What determines price for environmental insurance?

Environmental insurance — spanning pollution legal liability (PLL) and contractors pollution liability (CPL) — breaks the fundamental assumption underlying most commercial lines ratemaking: that expected losses scale proportionally with a measurable unit of business activity. Pollution risk is a property of the site or the work type, not a function of revenue, payroll, or headcount. A dormant industrial parcel with zero employees and zero revenue can carry more contamination liability than an active facility generating millions in receipts. This structural reality forces non-standard exposure bases, sub-linear factor structures, and a pricing process that functions more as individual risk underwriting than algorithmic ratemaking. Here's what shapes environmental premiums differently from anything else in commercial casualty.

• CAS literature explicitly documents the primary exposure base of number of locations (for PLL) as "less than linear," while comparable CAS documentation for revenue (for CPL) was not identified in the available evidence.

• Groundwater involvement is the single most rigorously validated binary rating variable in public actuarial literature for this line — it functions as a severity threshold, not a gradient.

• The known loss / fortuity doctrine generally treats losses the insured already knows about or that are substantially certain to occur as ordinarily uninsurable across lines of insurance, including liability and property coverage, rather than creating a special contract-law gate unique to undisclosed pre-existing contamination.

• CPL claim frequency jumped 60% in two years (1 in 21 contractors in 2014/15 to 1 in 16 in 2016/17), with mold and microbial matter driving 29% of claims.

• PFAS MDL cases grew from over 6,000 in April 2024 to more than 11,000 in 2025, and some industry reports have compared PFAS liability risk to asbestos as a major emerging exposure.

Exposure measures unique to environmental insurance

Standard GL exposure bases — payroll, receipts, square footage — assume more business activity generates more loss-producing events. Environmental contamination exists independently of current operations, breaking that causal link entirely.

PLL is often rated in part using square footage and/or the number of insured locations, along with the types of activities performed at those locations. CPL uses revenue, also sub-linear, because larger contractors typically maintain more sophisticated environmental compliance programmes, reducing expected loss per revenue dollar at higher tiers. Underground storage tank (UST) policies often use tank count, where each tank is a discrete release point with independent age, construction type, and leak detection characteristics, and the EPA reports 581,676 confirmed UST releases as of September 2025. Cost cap and remediation cost overrun policies use the remediation cost estimate itself as the exposure base, eliminating the proxy problem entirely, but requiring detailed site-specific environmental assessment and remedial planning that can make the product uneconomic for smaller cleanups, with some insurer guidelines setting minimum projected expenses around $2 million.

Rating factors that shape environmental premiums

Site characteristics and contamination history (PLL)

Property use classification is the primary PLL segmentation variable, distinguishing commercial, habitational, redevelopment, and industrial risk classes. But NAICS/SIC codes alone have limited predictive value — CAS literature explicitly warns that average cleanup costs by industry code are unreliable when thousands of sites lack cost data altogether. What matters more is the site-specific contamination profile: known conditions, regulatory status, contaminant types, and whether a "no further action" letter truly means clean (it often doesn't — re-opener risk persists when new contaminants are identified or standards tighten).

Groundwater involvement can materially affect environmental loss severity. The 1994 CAS Forum paper uses it as a binary model input because once contamination reaches groundwater, remediation shifts to pump-and-treat regimes with multi-decade timelines, expanded plume delineation costs, and potential drinking water impacts triggering third-party claims. It's a threshold effect — the actuarial significance sits in the categorical distinction, not a continuous depth gradient.

Storage tank characteristics

Tank age, wall construction (single-wall vs. double-wall with interstitial monitoring), cathodic protection status, leak detection method, and stored product type each independently drive release probability. California regulations distinguish between single-wall and double-wall systems, reflecting a policy view that construction type matters for leak prevention. Installation era matters: pre-1988 tanks without secondary containment carry materially higher frequency expectations.

Contractor operations and trade classification (CPL)

The primary CPL question is "what do you do?" IRMI organises contractor operations into explicit hazard tiers. High hazard — residential/habitational work and similar exposures — often requires minimum $1M CPL limits with five-year completed operations tails. Medium hazard — general and trade contractors — requires $1M minimums with two-year tails. These classifications have become more important in underwriting, and project owners may require CPL as a contract condition, especially for projects involving sensitive environmental exposures or hazardous materials.

Naturally occurring hazardous substances (silica, arsenic, pyrite, asbestos) create a geographic/geological rating layer that operates independently of the contractor's own chemical handling. Excavation contractors in high-NOH geology face materially higher loss frequency than the same trade classification in low-NOH areas — a multiplicative interaction standard class plans don't capture.

Binary gates versus continuous adjustments

Environmental insurance relies on knockout criteria grounded in contract law, not just underwriting preference. Known contamination may implicate the fortuity or known-loss doctrines, and a material misrepresentation or nondisclosure can make the policy voidable from inception. Active regulatory enforcement orders, NPL listing with ongoing CERCLA action, documented intentional non-compliance, and abandoned property status all function as hard declination gates. This is structurally different from GL, where a manufacturer with a documented defect history is written at a loaded premium rather than declined.

