Environmental Liability

February 1, 2005

Environmental Liability



Jonty T. Bogardus

February 2005




“Environmental Liability” is a significant concern that now pervades all corners of the economy. Historically only a (minor) concern of industrial organizations working with products obviously hazardous if released into the ecosystem, today’s highly regulated society requires all to be mindful of the consequences of their interactions with the environment. From major agricultural, resource extraction and manufacturing companies to homeowners wondering what to do with that extra half-can of paint, the potential for exposure is all pervasive.

Help may be at hand in the form of insurance coverage. Commercial entities focussed on activities that have the well-understood potential to cause environmental harm can purchase specialized environmental liability policies. For the rest of us, the ubiquitous commercial general liability (“CGL”) policy can, in certain circumstances, afford some degree of coverage. Accordingly, this paper will focus on the extent CGL policies provide coverage for environmental liabilities. While CGL policies will be specifically referred to throughout, much of the discussion is nonetheless applicable to other types of insurance, such as professional liability and D&O policies.

Subject to specific exclusions, CGL policies generally state that the insurer:

will pay those sums that the insured becomes legally obligated to pay as compensatory damages because of “bodily injury” or “property damage” to which this insurance applies.

In considering whether coverage exists for any particular environmental loss an insured will have to consider whether the loss constitutes “compensatory damages” that the insured is “legally obligated to pay” and, in the case of property damage whether the insurance covers the damaged property. As will become clear throughout the course of this paper insurance will not cover all costs incurred in remedying environmental problems, however it is always worthwhile to consider obtaining coverage for environmental liabilities and referring to existing policies when an environmental problem arises.

CGL policies, along with most everything else, have evolved over the past few decades. New CGL policies often contain “absolute” environmental liability exclusion clauses.

However, just because a loss has an “environmental” component does not mean that it is will be “absolutely” excluded from coverage under these newer policies. Older policies contain a more limited exclusion that excepts liability for occurrences that are “sudden and accidental”. Still older policies provided coverage for any “occurrence” that caused unintended bodily injury or property damage regardless of any environmental component.

The following sections set out a number of matters relevant to CGL coverage for environmental matters including the evolution of the CGL environmental exclusion, a number of considerations to take into account in considering what is and is not covered and the extent to which coverage can be engaged in the event of regulatory charges. To begin, however, a run-through of the sources of environmental liability is necessary as the type of claim and the relief sought may affect whether coverage is owing.


There are three primary sources of environmental liability: statutory, tort (e.g., negligence, nuisance) and contractual. Each raises different coverage issues, generally related to the questions of whether the liability takes the form of “compensatory damage” and whether the insured is “legally obligated to pay” the damages. Accordingly, a brief discussion of the differing aspects of these three sources of liability is warranted.


Virtually any business activity that an insured may be involved in is governed by a federal or provincial statute relating to the environment. Obvious examples include pulp mills, chemical manufacturing and hazardous waste disposal. Ironically, it may be those businesses that do not have any obvious environmental concerns that are most at risk. This is because managers of such businesses may not be aware of legislation that has the potential to result in liability for them. For example, a condominium developer (a business which causes no apparent environmental damage) may have built adjacent to contaminated land. If the contamination migrates to the developer’s property (for example, by way of groundwater), the resulting liability may be significant. If the owner of the adjacent contaminated property is insolvent, the developer, and its insurer, may be left “holding the bag” for substantial cleanup costs that it neither knew of nor caused.

What follows is a summary of some of the federal and provincial statutory provisions which impose environmental liability.

a.      Federal

Two of the most relevant federal statutes are the Fisheries Act and the Canadian Environmental Protection Act. The Fisheries Act is broadly worded and has been applied by Canadian courts in a number of circumstances. Generally speaking, it applies to all acts that affect “fish habitat”. The Act defines “fish habitat” as: “spawning grounds and nursery, rearing, food supply and migration areas on which fish depend directly or indirectly in order to carry out their life processes”. In other words, nearly every body of water that contains fish may be subject to the provisions of the Fisheries Act.

The Act prohibits persons from carrying on or undertaking any work that results in the harmful alteration, disruption or destruction of fish habitat. Furthermore, the deposit of any type of deleterious substance in water frequented by fish or in any place where it may ultimately find its way to water frequented by fish is prohibited. Therefore, the Act may apply even if a spill does not appear to be in the vicinity of fish habitat. As long as the substance ultimately finds its way to a fish habitat (for example, if it enters into a drainage system leading to a river), liability under the Fisheries Act may be incurred. The fact that a deleterious substance may be significantly diluted once it is added to water is irrelevant.

