Value Packs / Insurance / Claims Severity & Reserve Early-Warning / Brief 02

Claims Severity & Reserve Early-Warning · Brief 02

Attritional Severity Running Hot — Commercial Trucking

A steady claim severity × a frequency the rate never priced — the reserve-and-rate adequacy question behind a book that never dips under break-even

DataGenie Brief · autonomous
25+ programs · loss-runs + bordereaux + exposure · 0 raw rows moved

The one line you configure

Is our attritional claim cost on any program structurally outrunning what the rate and reserves assume — quarter after quarter, with no single event to point at?

In a nutshell

Commercial Trucking carries an attritional loss cost the rate never caught up to — a steady ~$30K claim severity at a frequency of ~25 per $M earned, holding flat quarter after quarter with a benign 3.7% large-loss ratio (no tail to reserve for). Held at a book-level 40% reserve ratio, the program still runs above break-even every month — a reserve-and-rate adequacy question, not a shock to reserve against.

At a glance

Claim severity$30Ksteady, ~book levelattritional
Claim frequency~25/$Mflat — not spikingstructural
Large-loss ratio3.7%benign — no tail to reserve for
Reserve ratio40%book level — adequacy in question

What's happening · the combined ratio has not re-entered the healthy band in a year

Commercial Trucking combined ratio (%)+8 pts
88%94%100%106%112%chronic — never dips under 100%Aug-25Oct-25Dec-25Feb-26Apr-26Jun-26
Now104%
Break-even100%
Months over 100%12 of 12

No spike, no event — just a flat ceiling above break-even. A tail-severity view sees nothing here; the chronic-combined view sees a program that has not earned its keep in a year.

Why · steady frequency and steady severity — attritional, not a shock

40/$M30/$M20/$M10/$M0/$MQ3-25Q4-25Q1-26Q2-26now
Claim frequency (per $M earned)
Average severity ($K)

Both lines are flat — frequency ~24–25 per $M, severity ~$30K. Neither is spiking; the loss cost is simply structurally high for commercial auto and the rate was set below it. This is the mirror image of Propane's episodic tail.

Where the 104% comes from · loss cost plus a heavy expense load

Loss ratio (paid + reserves)attritional BI · auto liability · cargo74.5%
Ceding commissionproducer acquisition cost16.0%
LAEclaim-handling on a litigious line (22%)7.8%
Operating expense5.5%

The four components sum to the 104% combined. The loss ratio alone (74.5%) is not alarming — it's the loss cost plus a 16% commission and a litigious-line LAE load that push the whole program past break-even. Levers: rate on the loss side, commission terms on the expense side.

Commercial-auto attritional cost — structural, all four states

Trucking bodily-injury, auto-liability and cargo losses arrive at a steady ~$30K severity and a claim frequency the plan never re-based. There's no venue or event to isolate — GA, MA, PA and TX all run alike — so the answer is a rate/terms and reserve-adequacy decision on the whole program, not a per-claim intervention.

So what

Commercial Trucking's loss cost is quiet and permanent — a steady attritional severity × frequency that keeps the program above break-even every month. Its large-loss ratio is benign at 3.7%, so there is no shock to reserve against; the whole gap is attritional loss cost the rate has never priced for, held at a book-level reserve ratio. That makes it a rate-and-reserve-adequacy decision, not a per-claim one — and because it's structural across all four states, the filing is program-wide.

The points that matter

Chronic, not episodic

The combined ratio has stayed between 100% and 108% for twelve straight months — no single event, no tail; the program simply runs above break-even.

12/12months over 100%

No tail to reserve for

The large-loss ratio is a benign 3.7% — this is attritional frequency × severity, so the answer is rate and terms, not a reserve strengthening or a per-risk cap.

3.7%large-loss ratio

It's structural across the footprint

GA, MA, PA and TX all run ~104% — there's no venue to trim, so a program-wide rate filing is the lever, not a state-by-state cull.

4 statesall ~104%

Questions it already answers

Is this a deteriorating book or just priced too cheap from the start?

Priced too cheap — the combined ratio is flat, not rising, so the loss cost isn't worsening; the rate was set below a commercial-auto loss cost that runs structurally hot.

Would a per-risk cap or reinsurance help, like it does on Propane?

No — the large-loss ratio is 3.7%, so there's no tail for a cap to catch; ceding attritional loss just cedes margin. This is a rate/commission decision.

Is any single state or producer dragging the average?

No — all four states sit at ~104% on one carrier (Clear Blue); the drift is the whole program, which is why the filing is program-wide.