August 1, 2015
The Role of Fortuity in Builders’ All Risk Policies
The Role of Fortuity in Builders’ All Risk Policies
Eric A. Dolden and Paul C. Dawson
|Eric Dolden||Paul Dawson|
THE ROLE OF FORTUITY IN BUILDERS’ ALL RISK POLICIES
By: Eric A. Dolden
The complexity and scale of modern construction projects have made the concept of a property insurance policy covering all conceivable risks advantageous to both insureds and their insurers. Insureds benefit from the extensive nature and scope of the coverage, and insurers benefit from the economies of managing and marketing a policy which provides them with reasonable underwriting certainty. For these reasons, the Builders’ All Risk policy (sometimes referred to as a Builders’ Risk or Course of Construction policy), is attractive to both the insurance industry and consumers.
However, “all risk” forms of property insurance do not actually insure against all possible risks. In this paper, we first review in Section II the origins and purpose of Builders’ Risk policies. In Sections III to VII, we demonstrate how in recent years the courts’ renewed emphasis on the principle of fortuity is re-shaping the approach to determining “all risks” coverage. We close in Section VIII by reviewing typical exclusions for non-fortuitous losses.
II. HISTORY AND PURPOSE OF BUILDERS’ RISK POLICIES
“All risk” property policies were originally developed in the context of marine insurance, and were intended to provide indemnity for all fortuitous losses other than those arising from the insured’s fraud or misconduct. That intent is reflected in the policy wording which typically states:
This Policy insures against all risks of physical loss of or damage to the property insured, except as hereinafter provided.
In policies insuring equipment and machinery the commonly-used wording provides:
This Policy insures against all risks of physical loss of or damage from any external cause except as hereinafter provided. (emphasis added)
“All risk” policies have evolved so dramatically that not only does they not indemnify for all losses, they do not indemnify for all risks. In order to understand why this has occurred, and the impact these changes have had within the insurance industry, it is useful to examine the origins of the such policies.
The market for property insurance began to develop after the Industrial Revolution. It was not until the great fire in London in 1666 that fire insurance for property became widespread. In the 17th Century, fire was the primary risk for property insurance. In the marine field, by contrast, ship-owners were exposed to a great number of potentially insurable risks. As a result, marine insurers eventually developed a form of “all risk” coverage.
The development of railways and the shipment of goods between commercial centres necessitated insurance which, although known as “inland marine insurance”, was in fact written as “all risk coverage.” With the subsequent development of other forms of transportation, coupled with the increased range of risks that these other modes of transport created, inland marine insurance gained greater market acceptance. In Insurance Company of North America v. Commissioner of Insurance it was noted that:
Inland marine insurance is comparatively new. It is an outgrowth of the development of land transportation. The Federal operation of railroads in the overcrowded conditions of World War I, the inauguration of shipment by motor truck and by airplane, greater mobility of population, and increased traffic in personal property were some of the factors accelerating the need for a type of insurance policy which would cover portable personal property other than at fixed location. The name itself is a misnomer, and includes many forms of insurance wholly lacking in marine features. While the chief characteristic is a relation to transportation, there are recognized categories which have no such relation…
Consumer demand for broader coverage on dwellings and the contents of those dwellings led to the adoption of “all risk” coverage in both residential and commercial settings, and ultimately, to its adoption in the construction context, as Builders’ Risk insurance.
Most construction projects today are undertaken with the benefit of coverage under a Builders’ Risk policy. A Builders’ Risk policy is an “all risk” policy which is maintained throughout the period from the commencement of construction until completion of the project. Most owners, as a term of any contract, will insist that the contractor obtain Builders’ Risk coverage during the course of construction. Illustrative of this is the Canadian Construction Association Document 2, of which General Condition 20 states:
[T]he contractor shall provide, maintain and pay for the insurance coverages listed in this General Condition … (d)… (i) All risks property Insurance shall be in the joint names of the Contractor, the Owner and the Consultant … The form of this Insurance shall be the latest edition of CCDC Form 201 and shall be maintained continuously until ten (10) days after the date of Total Performance of the Work, as set out in the certificate of Total Performance of the Work.”
In purchasing a Builders’ All Risk policy, contractors believe indemnity will be available in the event of an accident or damage on the construction site arising as a result of a party’s carelessness or negligent acts. All of the parties desire an assurance that in the event of a loss there will be sufficient funds on hand to rebuild the project. That is why, in keeping with CCDC 2, the contractor is usually required to obtain insurance for not less than the value of the construction contract. Canadian courts acknowledge the desirability of this practice. Mr. Justice De Grandpre, in Commonwealth Construction Company Limited v. Imperial Oil Limited. et al discussed the role of the Builders’ All Risk policy:
Whatever its label, its function is to provide to the owner the promise that the contractors will have the funds to rebuild in case of loss and to the contractors the protection against the crippling costs of starting afresh in such an event, the whole without resort to litigation in case of negligence by anyone connected with the construction, a risk accepted by the insurers at the outset. This purpose recognizes the importance of keeping to a minimum the difficulties that are bound to be created by the large number of participants in a major construction project, the complexity of which needs no demonstration. It also recognizes the realities of industrial life.
The passage above illustrates three of the perceived advantages of “all risk” coverage – whether marine, homeowners’, or Builders’ Risk:
- They cover all conceivable risks so that the consumer would be under no misapprehension as to the scope of coverage.
- They avoid overlapping coverage, where many individuals are involved with the insured property.
- They provide insurers with certainty and economies of scale in the marketing of insurance.
The first of these three benefits has historically been limited by the need to demonstrate both fortuity and external cause for indemnity. The Builders’ Risk policy does not indemnify for all loss nor does it cover all significant risks likely to be encountered when insuring property. In reality the term “all risk” is, in today’s insurance market, no more than a convenient label, which has little basis in fact.