The Act expressly preserves all civil remedies available at common law. In addition, the Act provides a statutory right to damages for the government and commercial fisherman in cases where there has been a loss due to a deposit of a deleterious substance in water frequented by fish. Those who may incur liability include persons who owned or had control of the deleterious substance, as well as those who caused or contributed to the deposit of a deleterious substance in water frequented by fish. Those who may incur liability include persons who owned or had control of the deleterious substance, as well as those who caused or contributed to the deposit of the deleterious substance. Damages may be claimed for all costs and expenses reasonably incurred to prevent the deposit or to counteract, mitigate or remedy any adverse defects that have resulted therefrom.

The Canadian Environmental Protection Act (“CEPA”) is comprehensive federal legislation intended to protect the environment. Among other things, CEPA deals with the regulation of toxic substances and provides for interim orders to be made on an emergency basis to protect the environment. The Act also deals with such matters as spill reporting, export and import of toxic substances and waste materials, international pollution and ocean dumping. Offenses are created for violations of the Act. Section 136 permits a civil cause of action for damages or an injunction as a result of conduct in contravention of CEPA.

b.      Provincial

Each Canadian province has its own environmental protection legislation. Although similarities do exist between such legislation, they are not identical, particularly with respect to the imposition of civil liability for environmental damage.

There are a number of B.C. Statutes that regulate the use of the environment. They include the Waste Management Act, the Water Act, the Forest Act, the Environment Management Act, the Transportation of Dangerous Goods Act, the Land Act, the Litter Act, and the Health Act. The basis for liability under each act varies. For example, the Health Act provides for the recovery of “all reasonable costs and expenses incurred in terminating a health hazard or unsanitary condition”…from any “person whose act, default or sufferance, the condition or health hazard was caused, or who was responsible for the health hazard or condition…”. The Environment Management Act permits recovery of costs associated with the cleanup of an environmental emergency “from the person whose act or neglect caused or who authorized the events that caused the environmental emergency in proportions the Court determines”.

The Waste Management Act is the central environmental protection statute in B.C. Under it direct liability may be imposed for pollution cleanup costs upon an order being issued by government officials. Liability is based not only on fault and culpability but also upon the relationship between a person and the polluted or polluting property.

Section 22 of the Act empowers an official, the Regional Waste Manager, when he is satisfied on reasonable grounds that a substance is causing pollution, to order remedial work against:

(a)      The person who had possession, charge or control of the substance at the time it escaped or was emitted,            spilled, dumped, discharged, abandoned or introduced into the environment;

(b)      Any other person who caused or authorized the pollution; or

(c)      The person who owns or occupies the land on which the substance is located or on which the substance            was located immediately before it escaped or was emitted, spilled, dumped, discharged, abandoned or            introduced into the environment.

Since liability for cleanup may now be based on a party’s relationship to land, persons as landlords or mortgagees in possession may find themselves having to clean up contamination caused by a lessee or mortgagor. This provision makes the business of lessors, lenders, and purchasers of property particularly difficult since no specific criteria, standards, or requirements have been set with respect to when the Regional Waste Manager may make such an order.

The Act also imposes criminal liability on officers, directors or employees of corporations that commit offenses under environmental legislation. To obtain a conviction it must be found that the accused authorized, permitted or acquiesced in the offence. Under Section 34.1 of the Act, a person convicted of a Waste Management Act offence may be required to pay a fine equal to the Court’s estimation of the amount of “monetary benefits (that) accrued to the person”. In some cases, profits realized from polluting activities, and hence the fine payable, may be enormous.

The above provisions are worded broadly and may be applied to recover cleanup costs from officers, directors and perhaps even shareholders of corporations whose acts fall within their ambit.

In each case, liability will turn on the facts, and in particular, the knowledge, means of knowledge, degree of control, ability to influence control and the action or inaction which the person in question took in the particular case.

Finally, the Waste Management Act also provides for civil recovery of “remediation costs” from any “person responsible” for the environmental harm. This aspect of the Waste Management Act has been relied upon significantly in the years since it came into effect to permit recovery of costs and allocation of liability between those that originally caused and are currently responsible for the environmental harm. Section 27(2) lists the various recoverable “costs of remediation”. This section has yet to be interpreted in the insurance context. The question, which will be touched on in greater detail below, is whether all “costs of remediation” (along with other forms of environmental liability imposed by statute) are “compensatory damage”. Section 27(2) states that “costs of remediation” includes not only the actual remediation costs but also:

(a)      costs of preparing a site profile,

(b)      costs of carrying out a site investigation and preparing a report, whether or not there has been a            determination under section 26.4 as to whether or not the site is a contaminated site,

(c)      legal and consultant costs associated with seeking contributions from other responsible persons, and

(d)      fees imposed by a manager, a municipality, an approving officer, a division head or a district inspector            under this Part.

