Transport Insurance Pricing Guide

What determines price for transport insurance?

Transport sits at the intersection of two distinct risk shapes that share a single technical platform. Rail is a low-frequency, high-severity catastrophe business where a single derailment can move a portfolio's entire result for the year. Logistics liability is a higher-frequency, medium-severity business now exposed to a litigation environment that has materially reshaped the right tail of its loss distribution. Both share a structural feature that breaks conventional rating: the risk moves, and loss potential does not track the insured asset's value the way it does in property or workers' compensation.

This guide covers the exposure bases that replace conventional per-unit measures, the rating factors that carry weight in transport pricing, the actuarial methods suited to thin-data and heavy-tailed portfolios, and the severity environment now shaping rate adequacy.

Transport actuaries face three intersecting challenges that do not appear together in any other commercial line, and they shape every element of how the line is priced:

• The exposure base must capture both frequency and severity at once. Vehicle counts and asset value each understate the risk in isolation, so transport pricing typically requires multiple proxies in parallel.

• The severity tail is now defined by litigation environment as much as by physical exposure. Nuclear verdict trends, social inflation, and venue concentration are reshaping the upper layers of transport loss distributions faster than experience data can catch up.

• Pricing must integrate experience rating, exposure rating, and credibility-weighted methods within a single risk tower. No single technique captures both the attritional layer and the catastrophe layer of a transport portfolio.

The takeaways below frame how each of these constraints translates into pricing practice.

• Transport policies typically require separate frequency and severity exposure proxies, a decomposition largely absent in property or workers' compensation rating. Long-haul carriers are commonly rated on mileage as a frequency proxy and on gross vehicle weight or payload capacity as a severity proxy, with cost of hire capturing subcontracted capacity that vehicle counts miss.

• Rail operator liability scales sharply with operational footprint. Large Class I freight railroads carry catastrophe exposure that short line operators do not, and limit purchases differ accordingly, though specific limit profiles are commercially sensitive and not consistently disclosed in public sources.

• The U.S. corporate liability environment now drives a meaningful portion of transport severity trend. Marathon Strategies recorded 135 nuclear verdicts of $10 million or more in 2024, a 52% increase over 2023, with total awards reaching $31.3 billion, and it identifies commercial trucking among the most heavily affected sectors.

• Single-event severity in rail can rival a small-country catastrophe loss. Norfolk Southern's 2024 10-K reports approximately $1.1 billion in 2023 and an additional $325 million (gross) in 2024 of costs associated with the East Palestine derailment, before insurance recoveries.

• Statutory liability caps materially constrain placement at the top of the rail tower. 49 USC 28103 sets an aggregate $200 million per-incident cap on rail passenger transportation liability, periodically adjusted for the Consumer Price Index, which compresses upper-layer pricing into a narrower distribution than primary loss data would otherwise suggest.

Together, these dynamics explain why transport pricing requires methods that span the attritional and catastrophe layers, with credibility weights varying by risk type, fleet size, and tower position.

Exposure measures unique to transport

Standard commercial lines map cleanly to a single exposure base. Property uses TIV because loss scales with asset value. General liability uses payroll or revenue because hazard scales with operational intensity at fixed premises. Transport breaks both assumptions: the risk moves, and the insured asset's value is a poor proxy for the loss it can cause to third parties.

Rail operator liability is typically rated on turnover or revenue, with per-passenger sub-limits stacked above a primary layer. Logistics liability follows similar logic: revenue or turnover, rather than payroll, because asset-light operators with automated workflows have headcounts that materially understate exposure. Throughput proxies both frequency and severity better than headcount in those operations.

Underneath the top-line exposure base, long-haul carriers commonly carry secondary rating proxies. Mileage approximates frequency exposure. Gross vehicle weight and payload capacity approximate severity exposure. Cost of hire captures subcontracted and owner-operator capacity that vehicle counts alone would miss. The structural point is that a single transport policy typically requires multiple exposure proxies operating in parallel, a design choice that other commercial lines rarely require.

Rating factors that shape transport premiums

Rail: consist composition and operating territory

Rail rating concentrates predictive power in a small number of variables relative to other lines. Train length, number of cars derailed in historical incidents, signalization and method of operation, and FRA track class are recognized in industry analysis as primary drivers of derailment severity, though specific multivariate weights are sensitive to dataset and methodology.