Continuous pricing variables include claims history, risk management quality, coverage structure (limits, retentions, retroactive date), Phase I/II assessment results, optional coverage parts (NODS, transportation, business interruption), and policy term. Term factors are explicitly sub-linear — fixed underwriting costs for environmental due diligence are amortised across the policy period, and longer terms expose the insurer to regulatory ratcheting risk as cleanup standards evolve.

Emerging contaminant surcharges

PFAS is no longer an emerging risk — it's an active repricing event. Carriers are responding with outright PFAS exclusions and enhanced underwriting, including in some cases sublimits or other policy limitations, particularly for sites with historical AFFF use. The DuPont/Chemours/Corteva settlement with New Jersey DEP reached $2 billion; 3M settled separately for $450 million. Underwriters scrutinise former manufacturing facilities, waste disposal sites, airports, and military bases for potential PFAS exposure in environmental risk assessment.

How actuaries price with structural data scarcity

No single method addresses environmental pricing's combination of thin data, extreme severity, long tails, and regulatory-driven loss emergence. Extreme value theory (Peaks-Over-Thresholds with generalised Pareto distribution) suits high-limit severity estimation because it extrapolates from moderate exceedances rather than requiring direct observation of extreme events. Monte Carlo simulation with separated catastrophic and attritional loss processes addresses the irregular occurrence of pollution catastrophes that destabilises standard development triangles. Bayesian credibility blending is structurally necessary because both insurer-specific experience and industry benchmarks are unreliable — the Bayesian framework forces explicit quantification of prior uncertainty rather than treating a sparse benchmark as fixed truth. Benchmark loss ratio / exposure rating serves as the default for individual accounts where no credible loss history exists, which is most of the book. Bornhuetter-Ferguson is commonly used for early development years where claims are most immature, because it gives limited credibility to early observed loss experience and relies more on an a priori expectation. Expert judgment with regulatory scenario stress testing is a required component, not a fallback — forward-looking scenarios for PFAS MCL enforcement, emerging contaminant designations, and judicial coverage reinterpretation address risks absent from any historical dataset.

What's shaping environmental pricing now

The environmental insurance market remains underpenetrated, with fewer than 20% of buyers purchasing coverage. Severity is the dominant trend driver: median nuclear jury verdicts have risen substantially—from about $21 million in 2020 to $44 million in 2023—and PFAS settlement values are reaching billions. PFAS MDL case volume nearly doubled in fifteen months. Mold, indoor air quality, and microbial matter continue to be important CPL exposure areas, with industry sources noting rising claims related to mold, fungi, bacteria, and Legionella. AM Best maintains its industry ultimate net environmental loss estimate at $46 billion, primarily reflecting legacy exposure, while environmental insurance books may face additional development uncertainty. Microplastics litigation is emerging — one Clean Water Act settlement exceeded $50 million — and climate-related litigation cases have more than doubled since 2015. Carriers are protecting loss ratios by restricting higher limits, tightening PFAS terms, and intensifying underwriting scrutiny on redevelopment sites where regulators are revisiting previously remediated conditions.

How hx supports Environmental insurance pricing

Configurable pricing logic for complex rating structures

Environmental's unique challenges require pricing logic that standard raters struggle to express. The hx Decision Engine lets actuaries implement these rules in native Python—including knockout criteria, coverage-specific calculations, and control interactions—then deploy changes with full governance and version control.

Environmental pricing may draw on a range of actuarial methods, depending on the risk and available data. The hx Decision Engine implements these multi-method frameworks in native Python with full mathematical transparency.

Submission triage aligned to appetite

Environmental submissions arrive with documentation that determines both insurability and pricing tier. hx Submission Triage extracts this data from unstructured broker submissions and surfaces it alongside appetite checks and indicative pricing, so underwriters can identify gaps before investing time in full analysis.

PLL insurability can turn on case-specific coverage issues that may be difficult to encode in standard workflow tools. hx Submission Triage routes based on Phase I/II assessment results, NPL listing status, and PFAS exposure flags before pricing begins.

Portfolio intelligence for aggregation management

Environmental's systemic risk requires portfolio-level visibility that policy-by-policy pricing can't provide. hx Portfolio Intelligence enables batch rating, what-if analysis, and concentration monitoring to support regulatory reporting requirements.

Environmental loss development is non-stationary—PFAS MDL cases grew from over 5,600 in August 2023 to more than 11,000 by July 2025, and evolving EPA PFAS drinking-water standards and state mini-Superfund laws may limit the reliability of historical LDFs for forward projections. hx Portfolio Intelligence stress-tests reserves against emerging contaminant scenarios and regulatory trigger changes across the entire book.

Audit trails for evolving regulatory requirements

With increasing regulatory scrutiny, actuaries need documented lineage from model assumptions to individual policy pricing decisions. hx captures every action automatically, creating the governance trail Environmental's regulatory environment demands.

Sub-linear factor structures for revenue and location count are explicitly documented in CAS literature, but applying them requires actuarial judgment on credibility weights and prior distributions that must survive audit. hx captures the rationale for Z-factor selection, EVT threshold calibration, and benchmark loss ratio adjustments with full version control.

Explore hx for Environmental insurance →

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