2.   TORT

Tort liability encompasses those wrongs that are not otherwise covered by codified statutory law (both criminal and civil) or arise out of contractual obligations. Torts give rise to a private right of action brought by one person against another in order to remedy, usually but not always by way of monetary award, an alleged wrong. Torts that can provide redress for environmental wrongs include negligence, nuisance, “the rule in Rylands v. Fletcher” and trespass.

For the purposes of this paper it is not necessary to explain the conceptual differences between each tort. Each of the above torts could be successfully applied, for example, to a case of the transfer of a hazardous substance from one person’s land to another’s, where harm results. The legal justification for alleging one tort over another often rests with the remedy being sought – and the choice of remedies is often important to determining the insurance coverage that might be available.

Remedies in tort include:

(a)      general damages;
(b)      aggravated damages;
(c)      punitive damages; and
(d)      injunctions.

General damages are wholly compensatory in the sense that one is being directly compensated for the actual quantum of the loss, but no more. They serve to “make one whole”, as if the tortious conduct had never occurred.

Aggravated damages are applied where the wrongdoer’s conduct was so malicious or outrageous that an additional award is justified to “compensate” for the plaintiff’s hurt feelings. Aggravated damages are deemed to be awarded to punish the wrongdoer as opposed to compensate the plaintiff. However, such damages are considered in law to be compensatory.

Conversely, punitive damages are not “compensatory” in any sense. They are also levied to “punish” the wrongdoer but do not correspond to any loss of the plaintiff for which compensation may be owing. Punitive damages relate solely to the egregious conduct of the wrongdoer and are intended to act as a deterrent against future bad conduct.

An injunction requires a party to a lawsuit to act or restrain from acting in a particular way. In the environmental context (though not the pollution context) injunctions are often granted in British Columbia to restrain environmental activists from preventing lawful logging or mining activities. There is no direct “damages” component to an injunction, though generally the harm sought to be alleviated by the injunction has already caused a loss or would in the future if the injunction requiring or restraining conduct were not in place.


Liability in contract results when a party to the contract refuses to perform on their contractual obligations. Remedies for breach of contract include damages, specific performance (i.e. requiring that the contract be performed upon) and rectification (e.g., fixing the problem caused by the breach of contract).

For the purposes of this paper, of interest are the situations where a party may be required by contract (usually by virtue of an indemnity clause) to assume responsibility for an environmental liability. The most common such case is in the context of landlord-tenant relations.

Lease documents often contain terms requiring the tenant to indemnify the landlord for any loss or expense occasioned by the tenant’s conduct. As noted above, under the Waste Management Act a property owner may be liable for the “costs of remediation” even where the owner did not cause the pollution to be remediated. Where an owner pays for the remediation and then seeks indemnification from its tenant, the question of interest here is whether the tenant’s insurer must respond to the claim made under the indemnity clause. Is the claim one for “compensatory damages” that the tenant is “legally obligated to pay”? While the answer is likely “yes”, insurers may take the matter to court once confronted with this question. However, the question may never arise as in practice all “potentially responsible parties”, including the tenant in the example above, will be named as defendants to a cost recovery action.


Now that the sources of environmental liability have been addressed we can move to consider under what circumstances an insurance policy will respond to defend or indemnify an insured faced with such a claim.

However, before discussing the specific ambit of any insurance policy, it is important to ensure a basic understanding of how insurance coverage under a particular policy is engaged. In particular, when discussing environmental issues it is important to determine whether the potentially applicable policy is an “occurrence-based” or “claims-made” policy. The distinction is important both when selecting a new policy and in considering which, if any, existing policy may respond to an environmental liability claim.

Insurance policies are time-limited. A typical insurance policy provides coverage for a one-year period, with rights of renewal occasionally extending the application of a specific policy over a period of years. Coverage is provided for the term of the policy. The question that arises is, when an event leading to liability occurs at one time and a claim is made against the insured party sometime later, and a new policy has incepted in the interim, which policy responds. Is it the policy in force at the time of the “occurrence” leading to liability or the policy in force at the time the “claim” is made and reported to the insurer?

The answer depends on the wording of the relevant policy. No type of policy uses exclusively one form of wording to determine the application of coverage. However, professional liability policies are more commonly claims-made policies and CGL’s are often occurrence-based policies. Note: since 1997 “claims-made” policies have added a requirement that both the claim be made and that it be reported to the insurer within the term of the policy.

The distinction between occurrence-based and claims-made policies is important when it comes to matters of environmental liability for, in particular, historic contamination. Claims are often made for property damage arising from environmental contamination that occurred years or even decades ago. As discussed below, many current policies exclude coverage for environmental contamination. However, many historic policies did not contain any such exclusion and if the historic policy was occurrence-based, such that the “occurrence” (e.g., spill) took place during the tenure of the policy (and “property damage” resulted at that time), a claim can still be advanced under the policy even though the term of the policy may have long since expired. Conversely, an old claims-made policy cannot be relied upon unless the insurer was notified of the potential for liability during the term of the policy (an unlikely event in the context of historic environmental losses).