Commodity type amplifies severity independently of car count. Toxic inhalation hazard cargo, pressure-liquefied gases, and other hazardous classifications carry severity loadings that the same loaded-car count would not predict for non-hazardous freight. Positive train control implementation is recognized in the market as a credit factor, but PTC's effect remains lagged in historical loss data, creating a pricing dynamic where prospective credit decisions outrun retrospective experience.

Logistics: fleet characteristics, regulatory compliance, and activity type

Logistics liability rating draws on a wider set of factors. Activity sector, axle configuration, fuel type, and fleet size all show statistically significant relativities in published GLM analyses, though specific coefficient magnitudes vary by dataset and study. The structural point that holds across studies is that univariate one-way analyses overstate the effect of any single factor relative to multivariate adjustment, which is a meaningful argument against pricing transport risks on uncontrolled rate tables.

Regulatory compliance variables function as a separate axis. FMCSA Compliance, Safety, Accountability (CSA) BASIC scores correlate with crash rates, with the directional relationship widely accepted, though the precise multiplier depends on the threshold and methodology used.

Factors that have shifted from modifier to prerequisite

A subset of transport rating variables no longer operates as a modifier on the rate algorithm. They function as binary eligibility gates instead, and a carrier failing one of them receives declination rather than a surcharge.

For trucking, filed FMCSA operating authority, an MCS-90 endorsement where applicable, and the absence of an Unsatisfactory safety rating function as binary screens. For UK-based rail operators, ORR licences and ROGS safety certificates or authorisations are statutory requirements, and applicants must demonstrate compliance with Safety Management System requirements set out in the relevant retained regulation and ERA guidance. Distinguishing where a factor switches regimes from continuous modifier to binary prerequisite is an underwriting judgment call rather than a mechanical rule.

How actuaries price with thin data and heavy tails

Transport pricing is rarely solved by a single technique. The line requires methods that integrate across the risk tower, because no individual approach captures both the attritional layer and the catastrophe layer.

Experience rating with shock loss separation handles the attritional layer well, with attritional losses priced over a five to ten year window and shock losses extracted and loaded over a longer 15 to 25 year horizon to prevent a single event from dominating trend selection. Exposure rating with increased limit factors anchors the upper layers and is often the only credible technique for immature accounts and high excess layers, where loss history is too thin to support direct experience rating. Multifactor credibility methods blend lower-layer experience into upper-layer pricing through covariance structure rather than the mechanical bilateral blend of experience and exposure that simpler approaches use.

Bayesian methods supplement thin transport data with informed priors from analogous lines and formally propagate uncertainty into pricing ranges. GLMs with regularization shrink sparse-cell parameters toward the class mean, mathematically equivalent to credibility weighting and useful for hazmat, intermodal, and last-mile sub-segments where individual cells lack standalone credibility. Bornhuetter-Ferguson is typically required for the most recent three to five accident years, where long tails leave chain-ladder estimates dominated by early random large losses. ASOP No. 39 addresses the identification and treatment of catastrophe losses in property/casualty ratemaking, and separating catastrophe and shock losses from attritional experience is important so that the parameters of other methods are not distorted.

What is shaping transport pricing now

Severity is deteriorating while frequency is improving in several transport sub-segments, a divergence that punishes single-factor trend selections. The U.S. corporate liability environment is the largest single driver. Marathon Strategies recorded 135 nuclear verdicts in 2024 with total awards of $31.3 billion, a 116% year-over-year increase in total value, and identifies commercial trucking among the sectors most affected. Swiss Re's Social Inflation Index provides a defensible severity trend floor for U.S. liability lines, attributing a meaningful share of recent claims growth to non-economic factors.

Rail is exposed to the same severity environment with thinner data. The cumulative cost of the East Palestine derailment, as reported in Norfolk Southern's 10-K, illustrates how a single rail event can produce more than $1 billion of incident-related costs across multiple accident years. Statutory caps such as 49 USC 28103 constrain pricing at the top of the rail passenger tower and reshape how excess capacity is structured.