However, as outlined below, the courts’ evolving interpretation of fortuity has gradually eliminated the requirement that the loss be external, and the focus on fortuity is prompting a new, more consistent approach to the interpretation of the many exclusions found in “all risk” policies.
Our analysis below begins by reviewing several limitations on coverage implied by virtue of traditional insurance concepts, such as:
- the requirement that the loss be fortuitous;
- the need for the insured to demonstrate that the proximate cause of the loss was a covered risk; and
- the need for the insured to demonstrate that the cause of the loss was external in nature.
We then turn to examine the effect of the “modern trend” on the interpretation of specific express exclusions commonly found in “all risk” policies, such as:
- loss or damage caused by “faulty workmanship”;
- loss or damage caused by a “defective design”;
- non-fortuitous loss or damage, such as might be caused by “wear and tear”, “inherent vice”, “latent defects”, “mechanical breakdown”; “mysterious disappearance”, etc.
III. CIRCUMSTANCES OF THE LOSS MUST BE FORTUITOUS:
Most within the insurance industry recognize that only fortuitous losses are insurable. Insurance is not provided for certainties. The concept of fortuity is central to the very concept of “all risks” insurance:
It is important to begin with an understanding of what is meant by an “all risks” coverage and, also, by the expression “more specifically insured”. According to Couch on Insurance, 2nd Rev. ed. (1982), at p. 48:141: “[R]ecovery under an ‘all-risk’ policy will, as a rule, be allowed for all fortuitous losses not resulting from misconduct or fraud, unless the policy contains a specific provision expressly excluding the loss from coverage.” To the same effect is the definition approved by the House of Lords in British & Foreign Marine Insurance Co. v. Gaunt,  2 A.C. 41 at pp. 47 and 51-52,  All E.R. Rep. 447: “all risks . . . were intended to cover all losses by any accidental cause of any kind” (referring to Schloss Bros. v. Stevens,  2 K.B. 665 at p. 673, 75 L.J.K.B. 927 (C.A.)). See also Keeton and Widiss, Insurance Law (1988), at p. 19: “The term ‘all-risk’ insurance is generally used to refer to a method of defining insured events as all fortuitous losses.”
Given that Builders’ Risk insurance constitutes a promise to pay only upon the fortuitous happening of loss or damage from any cause, it might seem that exclusions for inevitable or virtually inevitable loss (e.g., wear and tear, inherent vice and latent defects, all discussed below) should be unnecessary – by definition, such losses would not be fortuitous. Nonetheless, it is in fact very important, from the insurer’s point of view, that these exclusions be specifically included in Builders’ Risk policies because, in recent decades, the courts have moved to define the requirement of “fortuity” in a manner which strongly favours the position of the insured.
This shift has occurred because the test for fortuity is a subjective one. If the insured was actually not aware of the defect or vice inherent in the insured property, or if the damage was actually unanticipated or unforeseen, the loss was fortuitous from the standpoint of the insured and therefore within Builders’ Risk coverage.
This is well illustrated by the decision in Compagnie des Bauxites de Guinea v. Insurance Co. of North America. On a preliminary motion the Court held that an unknown design defect did satisfy the fortuity requirement. In considering the element of fortuity, one does not, with the benefit of hindsight, ask whether the loss would inevitably have arisen. Instead the issue is whether, based upon what the insured actually knew and appreciated prior to the loss, the loss would have seemed inevitable. This approach, based upon the insured’s actual knowledge at the time the policy was entered into, is clearly articulated in a later decision, Standard Structural Steel Co. v. Bethlehem Steel Corp., where the Court stated:
A fortuitous event is one which occurs accidentally, as a layman, and not as a technician or scientist would understand the term. It is an event `which happens by chance … unexpectedly or without known cause, one which is unplanned. An event which is certain to take place or is inevitable cannot be fortuitous, but a court must exercise restraint in its determination of whether an event was inevitable. It must also gauge inevitability by standing in the shoe of the parties at the time the contract of insurance was made.
More recently, the Supreme Court of Canada commented on the concept of fortuity, albeit in the context of a CGL policy, in Progressive Homes Ltd. v. Lombard General Insurance Co. of Canada:
Fortuity is built into the definition of “accident” itself as the insured is required to show that the damage was “neither expected nor intended from the standpoint of the Insured”. This definition is consistent with this Court’s core understanding of “accident”: “an unlooked-for mishap or an untoward event which is not expected or designed” (Gibbens, at para. 22; Martin v. American International Assurance Life Co., 2003 SCC 16,  1 S.C.R. 158, at para. 20; Canadian Indemnity, at pp. 315-16; originating in Fenton v. J. Thorley & Co., Ltd.,  A.C. 443, at p. 448). When an event is unlooked for, unexpected or not intended by the insured, it is fortuitous.
The emphasis on the insured’s pre-loss perception of risk (i.e., an event “not expected or designed”) as the litmus test of fortuity means that a loss resulting from the use of defective materials or work methods may still be fortuitous. This interpretation has significantly altered the courts’ approach to the question of whether loss must be caused “externally” if it is to be covered under an “all risks” policy, a topic we address in greater detail in Section V, below.
Applying the principle of fortuity requires taking into account differences between insureds. For example, a contractor might not be aware of a risk created by a sub-contractor’s deliberate use of defective materials. Similarly, an “all risk” property insurer was denied summary judgment in a case where a building had suffered structural damage due to the deliberate acts of a tenant; the court accepted that the damage might be fortuitous from the perspective of the unsuspecting landlord: Chanore Property Inc. v. ING Insurance.