Occurrence-based policies, as discussed above, respond to “occurrences” that took place during the tenure of the policy. However, is the “occurrence” the initial exposure to the harm or the manifestation of the harm? For example, the ubiquitous asbestos litigation of the 1990’s dealt with exposure to asbestos, whether on one occasion or over the course of many years, that only (allegedly) resulted in “injury” long after the initial exposure. Which policy responds? The one that was in force when the initial exposure occurred? The one in force when the greatest exposure occurred? The one in force when the injury first manifested?

The courts have set out a number of principles to guide in the decision as to which policy is “triggered”. In historic environmental contamination cases the theory that is most applicable, and is coming increasingly into vogue, is the “continuous trigger” or “triple-trigger” theory. Under this theory all policies from the time of exposure through to the manifestation of injury or damage are triggered and must respond.

The application of the continuous trigger theory can be very helpful to insureds where one of two conditions apply: (1) the insured did not have or cannot find evidence of insurance for the entire period, and (2) the insured’s policies do not all have the same limits and some do not cover the value of the loss. In the first scenario, a duty to defend can still be compelled of the insurer(s) the policyholder knows of or can demonstrate was on risk during the period of exposure even if there are no policies for some or most of the remaining years.

With respect to the second scenario, if the “occurrence” was reduced to a defined date, and the policy in force at the time of the occurrence had insufficient limits to cover the loss, the insured would be liable for the remainder. Under the continuous trigger theory the insured can pick the policy in place during the duration of the exposure that has the highest limits, thereby maximizing coverage. However, under this latter scenario it must be realized that only one policy, and not all of the potentially applicable policies, can be triggered; hence, no “stacking” of policies. The insurer picked to provide the indemnity can then seek equitable contribution from the other “triggered” insurers on a pro rata basis.

An interesting consequence of the ability of insurers to seek equitable contribution where the loss transpired over the course of a number of years can be seen in the case of Surrey (District) v. General Accident Assurance Co. Surrey constructed a flume to carry water over the Peace Portal golf course. Over the years the concrete flume deteriorated. After litigation between Peace Portal and Surrey resulted in Surrey being found liable for the cost of repairing the flume it sought coverage from its onetime insurer.

The degradation of the flume occurred over a number of years. Surrey could only produce policies for approximately one third of these years. Surrey sought recovery of the full amount from the insurer on the basis of the continuous trigger theory. However, the court found that while the insurer was obligated to grant an indemnity under this theory the insurer was nonetheless entitled to equitable contribution from other “insurers” on risk during the term of the loss. As Surrey could not identify other insurers the court considered it to be “self-insured” during this period and as a result only required the insurer to indemnify Surrey on a pro rata basis based on its years on risk.


Once one determines the applicable policy it is then necessary to review the specific wording of the policy to determine whether it will provide coverage for the specific environmental liability of concern. Fortunately, most CGL policies in effect at any one time are similarly worded; the insurance industry generally follows standardized language. Unfortunately, policy wordings particularly as they relate to coverage for environmental liability have evolved significantly over the past 40 years. The following provides a general run-through of this evolution, which will assist in informing the basis for and extent of the current “absolute” exclusion.

Before 1966 liability coverage was generally limited to claims for an “accident”. “Accident” was generally construed as requiring a “distinctive event that takes place by some unexpected happening at a date that can be fixed with reasonable certainty.” Increasing demand for “occurrence” coverage (for reasons irrelevant to our discussion) led to most policies being changed to provide coverage for any “occurrence”, meaning:

an accident, including continuous or repeated exposure to conditions, which results in bodily injury or property damage neither expected or intended from the standpoint of the insured.

What was not “expected or intended” by the insurers, however, was that this wording would open up claims for deliberate pollution where the resulting injury or damage was not “intended”. The policy wordings were amended again shortly thereafter. However, for those policies in effect from 1966 to 1973 (or thereabouts) the coverage can be quite broad and cover all manner of environmental losses regardless of the actions of the insured that led to the loss.

In 1973 a form of “qualified pollution exclusion” was incorporated into the body of standard CGL policies. Policies no longer applied:

to bodily injury or property damage arising out of the discharge, dispersal, release or escape of [pollutants] but this exclusion does not apply if such discharge … is sudden and accidental.

This change in wording focussed the coverage inquiry from the result of the event to the cause of the event. While, as discussed further below, most current policies contain an “absolute” environmental exclusion, there remain a few CGL policies still in existence that continue to utilize a qualified environmental exclusion which contemplates a limited cover for only those losses which are “sudden and accidental”. Accordingly, further discussion of the effect of this exclusion is warranted.