How hx supports transport insurance pricing

Transport's structural challenges, sparse data, mobile risk, and a litigation-driven severity environment, require pricing infrastructure that handles complexity without forcing actuaries into spreadsheet workarounds. The hx platform addresses each of these constraints directly.

Configurable pricing logic for complex rating structures

Transport pricing logic combines knockout criteria, multiple parallel exposure bases, commodity-specific severity treatments, and regulatory compliance lookups. The hx Decision Engine lets actuaries express these structures in native Python, with knockouts, control interactions, and coverage-specific calculations deployed under full version control and governance, replacing hard-coded rate tables with auditable pricing code.

Submission triage aligned to appetite

Transport submissions arrive with documentation that determines both insurability and pricing tier, including operating authority, safety ratings, fleet schedules, and commodity profiles. hx Submission Triage extracts this data from unstructured broker submissions and surfaces it alongside appetite checks and indicative pricing, so underwriters can identify gaps and route risks for review before investing time in full analysis.

Portfolio intelligence for aggregation management

Transport's bimodal loss profile, combining higher-frequency cargo and trucking liability events with lower-frequency catastrophic rail events, requires portfolio-level visibility that policy-by-policy pricing cannot provide. hx Portfolio Intelligence supports batch rating, what-if analysis, and aggregation monitoring across exposure dimensions such as commodity type, territory, and fleet characteristics, enabling renewal-book scenario testing under shifting limit and severity assumptions.

Audit trails for evolving regulatory requirements

Transport pricing involves separating attritional from shock loss experience, applying different credibility weights across layers, and adjusting severity curve parameters as social inflation evidence develops. hx captures every action automatically, maintaining lineage from model assumption to individual policy decision, version control over rating logic, and the documentation trail that regulatory rate filings and reinsurance negotiations require.

Explore hx for transport insurance.

Frequently asked questions

How is transport insurance different from commercial auto insurance?

Commercial auto is a sub-component of the broader transport insurance landscape, focused primarily on vehicles operated for business use. Transport insurance encompasses commercial auto along with rail, marine cargo, inland marine, and logistics liability, each with distinct exposure bases and rating mechanics. Transport pricing typically requires multiple parallel exposure proxies and methods that span both attritional and catastrophe layers, a complexity that standard commercial auto rating rarely encounters.

Why does cost of hire matter for trucking pricing?

Cost of hire captures subcontracted and owner-operator capacity that owned-vehicle counts miss. A motor carrier that handles peak volume through hired capacity has exposure that does not appear on its fleet schedule, but the loss potential is real and falls within the policy. Cost of hire functions as a severity-and-frequency adjustment that brings rated exposure closer to actual operational footprint.

What does it mean to separate shock losses from attritional experience?

Attritional losses are the day-to-day claims that accumulate predictably across a portfolio. Shock losses are infrequent, very large events that distort trend if mixed into the attritional dataset. Separating the two allows actuaries to project attritional experience over a shorter window and to load shock loss expectations from a longer horizon, without letting a single event dominate the rate.

Why is positive train control treated as a prospective credit?

Positive train control reduces the probability of certain accident types, but its full effect has not yet emerged in historical loss data because the technology was implemented relatively recently across most U.S. mainline track. Crediting PTC retrospectively understates its risk-mitigation effect, which is why some underwriters extend prospective credit ahead of fully developed experience.

How do liability caps affect rail insurance pricing?

Statutory caps such as the one in 49 USC 28103 set a ceiling on aggregate per-incident liability for rail passenger transportation, adjusted periodically for inflation. The cap compresses the upper tail of the loss distribution into a narrower range, which changes how reinsurance layers attach and how excess capacity is structured. Pricing models that ignore the cap will overstate severity at high attachment points.

What is the practical difference between experience rating and exposure rating in transport?

Experience rating prices a risk based on its own historical loss record. Exposure rating prices a risk by applying industry severity curves and increased limit factors to its current exposure base, regardless of its own loss history. Experience rating dominates the lower layers of mature accounts. Exposure rating dominates the upper layers of any account and the entire pricing problem for accounts with insufficient history.

Explore hx for Transport insurance →

This guide is part of Hyperexponential's insurance pricing resource library. For more information on how hx supports Transport pricing, contact us.