It is obvious from the cases discussed above that the insured can more readily satisfy the requirement of fortuity. Conversely, an insurer cannot be confident of avoiding coverage for an apparently inevitable loss unless the policy contains specific exclusions clearly intended to meet the contingency of the loss. In terms of policy interpretation the Canadian courts typically use this approach to favour insureds on coverage disputes. From an evidentiary standpoint the insured need only prove that a loss occurred within the policy period. The insurer will have the burden of establishing both non-fortuity and the requirements of any applicable exclusion (Feuiltault Solution Systems inc. v. Zurich Canada, citing British & Foreign Marine Insurance Company v. Gaunt15 and Nelson Marketing International Inc. v. Royal and Sun Alliance Insurance Co. of Canada.)
It is important to note that an insurer is not required to cover a loss simply because the insured failed to anticipate it; in other words, all covered losses must be fortuitous, but not all fortuitous losses are necessarily covered. Fortuity does not guarantee coverage – it only means that the loss is not automatically excluded (Algonquin Power v. Chubb Insurance Co. of Canada).
IV. DOES PROXIMATE CAUSE MATTER ANYMORE?:
To be covered under a Builders’ Risk policy, the “proximate cause” of the loss must be within the scope of coverage, and not otherwise excluded; where an insurer seeks to rely on an exclusion for a specified risk, the insurer must demonstrate that the excepted peril is the proximate cause of the loss.
To be merely a cause is not sufficient – the idea of “proximate cause” connotes an important or significant cause. In Triple Five Corp. v Simcoe & Erie Group, the Court described “proximate cause” as “the cause required for insurance liability under the policy, and also…a critical or dominant event in the chain of events that led to the loss”. The Court in that case concluded that the proximate cause of a rollercoaster accident was defective design, an excluded peril, and so the loss was not covered.19 Another case describes “proximate cause” as meaning “the direct, dominant, operative, and effective cause”.
Accidents in the construction context will only rarely have a single cause, or even a single “dominant” cause. Far more common are situations involving multiple causes that either consecutively or concurrently produce a loss. Determining coverage in such situations can be far more complex. For example, in Canevada Country Communities Inc. v. GAN Insurance Co. a sudden freeze damaged the sprinkler pipes in an almost-completed building project; the subsequent thaw released water which damaged the walls, floors, and other building components. The Builders’ Risk policy excluded damaged caused by “directly or indirectly” by freezing. The court ruled that the exclusion only applied to the damaged sprinkler pipes; the “proximate cause” of the rest of the damage was the release of water, a covered peril, even though the release was precipitated (caused “indirectly”) by freezing.
However, the Supreme Court of Canada in Derksen v. 539938 Ontario Ltd. fundamentally altered the approach courts must follow in addressing concurrent causes in insurance coverage disputes. In Derksen, the defendant’s employee negligently loaded a truck with an unsecure load. When the truck was driven the load came loose, flew off, and hit a school bus, injuring children aboard the bus. The defendant had an auto policy and a commercial general liability policy (“CGL”), each of which could potentially respond to this loss.
The CGL policy excluded bodily injury or property damage arising out of “…the use or operation of …any automobile.” The CGL insurer argued that one of the proximate causes of the loss was the operation of the truck and, as a result, the loss should be excluded. The Court rejected this analysis and stated that the utility of the “proximate cause” analysis with respect to insurance policies is questionable. The Court concluded that the accident was caused by two concurrent causes, and in such circumstances, it is “undesirable” to attempt to select either as the “proximate” cause.
Instead, the Court stated that where there are concurrent causes of a loss, there is no presumption that all coverage is ousted if one of the concurrent causes is an excluded peril. An insurer who wishes to oust coverage in such cases must state that expressly in the policy. The Court stated further that whether an exclusion applies in a particular case of concurrent causes is a matter of interpretation. In this case, the words “arising out of” in the exclusion do not indicate that coverage for an insured risk will be inoperative if an expressly excluded risk constitutes an additional cause. Since the loss arose partly from the operation of the truck, the exclusion is triggered, but only as to the portion of the loss attributable to the auto-related cause. The negligent clean up of the work site was a non-auto-related cause and, absent express language to the contrary, the CGL should cover this risk.
However, the Court in Derksen explained that insurers can use “anti-concurrency” language to avoid undesirable expansions of coverage resulting from concurrent causation. For example, an exclusion that applies to all loss “regardless of the cause of the excluded event, other causes of the loss, or whether other causes acted concurrently or in any sequence with the excluded event to produce the loss” allowed an insurer to successfully exclude loss resulting from both covered and excluded causes. In the “all-risk” context, a policy that excluded loss caused “directly or indirectly” by water damage did not cover the cost of remediating mould contamination, since the mould was supported by moisture damage.
Derksen has been applied many times in the context of “all risk” property insurance policies, including in particular situations in which residential properties have been damaged by water and/or earth movement, as well as damage to insureds’ products arising from power interruptions. In the Builders’ Risk context, courts have in several cases distinguished Derksen by concluding that the loss arose from a single “proximate cause” rather than concurrent causes.
In summary, the concept of “proximate cause” is still relevant to the interpretation of Builders’ Risk and other “all risk” property policies, but an insurer that wishes to rely upon the concept to exclude a claim will likely have to demonstrate that the loss results from only one cause – or else have inserted “anti-concurrency” language into its policy. Otherwise, where a loss is caused by two concurrent causes, some of which are covered and others are not, the approach prescribed by Derksen will typically result in coverage.
V. MUST THE CAUSE OF LOSS BE EXTERNAL IN NATURE?:
As noted at the outset of this paper, it used to be a hallmark of “all risks” policies that they would only cover loss resulting from a cause external to the insured property itself. Externality was closely related to the concept of fortuity, as it was understood at the time, i.e., in that the damage could not result from an inherent defect or some inevitable condition within the property itself. Historically, in the context of marine policies, the expression “external cause” was sometimes expressly added to the policy language, the intention being to exclude from coverage:
- losses resulting from negligent acts of the owner or master;
- losses resulting from normal wear and tear; and
- losses resulting from internal decomposition or deterioration of the insured property.