Judicial interpretation of such clauses in both Canada and the U.S. has sometimes favoured the insured by adopting a view that emphasizes the “accidental” aspect of such spills while minimizing the importance of the degree to which the spill has to be “sudden”. However, in the BP Canada Inc. v. Comco Service Station Construction & Maintenance Ltd., a decision of the Ontario Supreme Court, the plain meaning of the word “sudden” is given effect.

The facts of the BP Canada case are relatively straightforward. In 1970 the insured purchased a retail gasoline outlet which it continued to operate. In May, 1980, an underground gasoline leak was discovered. For the purposes of the motion before the Court, it was accepted that the gas had leaked from a cracked coupling in the storage system on the property, that the coupling had been defective from the time of its installation in 1967 and that the leak had been ongoing for a considerable although unspecified period of time. The soil and ground water of the insured and its neighbours became contaminated by the gasoline. BP was required to carry out the necessary cleanup.

In February 1983 BP commenced an action in damages against the insured in contract and in tort. The issue was whether, on the facts which were agreed for the purposes of the motion, the damage in question arose from a release or escape of gasoline which was “sudden and accidental”.

Can damage resulting from a gradual leak of pollutants ever be “sudden and accidental”? After pointing out that the onus is on the insured to bring itself within the exception to the exclusion, the court proceeded to review U.S. and Canadian case law that has considered such clauses. Early American law found that a gradual leak could be considered “sudden” as there was one distinct moment when the leak commenced. Later American cases found the opposite. The court sided with the more recent law in concluding that the plain meaning of the term “sudden and accidental” must include a temporal element and should not be extended to include unintended consequences which are not sudden, but occur over time.

The BP Canada case still applies, at least in Ontario, but the principle continues to be disputed in several American jurisdictions. However, it is possible to envision situations where the insurer might be required to provide coverage limited to damage caused by the initial, sudden leak but not for the incremental, gradual damage which the leak causes thereafter.

Regardless of the outcome of the BP Canada case, the facts of each case must still be carefully investigated to determine whether the exclusion applies. For example, an underground gasoline storage tank might be found to have a crack at a level that results in sudden, low volume spills of short duration each time the tank is filled. On such facts, an argument might be made that the damage was caused by a series of, “sudden and accidental” spills, notwithstanding that it may be years before the damage is discovered.

On the “accidental” side of the equation, it is clear that “accidental” means unintended and implies the intervention of a fortuitous cause. Once an insured has knowledge that environmental harm is occurring, and fails to take preventative action, the subsequent harm from that point forward will not be covered.

Finally, as mentioned above an “absolute pollution exclusion” is now a typical component of CGL policies. The absolute exclusion was put into effect in 1985 in reply to court interpretations of the “sudden and accidental” exclusion to permit coverage for gradual but unintentional pollution. The absolute exclusion generally provides that there is no coverage for:

bodily injury or property damage arising out of the actual, alleged or threatened discharge, dispersal, release or escape of pollutants …

The absolute exclusion has, for the most part, allowed insurers to avoid responsibility for “traditional” environmental damages. The question being raised most commonly today is whether the absolute exclusion extends so far as to limit coverage for non-traditional environmental liability, for example bodily injury caused by the escape of carbon monoxide “pollution” from an apartment building furnace. The answer, generally, is “no”; such harm is what was traditionally intended to be covered by liability policies and is not now excluded by the “environmental” policy exclusion. Recent Canadian cases considering the absolute exclusion are discussed toward the end of this paper.


Given the varying ways in which coverage can be engaged and the evolution of policy language over the last 40 years, it is not surprising that the courts have had to step in on occasion and resolve conflicts that have arisen between insurers and insureds. Of course, coverage is governed by the specific terms of the policy in question. However, as many policies have similar or identical wording one would expect, in theory, consistent approaches to be adopted. Unfortunately, neither the Canadian Courts or their American counterparts, who have had much more exposure to environmental liability cases, have managed to make consistent decisions regarding the scope of insurance coverage and the applicability of exclusion clauses.

In general, the interpretation of insurance contracts is a two-step process. First, the ordinary rules of interpreting contracts must be applied. Effect must be given to the intention of the parties as expressed from the words used in the policy. This usually means applying the literal meaning of the words in the contract. However, the literal meaning will be rejected where the result is either unrealistic or one which was not reasonably contemplated in the commercial atmosphere in which the insurance contract was created. The second step of interpretation applies when the policy contains an ambiguity. In such cases, the contra proferentem rule will be applied. This rule requires that ambiguities be resolved in favour of the insured. This rule is frequently applied in the case of exclusion clauses.

In cases dealing with environmental liability (as with any other claim under a CGL policy), the first question which must be answered is whether the damage is covered having resulted from either an “accident” (with respect to some CGL policies) or an “occurrence” (with respect to other CGL policies). Usually, these conditions are satisfied if the damage was neither expected nor intended from the standpoint of the insured. In practice, whenever the insured is engaged in a business that, arguably, would involve a reasonable expectation that pollution may occur, then the issue arises of whether the specific damage at issue was expected or intended.

