Splitting the Baby: Apportioning Environmental Liability Among Triggered Insurance Policies

Abstract

CERCLA was intended to solve the nation's hazardous waste disposal problems. Not only would the law clean up hazardous waste sites, but, through retroactive strict liability, it would also make polluters pay the costs associated with this cleanup. Faced with staggering cleanup costs, Potentially Responsible Parties ("PRPs") sought to reduce their liability by invoking their Comprehensive General Liability ("CGL") insurance policies. PRPs often claimed coverage under multiple insurance policies issued over the course of many years. The resulting litigation frequently involved scores of insurance policies covering numerous industrial sites for the better part of the century. A policyholder might have purchased intricate layers of insurance coverage within each policy period. As an added complexity, there were often gaps in insurance coverage. With no statutory or contractual direction, common law courts had to fill the gaps and chart the interplay between state insurance law and federal environmental law. Unfortunately, the jurisdictions confronting this question responded fitfully and unsystematically. While nominally applying the same principles of insurance contract interpretation, courts reached disparate, if not downright contradictory, results. Further amplifying the confusion, courts almost uniformly failed to articulate the reasoning driving a particular allocation remedy, making it virtually impossible to project future outcomes. PRPs and insurers are left without predictable guidelines: facing enormous but uncertain liabilities. This article examines several core questions in the allocation process. When is insurance triggered to cover a loss? How should responsibility for covering a loss be divvied up? Should the policyholder be assigned any share of the loss? If so, under what circumstances — any time there is a gap in coverage, or only when that gap is rooted in a decision to self-insure? These questions implicate fundamental assumptions of existing insurance jurisprudence, particularly assumptions about the nature of the contract between the parties. The answers necessarily reflect policy choices about fairness and about the purpose of insurance. I contend that the existing law can be resolved into two fundamentally different allocation models that all courts have implicitly employed. Taking a holistic approach, this article seeks to reconcile, to the extent possible, these competing models, and proposes a new allocation system that maximizes the identified goals of the allocation process, while minimizing any disadvantages.