Modern cases have noted that the relationship between externality and fortuity is more complex, as noted by the Alberta Court of Appeal in Triple Five, supra, at para. 9:
 The trial judge concluded also that this “design error” was an “external” cause, and thus there was coverage under this “all-risk” policy unless the peril was excluded. For the Mall, this was an essential finding. An all-risk policy has been held to insure against only perils external to the property, as opposed to perils that are somehow inside the property. British and Foreign Marine Insurance v. Gaunt  2 A.C. 41 (H.L.) at 46. For the purpose of this appeal, we accept the finding that the design error was external to the roller-coaster. But we point out that all errors in the course of manufacture are in this sense “external” to the manufactured goods…
More recent decisions from the Supreme Court of Canada have rejected the concept of externality as an artificial false dichotomy. As noted in Canadian National Railway v. Royal and Sun Alliance, the “internal” elements of a design (or, we might add, a product) will always operate in “external” conditions. As noted above, the Court in Progressive Homes, supra, has held that “property damage” need not imply damage to property other than that on which the insured is working, but can include damage to the insured property itself. The most recent pronouncement on this issue, from the Alberta Court of Appeal in Ledcor, supra, definitively dismisses “externality” as a requirement for coverage:
 Progressive Homes confirms that the analysis starts with the wording of the policy. Firstly, the policy is said to be “all risks” and defines the coverage broadly: “direct physical loss or damage except as hereinafter provided”. “Damage” can refer to damage to one’s own property, and is not confined to damage to the property of third parties, nor is it confined to damage caused by forces external to the property: Progressive Homes at paras. 34-7.
The more modern approach adopted in these recent decisions is to subsume “externality” within the requirement of “fortuity”; risks are covered not because they are caused “externally”, but because they are fortuitous, which can include unexpected and unintended “internal” causes, such as design defects, or the unanticipated results of an improper work technique.
In light of these decisions, the traditional “all risks” insuring agreement (i.e., all risk of direct physical loss or damage to insured property) likely no longer requires as a precondition to coverage that the loss be caused externally.
What, then, can we make of policies that still explicitly require externality? Policies that insure equipment and machinery typically insert the concept into their insuring agreements (i.e., “all risks of direct physical loss or damage to the property insured from any external cause“). For example, in Farry Excavating and Grading Ltd., supra, the insured sought indemnity for repairs to a front-end loader allegedly damaged by vandals. The insurer contended that the damage resulted from improper maintenance, as sand had been found in the transmission. Like most “all risk” policies covering equipment, the policy only covered loss attributable to an external cause. The Court relied upon an Ontario case, Nadrofsky Steel Erecting Ltd. v. Hartford Fire Insurance Co., in which Pennell, J. stated:
The phrase `external cause’ is concerned with an outward source or origin or an instigating agent. A cause which has an external source or origin is not rendered internal by the fact that its effect is internal, since it is the means and not the injury itself to which the phrase refers …
The word ‘cause’ has a number of meanings. The choice of the real cause must be made by applying common sense standards: … The real cause, involved as it may be with other causes, must be at least something without which the event would not have happened. It is a substantial factor in producing the event.
When these passages are read together it is clear that “external cause” must mean “a substantial factor coming from without”. On the evidence, the Court accepted that the loss was likely caused by vandals putting sand into the transmission, and was thus covered, even though the sand’s effect only manifested later in the form of an apparently “internal” breakdown.
The requirement of externality in equipment policies has caused the courts much difficulty over the years, and has led to arguably contradictory results. For example, the Court in Nadrofsky, supra, concluded that improper operation of a crane, when combined with pre-existing damage, was an external cause of the crane’s collapse, and so the loss was covered. However, in Krane Service Ltd. v. American Home Assurance Co. the operator’s failure to install a cotter pin in a mobile crane, together with everyday vibration, caused the crane to tip over; the Court concluded that the missing pin was an internal cause, and so the loss was not covered. In both cases, a combination of “internal” conditions and “external” operator error caused the loss, but different interpretations of “externality” and causation led to opposite results.
Both Nadrofsky and Krane Service highlight a further difficulty with externality – external to what? The courts may or may not treat separate parts of a piece of equipment as one indivisible unit. Damage originating from one part can be “from without” if it causes harm to another part. In John Kmita v. Saskatchewan Government Insurance Office, the insured serviced oil wells. A pump installed at a well site was damaged when a relief valve located some distance away on a discharge line failed to open under extreme pressure. The Court concluded that the relief valve was external to the pump and that indemnity was available for the pump. The valve and the pump were considered separate units and a fault emanating from the valve to the pump was “from without”. Would the Court have reached the same conclusion if the valve had been closer to the pump, or perhaps installed in the same housing?
The contradictions and complications illustrated by Farry Excavating, Nadorfsky, Krane Service, and John Kmita may potentially be avoided by the modern trend, typified by Canadian National Railway, Progressive Homes, and Ledcor, of discarding “externality” as a condition of coverage. Focussing on fortuity as the guiding principle, which imports an analysis of what the insured knew or ought to have known about the potential risk, allows the Court to sidestep the metaphysical parsing of internal and external causes and components. The result in each case will be guided by whether the loss is truly accidental – in the sense that it was unexpected or unintended from the perspective of the insured, and also in the sense that the loss was neither inherent in the property nor inevitable – taking into account all “external” and “internal” factors.
The renewed on fortuity is also leading to a reinterpretation of the many exclusions that have been developed over the years in “all risk” policies to exclude non-fortuitous losses, and to which we now turn below.