The answer to this question is largely fact dependent, which perhaps explains the divergence of opinion in American case. It is generally the case that Canadian Courts interpret the meaning of “occurrence” or “accident” liberally. In Canadian Indemnity Company v. Walkem Machine Ltd., the Supreme Court of Canada held that an insured could take a calculated risk and if damage then resulted, it would be considered accidental. The Court’s reasoning would indicate that any occurrence resulting in loss is an accident unless it is done with intent to bring about loss or damage. The case dealt with a CGL policy that covered damage caused by “accident”. The same reasoning has since been applied to an occurrence based policy where coverage is predicated on damages being neither expected nor intended. The practical result is that a “calculated risk” amounts to no risk at all (for the insured), since the insurer is (most often) required to pay the damages that would result from the risk being realized.


Intentionally caused pollution will always be excluded from coverage. Just as with all other types of coverage, insurance does not cover damage, injury or loss that the insured intends to result from his actions. In the environmental context, what this means in practice is that a producer of hazardous waste cannot simply dump the waste on a neighbouring property (thereby minimizing its disposal costs) and then expect its insurer to cover the inevitable claim by the landowner to remediate the contamination.

However, there are circumstances where the discharge of pollutants in merely collateral to the activities of the insured. In such instances it is necessary to determine the intent of the insured.

For example, in one case hazardous material was properly treated and disposed of. Subsequently the waste was released from its containment through no fault of the insured. The court found that the release was not a “discharge of pollutants” by the insured and the release was “sudden and accidental” (under a non-absolute exclusion policy) such that the policy must respond.

In contrast, the owner of a uranium mill was aware that leachings from a holding pond were escaping into the environment. The owner argued, however, that it was unaware of the extent of the leaching and had thought that such minimal amounts were escaping as to not constitute “pollution”. The court found that while the harm may be “unexpected or unintended”, the knowledge of the leaching was what was important, rendering the escape “expected or intended”.


A further common exclusion in liability policies (as opposed to property insurance policies) is an exclusion for harm caused to one’s own property. This is important when considering environmental liability since as often as not pollution created by a person causes damage to their own property, as opposed to a neighbouring property or an adjacent water body. The simple answer is that a liability policy will not compensate an insured for the costs of remediating environmental damage to their own property. The exception to this conclusion, dealt with in more detail in Section XI below, is that government-ordered cleanup and remedial costs have been held to constitute damages as the insured is arguably “legally obligated to pay” such amounts within the terms of a typical CGL policy.


In Section II above there was a discussion of the various remedies that can result from statutory, tort or contractual claims. As is probably apparent by now a standard CGL policy will not provide coverage for each of these remedies.

For example, in tort it was noted that available remedies are general damages, aggravated damages, punitive damages and injunctions. In practice the vast majority of claims for punitive damages or an injunction are accompanied by a claim for general damages. Accordingly, if otherwise within policy coverage a duty to defend will be triggered. However, punitive damages if awarded are not “compensatory”. Accordingly, courts have generally found that no indemnity is owing to an insured for any award of punitive damages made against it.

Historically, CGL policies applied generally to “damages”. Insurers, when faced with the issue, argued that it was against public policy (and the intent of the insurance policy) to pay out on punitive damage awards. However, due to inconsistent treatment by the courts the change was made to specify coverage for “compensatory” damages in order to avoid paying out insureds for their “intentional, malicious conduct”.

As for injunctions, by their definition they do not require “payment” of “damages” and hence there is nothing for insurance to indemnify.

When considering environmental liability under contract it must be remembered that an insurance policy is only going to respond to “damage” to person or property. It will not respond to a business loss that did not arise out of “property damage” or “bodily injury”.

As for the situation alluded to above in Section II, where a tenant (or other person) owes an indemnity obligation it is akin to the landlord making a claim for “compensatory damages” the tenant is “legally obligated to pay”. As with insurance contracts, indemnity obligations are decided on the specific wording of the indemnity and broad statements in the absence of specific wordings are inappropriate.

However, and in conclusion, suffice to be said that courts have, in recent years at least, generally taken an expansive view of what constitute “damages”. If a monetary payment is required as a result of wrongdoing, then it will almost invariably be held to constitute “damages”. This issue is further addressed in Section XI when considering whether a regulatory fine constitutes damages.


Most insurance policies provide two components to coverage. First, there is an obligation on the insurer to defend any claim made against the insured that potentially comes within the scope of coverage offered by the policy. Second, if damages are awarded against the insured for liability that comes within the scope of coverage the insurer must indemnify the insured for the award of damages.

