Commercial auto is still the stubborn line in an otherwise softer insurance market. For delivery operators, that matters because the renewal is not just a back-office bill. It is a lane constraint, a bid constraint, and sometimes a growth constraint.
The current signal is mixed. Insurance Journal, citing The Council of Insurance Agents & Brokers' Q4 2025 survey, reported that commercial property/casualty pricing overall was the softest since 2017, with average premiums across account sizes up only 0.2%. But commercial auto remained the exception: the line averaged a 6.6% increase in Q4 2025 and had logged 58 straight quarters of increases, according to the same CIAB survey summary.
That is the key operating takeaway. A DSP or independent final-mile carrier may hear that the market is softening, but the van fleet renewal can still arrive higher. The reason is not mystery underwriting; it is loss experience. Insurance Journal separately reported AM Best's 2025 view that the U.S. commercial auto segment produced about $4.9 billion of underwriting losses in 2024. AM Best's problem area was liability, not vehicle physical damage: commercial auto liability posted a 113 combined ratio in 2024, while physical damage was profitable at an 88.6 combined ratio, according to Insurance Journal's summary of the report.
For operators, that split explains why brokers keep asking for driver files, telematics exports, camera status, MVR cadence, claims runs, radius, and use-case detail. The carrier is not only pricing the metal. It is pricing the chance that a stop, backing event, intersection crash, pedestrian incident, or plaintiff attorney turns a route-level incident into a large liability file.
Carrier appetite is therefore becoming more selective even when capacity exists. In the same CIAB-based Insurance Journal article, respondents reported less capacity for commercial auto risk. That does not mean every fleet is uninsurable. It means clean, documented fleets have more leverage than fleets that show up at renewal with missing driver qualification files, open claims, no coaching log, or unclear vehicle use.
Last-mile fleets should treat insurance as a controllable operating system, not a once-a-year purchase. The weekly disciplines matter: MVR checks before dispatch, fast incident reporting, photo evidence, camera retrieval, route-risk coaching, and documented corrective action after preventable events. If your insurer, MGA, or broker offers a safety portal, use it and keep proof. At renewal, the question is not whether you say you run safely; it is whether an underwriter can see it without chasing you.
The margin pressure is sharper because insurance is rising alongside other fleet costs. Fuel should be modeled on gasoline for last-mile vans, not diesel. The U.S. Energy Information Administration (EIA) gasoline figure provided for this assignment is $4.187 per gallon as of 2026-06-18. A fleet running Ram ProMasters, Ford Transit 350s, gasoline Sprinter 2500s, or gas step vans should use gasoline in cost-per-route and cost-per-stop math. When insurance adds another few points on top, the operator's room for error narrows: more paid miles, avoidable idling, reattempts, and unplanned rentals all compound the same daily margin problem.
The insurance discussion also reaches the bid desk. If a route package requires more vans, the incremental economics should include the expected insurance impact before the operator accepts the work. A new van can carry not only premium, but also deductible exposure, loss-run volatility, and the administrative load of adding drivers quickly. A route that looks acceptable on revenue per stop can turn weak if it requires hiring marginal drivers or stretching safety supervision.
ATRI's November 2025 research notice is useful even though it is trucking-wide rather than van-specific. The American Transportation Research Institute said trucking auto liability premiums rose 36% per mile over eight years and that insurance cost/availability and lawsuit abuse ranked among the industry's top issues in 2025. Last-mile operators should not import heavy-truck diesel economics from that research, but the liability-market lesson applies: insurers are reacting to severity, litigation risk, and sustained underwriting losses.
FleetOwner's October 2025 reporting points to the same management response. Its article emphasized that fleets are turning to ADAS, AI dashcams, and better legal-defense records as nuclear verdict pressure raises insurance and legal costs. The practical read for small and midsize delivery fleets is not to buy technology blindly. It is to make sure any system produces usable evidence: time-stamped video, driver coaching records, speed and harsh-event trends, maintenance documentation, and a clear process for preserving data after a crash.
Coverage design also deserves attention. Progressive Commercial's delivery and courier insurance page says cost varies with factors such as vehicle type, business size, location, coverage, number of drivers, and claims history. That list is a useful checklist before renewal. If your fleet changed zones, added peak vehicles, shifted from parcels to heavier items, or took on drivers with limited tenure, tell the broker early rather than letting the carrier discover it late.
The operator playbook for 2026 is straightforward. First, separate liability and physical damage decisions instead of treating commercial auto as one lump bill. Raising a physical-damage deductible may be reasonable for a strong balance sheet, but liability limits protect the company from the claim that can end the business. Second, build a renewal packet 90 days out: vehicle schedule, driver roster, MVR process, loss runs, telematics summary, camera policy, maintenance cadence, and safety-meeting logs. Third, push claims hygiene. Close small files, document subrogation opportunities, and make sure reserves are not stale.
Insurance is not soft for delivery fleets just because some commercial lines are softening. The more useful view is this: liability severity is keeping carrier discipline high, and operators with clean data can compete better for capacity. In a low-margin last-mile model, that is not an insurance department issue. It is fleet economics.