VI. THE “FAULTY WORKMANSHIP” AND “DEFECTIVE DESIGN” EXCLUSIONS:
Builders’ Risk coverage was intended to protect contractors and owners from the financial ruin which could result from major damage to property, including loss during construction. Given this background it is not surprising that contractors and owners would view with concern the “faulty workmanship” and “defective design” exclusions found in most Builders’ Risk policies. The precise wording of these exclusions varies depending upon the insurer. By way of an example, the IBC Broad Form (Form #51208) exclusion is worded as follows:
This Policy does not insure…..
(a) the cost of making good
(i) faulty or improper material;
(ii) faulty or improper workmanship;
(iii) faulty or improper design or loss or damage caused directly or indirectly therefrom provided however to the extent not otherwise excluded under this Policy, resultant damage to the property insured caused by fire or explosion shall be insured.
It is the nature of the exclusion that creates problems. On the one hand, owners and contractors recognize that negligent conduct is the most common source of loss on construction site. For that reason, any type of property policy which does not afford indemnity in those circumstances is of limited commercial value to the construction industry. On the other hand, the insurance industry fears that without these exclusions insureds would be given carte blanche to engage in negligent construction work. To provide indemnity in those circumstances would discourage proper workmanship and design.
Upon their introduction the “faulty workmanship” and “defective design” exclusions were given a very limited scope. The United States courts were concerned that to do otherwise would eliminate indemnity in circumstances involving mere negligence on the work site. The Canadian courts initially took the same approach, but for several decades they tended to widen the operation of the exclusions so as to deny indemnity if the loss is attributable to the insured’s own carelessness – seriously undermining the original purpose of Builders’ Risk coverage. However, the most recent decisions concerning “all risks” policies are again limiting the scope of this and other exclusions, by emphasizing the importance of fortuity.
A. THE INITIAL U.S. APPROACH:
We begin our survey with early American authorities such as City of Barre, Vermont v. New Hampshire Ins. Co., in order to illustrate how the exclusions used to be interpreted. In that case, strong winds caused a number of support arches at a construction site to collapse. Upon investigation, it emerged that the contractor had installed only two of the six guy cables specified in the architect’s plans, and that the two cables which had been installed were under-strength. The Court interpreted the “faulty workmanship” narrowly, stating that the term was not synonymous with negligence:
[I]f the intent of the policy was to exclude liability from loss caused by the negligence on the part of the contractor, it is not unfair to require that such intent be clearly expressed, particularly when one exclusion from coverage refers to “loss caused … by … any intentional act on the part of the insured…”. Express exclusion of an intentional act would seem a clear indication the consequences of a negligent act were within coverage, and not excluded.
American courts also initially limited the scope of the “faulty workmanship” exclusion by viewing “faulty workmanship” as referring solely to the product or building, not to errors or mistakes made by construction workers. This approach is best exemplified by the decision in Equitable Fire & Marine Insurance Company v. Allied Steel Construction Co. In that case, pipe fittings had been improperly fitted although, as products, they were entirely suited to the intended task. In concluding that the “faulty workmanship” exclusion did not apply the Court stated:
The non-insured risks deal with design, specifications, workmanship and materials. If [the Insurer’s] … meaning of workmanship is used, the continuity of that work group is disrupted. However, by using the plain and ordinary meaning of the word “workmanship” – the art of skill of workmen – the internal consistency of the paragraph is maintained. We think it plain that … the non-insured risks dealt only with loss caused by defects in the pipe or piers.
Neither of these methods of restraining the scope of the “faulty workmanship” and “defective design” exclusions are entirely satisfactory. The former distinguishes “fault” from “negligence”, but does not explain the difference between the terms, while the latter seems to read workmanship and design out of the exclusion altogether. As will be seen further below, modern Canadian cases are squarely addressing both problems by emphasizing the principle of fortuity.
B. THE INITIAL CANADIAN APPROACH:
Early Canadian authorities also treated the exclusions narrowly, often by distinguishing between loss caused “on-site” or “off-site”. The courts reasoned that the exclusion only applied to problems occurring at the insured premises; problems originating “off site”, e.g., through defective design, could not therefore be “faulty workmanship”.
Representative of these early Canadian cases is the decision in The Foundation Company of Canada Limited v. Aetna Casualty Company of Canada et al. The contractor had improperly compacted the material used for pouring a cement floor, and an issue arose as to whether the exclusion for “faulty or defective workmanship” barred recovery. In Foundation (supra) the insurer also had the benefit of an exclusion which barred recovery for “faulty or defective construction” (emphasis added). Notwithstanding the existence of both exclusions, the Court permitted the insured to recover.
A related method of limiting the “faulty workmanship” exclusion was to characterize the cause of the problem as a design deficiency. This avoided any suggestion that “faulty workmanship” was involved. That provided relief for insureds with the benefit of extended coverage for design defects, in cases where the policy omitted any exclusion for this risk.
Maclab Enterprises Ltd. v. Commonwealth Insurance Company et al, a decision of the Alberta Queens Bench, illustrates both of these approaches. Brick panels on an office building began to come loose thirteen years after its construction, and portions of the brick wall ultimately had to be replaced. The insurer’s Builders’ Risk policy excluded indemnity for “faulty workmanship” but did not exclude “faulty design”. Expert evidence was led at trial that the mortar specified by the architect was unsuited to the particular application. The Court accepted the premise that “design defect” had a very wide meaning and included within its ambit the architect’s directions necessary for the construction of the building. By accepting that “faulty design” encompassed the “mental concept” of translating the plans and specifications into a physical object, and by limiting “workmanship” to those matters which clearly arose on the job site, the Court concluded that the use of improper mortar did not fall within the “faulty workmanship” exclusion.
This restrictive approach to the “faulty workmanship” exclusion was again evidenced in Simcoe & Erie General Insurance Company v. Willowbrook Homes (1964) Ltd., a decision of the Alberta Court of Appeal. A contractor, hired to erect warehouses made out of concrete sections, used inadequate bracing to support the partially-built warehouses. That factor, in combination with high winds, caused the warehouses’ walls to collapse. A consulting engineer had been involved in both the original formulation of the bracing procedure and had attended at the work site to ensure proper implementation of the recommended procedure.