While a detailed discussion of the law in this area is far outside the scope of this paper, suffice to say that the duty to defend is much broader than the duty to indemnify. Accordingly, while for example the absolute pollution exclusion may appear to exclude coverage for certain liability, the insurer may still be obliged to defend the claim until a determination can be made as to the scope of the indemnity obligation. Also, where the primary relief sought against an insured is an injunction, which is not “compensatory damages” for which the policy will eventually owe and indemnity, generally there will be some form of additional “damages” claim that triggers the duty to defend. However, in a case where only an injunction or some other “non-damage” relief is sought there may be no duty to defend. The cases appearing in Section XII below highlight the extent to which a claim that may not strictly fall within coverage may still trigger the policy’s duty to defend obligation.


As discussed above, under many CGL policies the insurer agrees to pay on behalf of the insured all sums that the insured becomes legally obligated to pay as damages because of property damage caused by an “accident” or an “occurrence”. Property damage is generally defined to include both physical injury or destruction of tangible property or a loss of use of tangible property.

Government ordered cleanup costs are clearly legal obligations to pay. However, can they be said to be in respect of “property damage”? Do they amount to “damages”?

Should it make any difference whether cleanup costs are incurred before a formal order is made? Finally, should the insured be permitted to recover costs that extend beyond mere cleanup to the provision of works or devices which would prevent another pollution problem?

No doubt arguments may be made in support of both sides of these issues. CGL policies are basically designed to cover liability for damages incurred by third parties, not to pay monies which are required in respect of an insured’s own property.

On the other hand, in a situation where a government agency cleans up pollution and then seeks to recover it from a polluter or owner of property, its position may be analogous to that of the third party. The insured would argue that there is no real distinction between a government claiming reimbursement for cleanup costs and a damage claim for injury to government property. The only difference is that in the former case, the government incurs loss (in the form of cleanup costs) with respect to the insured’s property, while in the latter case the government’s loss is in respect of its own property. In both cases, the fact is that the government itself is incurring a cost that it later seeks to recover.

The Canadian cases on this point are old but apparently still applicable. In two British Columbia Supreme Court cases government-ordered cleanup and remediation costs incurred by third parties were held to constitute “damages” by reason of the “liability imposed by law” term of the relevant insurance policies. The American cases are inconsistent but trend toward including government-ordered cleanups within the term “damages”, regardless of whether a legal action is commenced. In one American case, Morton International Inc. v. General Accident Insurance Co. of America, the insured’s own remediation expenses were held to be damages for which the insurer must indemnify.

It is submitted that apart from the obvious financial aspect, the real issue of concern to insurers associated with government ordered cleanup costs is the fact that the specific risks may not have been fully addressed when the CGL policy was purchased. Now that such risks have been recognized, insurers have and continue to develop new policy wordings to give effect to their original intention concerning the scope of coverage that is being purchased.

Nonetheless, given the unique nature of environmental damage, we consider it likely that the courts will attempt to encourage insurance coverage for cleanup costs, without the need for a finding of third party damages as a precondition. Simply put, the courts will not want to provide an incentive for a polluter to wait until a discharge damages a third party’s property before the polluter may receive insurance coverage. Judicial treatment of this matter in the United States has not been entirely consistent.

One American case on point that cuts both ways is that of AIU Insurance Company v. Superior Court of St. Clara County. In that case, the California Supreme Court held that the CGL policy in question covered the cost of reimbursing government agencies and complying with injunctions ordering cleanup under the Comprehensive Environmental Response Compensation Liability Act and similar statutes. However, the Court held that costs incurred to pay for measures taken in advance of any release of hazardous waste were not incurred because of property damage and were not within the coverage of the policies.


The 2002 Ontario Court of Appeal decision in Zurich Insurance Co. v. 686234 Ontario Ltd. provides a good overview of the principal problem currently facing interpreters of the absolute exclusion clause and states, in our opinion, the current status of the law in Canada. The principal problem with such clauses, alluded to above, is whether the “absolute” exclusion applies to only “traditional” forms of environmental liability (e.g., migrating toxic substances from one property to another) or whether it applies whenever the harm alleged includes an “environmental” component. In the context of carbon monoxide poisoning resulting from a defective furnace, the Court held:

The American authorities that have interpreted the absolute pollution liability exclusion in cases involving claims arising from carbon monoxide poisoning have not reached a uniform interpretation. One line of cases, as exemplified by Essex Insurance Co., has construed the exclusion literally and held that it bars claims arising from common business hazards such as carbon monoxide poisoning claims not normally viewed as pollution. Another line of cases, as exemplified by Koloms and Stoney Run, has held that the exclusion does not bar such claims. In reaching this result, these cases have declined to focus hyperliterally on the text of the exclusion, and have applied various interpretative approaches including finding ambiguity in the exclusion, considering the history of the exclusion clause and its environmental context, the purpose of the CGL policy, and the objectively reasonable expectation of the parties. As I have pointed out, these are the essential grounds relied on by the respondent and, generally, applied by the application judge, for construing the exclusion against Zurich.

