191,000
Excess percentage above trend (no-fault / tort)	3.5% / 4.3%	2.6% / 1.5%
Disposition rate (no-fault / tort)	41.9% / 53.9%	43.2% / 48.2%
Excess filings settled or judged	106,100	35,800
State excess value if used alone	$70.7 billion	$24.1 billion
Share of $143.6B to $173.1B, with federal $1.07B	41% to 50%	15% to 18%
Published 40/60 blend	64,000 filings; $42.77 billion; 25.3% to 30.5%

Figures 20, 23 and 24 of the white paper; standalone values computed as excess settled or judged filings times $687,000 for no-fault states and $660,000 for tort states. The 40/60 weights reproduce 63,920, which the paper rounds to 64,000; the footnote to Figure 24 reads as a simple average, which would give 70,950.

The filing counts differ by 24%, the disposition rates by a few points. The excess percentage is where the methods separate: scaling the 11 complete states up by their share of federal filings produces a 4.1% state filing CAGR from 2014, while building the series from partial state data produces 1.9%, against the same 2008 to 2014 baseline. In tort states the gap is 4.3% against 1.5%. Because excess percentage multiplies everything else, a factor of 2.9 in that term becomes a factor of 2.9 in the result.

The paper weights the state method at 60% and explains why: it is "built directly from partial, real state civil case data" and the authors expect its role to grow as more states publish. It also reports that a 50/50 weighting raises the excess value about 10% and a 30/70 weighting lowers it about 10%. On the 3.4% loss share, that is a band from roughly 3.1% to 3.7%, with the method endpoints at about 1.9% and 5.6%.

For an actuary carrying an excess-litigation load into an auto liability trend selection, those endpoints are the useful numbers. The blended 3.4% averages to about a third of a point of loss ratio a year across the decade, the state method to a fifth of a point and the federal method to over half a point. The paper's preliminary year-by-year allocation shows the share rising across the period, so the recent-year load sits above whichever average is chosen. A trend built on the headline figure inherits a 40/60 judgment the authors themselves describe as transitional.

A Count-Driven Excess Measured Against a Severity-Driven Base

The title compares litigation to claims frequency, and the paper's frequency is filings per nonfarm payroll employee from the St. Louis Fed's PAYNSA series: up 4.5% a year for motor vehicle tort against a 0.9% decline for other federal civil cases. Insurance claim frequency never appears. ISO data on Triple-I's own statistics page show bodily injury liability claims per 100 earned car years falling from 0.89 in 2015 to 0.80 in 2024, about 1.2% a year, while severity rose from $17,014 to $28,278, about 5.8% a year (Triple-I, citing ISO). Supplying that series sharpens the title: filings rose while claims fell.

The paper's own severity series then points the other way. Judgment amounts in controversy held at $873,000, $872,000 and $839,000 across three decades; settlement amounts fell from $774,000 to $706,000 to $668,000; and plaintiff awards, up across decades from $377,000 to $681,000, ran a negative 2.5% CAGR from 2015 to 2024 (Figures 12 to 14). Two attempts to add a severity component were abandoned. A credibility weighting at the 1,082 standard raised the countrywide average because thinly populated states had lower amounts, and a severity trend fit had no statistically sound coefficient of determination (Appendix A).

So the $43.84 billion is entirely a count effect, expressed as a share of an estimate built the other way. Announcing the October 2025 inflation report, Triple-I and the CAS wrote that across all liability lines "claim severity, not claim frequency, is driving loss increases" (CAS, October 2025). Morgan Bugbee of the CAS left the reconciliation to the reader: loss trends "may not be fully explained by claim frequency alone" (Triple-I and CAS release, October 1, 2026). The 25.3% to 30.5% share is a ratio of two different mechanisms, and the site's social inflation and litigation trends analysis finds the severity channel carrying most of the casualty-wide figure.

The remaining assumptions all sit in the state method the paper expects to dominate. States publishing only filings or only dispositions get a one-to-one clearance rate; states with no motor-vehicle-specific disposition rate get the general civil rate; and for some states the only source for some years was an aggregator that has ceased operating, so the prior period's figure was carried forward. Each of those is a choice that holds the count down, and each will be replaced as states publish, in a direction nobody can sign.

That is the bind the 40/60 blend leaves. The authors say the state weight should rise as its data improves, and at 100% the headline falls from $43.84 billion toward $25 billion unless states also begin publishing award amounts, which none does today. The method change that makes the estimate more defensible each year also shrinks it, while the paper's own preliminary allocation says the per-year share is climbing. A litigation load selected from this series will be revised down by method and up by data in the same future edition, and an auto liability trend has to be set before either arrives.