The Builders’ Risk policy in issue in Willowbrook Homes excluded both faulty workmanship and defective design. The Alberta Court of Appeal concluded that the “faulty workmanship” exclusion did not apply because, until the building was completed, the bracing was dictated by design procedures, whether emanating from the plans, specifications or the architect’s recommended procedures, and so fell more clearly within the ambit of “faulty design” than “faulty workmanship”.
The decision in Willowbrook Homes reflects the courts’ willingness to limit the scope of the faulty workmanship exclusion by means of “locational” considerations. Distinctions are made based on whether the error occurred on the site as opposed to off the site. The decision correctly assumes that many traditional design functions can have a significant impact on the construction site, and that fact militates against the otherwise inevitable conclusion that every failing at the site is necessarily attributable to faulty workmanship.
Again, these methods of limiting the scope of the “faulty workmanship” exclusion are not entirely satisfactory. Most Builders’ Risk policies now exclude defective design as well; treating the cause of the loss as defective design simply displaces the coverage analysis to another part of the policy. The locational approach is also conceptually difficult to apply in modern construction settings. Assume that, as part of a design-build project, an engineer prepares incomplete or misleading performance specifications expected of roofing materials to be used on a project; a contractor bidding on the project proposes to use inappropriate materials, that meet neither the specifications nor the intended use; the contractor then installs the materials incorrectly, and the engineer fails to perceive any problems during site inspections. When the roof begins to leak, did it result from negligent design or workmanship? And did that negligence occur on or off the site?
C. THE WIDENING OF THE EXCLUSIONS:
To this point this paper has examined early cases which gave a narrow ambit to the “faulty workmanship” and “defective design” exclusions. Subsequent Canadian cases, wrestling with the conceptual difficulties discussed above, dramatically widened the scope of this exclusions, and thereby almost completely subverted the purpose and intention underlying Builders’ Risk insurance.
To understand why this has occurred any analysis must begin with the 1968 decision of the Australian High Court in Queensland Government Railways and Electric Power Transmission Pty Ltd. v. Manufacturers’ Mutual Insurance, Ltd. Flooding caused three concrete piers erected in a river bed to collapse. In arbitration it was determined that the insured was entitled to recover notwithstanding an exclusion for “faulty design”. The arbitration panel concluded that the culpability of the designer did not amount to negligence and that, at the time of its creation, the design reflected “state of the art” technology, shown by the fact that record flood levels had been required before the piers collapsed.
In overruling the arbitration decision the High Court indicated that even absent “negligence”, as that term is understood in tort law, indemnity was not available. In the Court’s view “faulty design” embraced conduct less culpable than negligence. The Court stated:
[T]he loss which occurred did arise from faulty design. Let it be accepted, as the arbitrator found, that the piers, as designed, failed to withstand the water force to which they were subjected because they were designed in accordance with engineering knowledge and practice which was deficient, rather than because the designer failed to take advantage of such professional knowledge as there was, nevertheless the loss was due to “faulty design” and the arbitrator has done no more than explain how it happened that the design was faulty. We think it was an error to confine faulty design to the personal failure or non-compliance with standards which would be expected of designing engineers on the part of the designing engineers responsible for the piers. To design something that will not work simply because at the time of its designing insufficient is known about the problems involved and their solution to achieve a successful outcome is a common enough instance of faulty design. The distinction which is relevant is that between “faulty”, i.e., defective, design and design free from defect. We have not found sufficient ground for reading the exclusion in this policy as not covering loss from faulty design when, as here, the piers fell because their design was defective although, according to the finding, not negligently so. The exclusion is not against loss from “negligent designing”, it is against loss from “faulty design”, and the latter is more comprehensive than the former.
This is a direct result of the use of the word “faulty”. In the Court’s view that term, in the context of an insurance policy, did not mean “negligence”. The Court considered “fault” as the term applies to people:
In ordinary English today the word “fault” is used in senses which are all indicative of the basic concept, a falling short, a shortcoming. When used in relation to a person. a fault is a falling short in conduct or behaviour some act or omission which whether wilful or negligent, predicates blame. In this sense the word has a wide meaning when used of a person.
and then as it applies to things:
We are concerned with the word “faulty” as descriptive of an inanimate thing. The words “fault” and “faulty” then have a different sense. They, again according to their derivation, connote a falling short; but not falling short in conduct or behaviour. They designate an objective quality of a thing. It is not up to a required standard.
“Faulty” design is not the same as “negligent” design. In other words, the Court went on to indicate:
Doubtless a faulty design can be the product of fault on the part of the designer. But a man may use skill and care, he may do all that in the circumstances could reasonably be expected of him, and yet produce something which is faulty because it will not answer the purpose for which it was intended. His product may be faulty although he be free of blame.
The philosophy which underlies thus Australian decision was adopted with enthusiasm by judges in Canada. In Pentagon Construction (1969) Co. Ltd. v. United States Fidelity and Guaranty Company, a contractor had been hired to construct a cement tank at a water treatment plant. The contractor failed to install reinforcing straps around the outside of the tank, as required by the structural design, before he filled the tank with water. The tank collapsed. The insurer denied indemnity on the basis of both the “faulty workmanship” and “faulty and improper design” exclusions.
The contractor argued that absent a specific act of negligence the “faulty workmanship” exclusion did not apply. In rejecting that limited view of the exclusion the Court stated that “workmanship”, in the context of an Builders’ Risk policy, referred to the “combination, or conglomeration, of all the skills that were directed to the building of the tank.” In accepting that the failure to follow a particular procedure or sequence fell squarely within the ambit of the “faulty workmanship” exclusion Mr. Justice Robertson stated:
If there was faulty or improper workmanship by any one person at one or more times, or by any several persons at one or more times, or by a combination of persons at one or at several times, or otherwise howsoever, and that workmanship caused the loss or damage, the insurance was not to cover the loss or damage. Once it is established that loss or damage in respect of which [the insured] makes a claim to be indemnified was caused by faulty or improper workmanship, it is neither necessary nor proper to inquire further. There it is. The claim is not covered.