I find the second line of American cases to be more persuasive than the line of cases that has literally interpreted the exclusion. In my view, in construing contracts of insurance, dictionary literalism is often a poor substitute for connotative contextual construction. When the full panoply of insurance contract construction tools is brought to bear on the pollution exclusion, defective maintenance of a furnace giving rise to carbon monoxide poisoning, like related business torts such as temporarily strong odours produced by floor resurfacing or painting, fail the common sense test for determining what is “pollution”. These represent claims long covered by CGL insurance policies. To apply an exclusion intended to bar coverage for claims arising from environmental pollution to carbon monoxide poisoning from a faulty furnace, is to deny the history of the exclusion, the purpose of CGL insurance, and the reasonable expectations of policyholders in acquiring the insurance.

Another decision where the absolute exclusion was held not to apply was Medicine Hat (City) v. Continental Casualty. In this case employees of the city sustained neurological problems from the use of methanol and lubrizol in city buses. It was acknowledged that lubrizol and methanol were pollutants. The court found that as the damage could occur without a “discharge” (e.g., during normal maintenance of the buses) the exclusion did not apply. The court stated:

Discharge, dispersal, release or escape of pollutants” is the language of improper or unintended events or conduct. It is not the language of intended use or consequences or of the normal operation of facilities or vehicles. In this case, the polluting substance or gas is part of and confined to the intended and normal operation of a transit garage and buses. This conduct and these events do not fall within the exclusion clause. In my view, the pollution exclusion clause is intended to protect the insurer from liability for the enforcement of environmental laws. The exclusion clause uses environmental terms of art because it is intended to exclude coverage only as it relates to environmental pollution and the improper disposal or contamination of hazardous waste.

Of somewhat similar effect, and of some comfort to those charged with the job of remediating contamination, is the case of Trafalgar Insurance Co. of Canada v. Imperial Oil Ltd. In this case the policyholder was in the process of cleaning up an oil spill in a home when, allegedly, its conduct caused the contamination to spread. The court found coverage for the policyholder’s negligence in permitting the spread of the pollution as the expanded pollution did not constitute a new “escape” beyond the continuous exposure occasioned from the initial escape. However, this decision has been criticized in an influential text as going beyond the bounds of the CGL wording at issue and turning the policy into, in effect, an Environmental Professional Liability policy (which would have cost an additional premium).

In Hay Bay Genetics Inc. v. MacGregor Concrete Products Ltd., the insured sold septic tanks. The court found that to deny coverage when a problem arose with one of the tanks would defeat the entire purpose of the CGL:

MacGregor would not have taken out this insurance coverage if it were not to cover potential pollution risks. MacGregor is not in the business of polluting the environment as a result of the nature of its business. Pollution may have been a risk, but it was not a probable consequence of carrying out its business.

In Pier Mac Petroleum Installation Ltd. v. AXA Pacific Insurance Co. the British Columbia Supreme Court held that the exclusion preclude coverage for the cost of repairs caused by a petroleum leakage resulting from the negligent construction of a gas bar. It appears that the court construed the exclusion literally.

In Great West Development Marine Corp. v. Canadian Surety Co., another British Columbia Supreme Court decision, the insured was the owner and developer of a condominium project. Fill from the project was sold to a person who later sued the insured, alleging that she had received poor quality soil containing construction debris that would leach toxic chemicals and contaminate her crops and groundwater. The court held that the insurer had a duty to defend as the underlying claim did not entirely rest upon the threatened escape of pollutants. As the thrust of the claim was that the fill was of poor quality, this by itself did not bring the claim within the scope of the exclusion.


The most commonly used alternate pollution exclusion clause in British Columbia is the so-called “limited pollution liability” (“LPL”) exclusion clause. The LPL exclusion can be added as an endorsement to the standard wording of the policy and replaces the absolute exclusion. The LPL is more forgiving than the absolute exclusion as it permits coverage for pollution liability where there is an unexpected or unintentional discharge of pollutants that is detected within a very short time after the commencement of the discharge and is reported to the insurer almost immediately. However, a policy with LPL wording still generally does not respond to gradual pollution or to cleanup of premises owned or controlled by the insured.


This paper demonstrates the benefits of retaining old policies and considering insurance coverage when a claim is made, and illustrates the significant limitations current CGL policies place on environmental liability coverage. Again, specific policy wording is paramount, and if you are concerned that your CGL policy does not provide sufficient, targeted coverage then consider obtaining insurance that specifically addresses environmental liability.

We're here to Help

Work With Us

Named One of Canada's Top Insurance Defence Litigation Boutiques by Canadian Lawyer magazine