The achievement of the result called for by the contract required a number of steps to be taken in a particular sequence; failure to take them in that sequence could constitute faulty or improper workmanship; all too obviously it did so here. It is of no consequence why the proper sequence was not observed, or what individual was to blame for the failure to observe it, or whether he was employed by [the insured], or that one cannot fix the blame on any particular person. What the reason for it may have been, there was improper workmanship and it caused the damage to the tank.
The Court also concluded that the loss was excluded as “faulty or improper design”, notwithstanding that the factor which actually triggered the loss was the failure of the sub-contractor’s employees to properly undertake the test procedures. In the Court’s view the exclusion was not limited to the design itself, but extended as well to the manifestations of the design concept, including the procedures required to implement the design.
Relying upon dictionary definitions of the term, the British Columbia Court of Appeal concluded that “design” was a broad term which included not only “structural” calculations, shape and location of materials, but their choice, as well as the choice of particular work processes… The Court stated:
In my opinion “faulty or improper design” … must have reference to “design” as contemplated by the construction contract, in the sense of including all the details covered by the drawings and specifications which that contract required to be followed. I do not think the word `design’ can properly be limited to some concept in the designer’s head to the exclusion of the recording of that concept through the drawings and specifications.
The effect of Queensland and Pentagon Construction was to give the workmanship and design exclusions a wide ambit indeed. The conclusion that a contractor’s failure to follow a particular procedure or system amounted to “faulty workmanship”, was echoed in the decision of the Saskatchewan Court of Appeal in Bird Construction v. U.S. Fire Insurance Co. et al., in which a sub-contractor was retained to fabricate and erect steel roof trusses for a fieldhouse. The policy wording was unusual, excluding indemnity in circumstances involving “faulty or defective workmanship or material” or “fault, defect, error or omission in design, plan or specifications”. The plans and construction specifications did not contain instructions on how to incorporate the trusses into the building, so the general contractor issued its own instructions – which he then failed to follow, resulting in the trusses’ collapse during construction. The Court concluded that since the erection procedure was not part of any design, plan or specifications, it could not constitute “defective design”, but it did amount to “faulty workmanship”. The erection of the trusses was part of the work, and the procedure the contractor chose to employ in carrying out the work was “faulty”.
The wide definition of “fault” articulated by the decision in Queensland Government Railways (supra), taken together with the wide meaning ascribed to the term “design”, set the stage for the application of the “faulty design” exclusion in situations where the design problem “spills over” onto the construction site. “Design” activities can merge into construction activities, or “workmanship”, especially where engineers or architects are more closely involved at the construction site. The distinctions between design and workmanship, and between on-site and off-site activities, begin to break down.
This problem becomes evident from reviewing two cases in which the facts and the results were very different. The first is Willowbrook Homes, supra, where the Court extended the design aspect to include all activities until completion of the project, because the design consultant had a significant involvement in the “on-site” procedures. The court stated:
In my opinion the bracing in this case came under the exception design and not workmanship. The temporary bracing required while the walls were in the course of construction and until the roof was put on was part of the design of the building.
That decision contrasts sharply with Todd’s Men & Boys Wear Ltd. et al v. Diamond Masonry (Calgary) Ltd. et al, where the contractor failed to secure a wall section to the building’s steel frame, and the wall collapsed. The Court concluded that the “defective design” exclusion did not apply, because the design consultant had not been involved in either the assembly of the wall section, the selection of the nature or type of bracing to be used, or in supervising the installation of the bracing on the site. Therefore, the failing was properly classed as “faulty workmanship” as opposed to “faulty design” and the insured was entitled to succeed.
Cases such as Pentagon and Bird Construction tried unsuccessfully to resolve unworkable distinctions between workmanship and design, and between on- and off-site negligence.
It became difficult for insurers or insureds alike to determine how the courts would interpret “fault”, or distinguish between “workmanship” and “design”. The practical result was that coverage under Builders’ Risk policies was denied in circumstances involving a broad range of human failings that would, as a matter of tort law, not be characterized as negligence. In one sense, the evolution in Canadian cases “solved” the conceptual difficulties of earlier cases by abandoning distinctions that were unworkable in practice, but at the cost of significantly narrowing the scope of coverage originally intended by “all risk” policies.
Furthermore, all of the cases discussed above suffer from a similar conceptual difficulty, in that they rather arbitrarily select one causal element as the governing, proximate cause of the loss. Older cases emphasize “off site” and design defects, regardless of subsequent faulty workmanship, and thereby sideline the workmanship exclusion. Later cases emphasize “on site” and faulty workmanship, regardless of antecedent design problems, and thereby apply the workmanship exclusion. Both approaches are inflexible, and preclude a nuanced appreciation of how the insureds and others involved in a construction project perceived and allocated risk amongst themselves. In short, the historical approaches to the interpretation of the exclusion are inconsistent with our modern understanding of fortuity.
D. THE “RESULTING DAMAGE” EXCEPTION:
The difficulties of interpreting the “faulty workmanship” and “design defect” exclusions are compounded by the fact that, unique among property policies, Builders’ Risk policies allow indemnity if the “faulty workmanship” or “defective design” causes property damage to third parties. This limited form of cover, in effect a grant of cover within an exclusion, is marketed as one of two wordings:
This policy does not insure the cost of making good faulty or defective workmanship, material, construction or design, but this exclusion shall not apply to damage resulting from such faulty or defective workmanship, material, construction or design.
This policy does not insure … faulty or improper workmanship; provided, however, to the extent otherwise insured and not otherwise excluded under this Policy resultant damage to the property shall be insured.
This limited coverage for “resulting damage” has given rise to two issues in the context of claims. Firstly, does the “resulting damage” exception afford indemnity for the insured’s own losses, or is it intended only to provide relief to third parties who sustain damage? Secondly, if the exception to the exclusion does cover the insured’s own losses, what portion of the insured’s property is covered by reason of this limited coverage? These issues are clearly intertwined, and will be considered together. In the end, they underscore even more clearly the courts’ historic difficulties interpreting the exclusions.
In order to take advantage of the “resulting damage” exception, insureds often tried to distinguish the portion(s) of the building on which they worked from other “property” damaged by their faulty workmanship. Such arguments were usually unsuccessful, as the courts typically treated the insured property as a single functional unit, such that none of the damage caused by “faulty workmanship” could be considered “resulting damage”.
A good example of this approach is Sayers & Associates Limited v. The Insurance Corporation of Ireland et al, a decision of the Ontario Court of Appeal. A sub-contractor was hired to install electrical equipment throughout a new building, and to ensure that any electrical work was protected from the elements. The sub-contractor failed to place protective coverings over some of the electrical openings, so rainwater was able to enter the electrical system. This formed a highly alkaline solution which ultimately caused an electrical malfunction. While conceding that the failure to use protective coverings amounted to “faulty workmanship”, the sub-contractor argued that the ensuing damage to other related equipment, as well as the cost of supplying temporary power to the building, was recoverable on the basis of the exception.
The Ontario Court of Appeal rejected this submission. In the Court’s view the electrical contractor’s work, together with the necessary protective measures, were an integral part of the entire work and all ensuing loss could be treated as “original” damage to the insured property, as opposed to damage resulting from “faulty workmanship”. Stated differently, since the protective measures were within the scope of the “work” to be performed, and not outside the scope of that work, the exception to the exclusion did not apply.
The reasoning in Sayers & Associates Ltd. (supra) was followed in other construction contexts. In Simcoe & Erie General Insurance Company v. Royal Insurance Company of Canada et al, a dispute arose following the collapse of a transit bridge over the Elbow River in Calgary. Evidence at the trial suggested that the collapse was caused by the design engineer’s failure to properly allow for imbalanced weight loads. The superstructure of the bridge had to be replaced, at a cost of $636,945.46.
The primary issue centred over which of two insurers had to defend claims brought by the City of Calgary. The engineer’s E & O insurers contended that the “exception” to the “faulty workmanship” exclusion obligated the “all risk” insurer on the project, Royal Insurance, to indemnify for amounts other than damage attributable to the design error, such as additional engineering costs in re-constructing a replacement bridge. The E & O insurers argued that the deficient section of the bridge, for insurance purposes, could be treated as being functionally separate from the other portions of the bridge which ultimately had to be replaced. This was in turn based on the premise that the damage to those latter sections was attributable to faulty workmanship and design. In concluding that the “design” was intended to encompass the entire bridge the Court rejected any arbitrary “severing” of the entire project into portions, stating instead that the error affected the entire structure:
[I]t appears abundantly clear to me that “design” encompasses the totality of the superstructure, and that each and every part of the superstructure was integral to the whole, and what in fact overturned into the Elbow River was the whole structure. The “design” was in my view fundamental to the whole, and when the design was in error, the whole of the superstructure was doomed to fail and did indeed fail.
None of the costs associated with the construction of the replacement bridge were treated as being within the exception to the exclusion. All of the steps taken to replace the totally failed structure were treated as being the direct consequence of failed workmanship and therefore none of the cost was recoverable under Royal’s Builders’ Risk policy.
Similarly, in Poole Construction Ltd. v. Guardian Assurance Company, coffer dams had been erected in a river bed for flood control purposes, and to facilitate construction of a bridge. The coffer dams failed and it was determined at trial that the loss was excluded by reason of the “design exclusion”. However, the court’s comments provide some insight into the intended scope of the exception to the exclusion. Bowen, J., stated:
[T]he clear intent of the section is to make it clear that the insurer will not indemnify the insured for costs caused by the insured’s own use of faulty workmanship, materials or design. To do otherwise would give the insured carte blanche to use faulty materials, workmanship or design. This opinion is bolstered by the fact that the second exception talks of “damage resulting from”, thereby alluding to an entirely different aspect of the matter. The original exception talks of the “cost of making good faulty design”. That cost of making good covers the entire structure in this case. To make good the faulty design, the Plaintiffs had to remove the structure and fill the excavation. Those are the costs for which they ask to be indemnified and they come precisely within the exception.
However, other cases have shown a greater willingness to consider segmenting construction projects and treat portions of the damage as “resulting”, and thus covered within the exception. In Poole-Pritchard Canadian Limited v. The Underwriting Members of Lloyds, a sub-contractor supplied pipes to the construction of an industrial plant. Steel bands used as fasteners corroded due to the presence of chlorides in the raw material. The issue was whether the insured could recover the cost of replacing the corroded steel bands. The Court said not, as the cost was attributable to the faulty workmanship. Consistent with the later decisions in Sayers & Associates and Simcoe & Erie General Insurance Co., the Court treated the entire project as one unit, incapable of being segmented for insurance purposes, but Mr. Justice Allan went on to speculate as to when the “resulting damage” exception might apply:
I think the words used in the first clause, namely, this exclusion shall not be deemed to exclude loss or damage arising as a consequence of “faulty workmanship, construction or design” are clearly referable to loss or damage other than the cost of repairing or replacing the faulty insulation work. For example, if, as a result thereof, the tanks themselves had been damaged or collapsed the loss thereby sustained might have been recoverable.
Although the decision in Poole-Pritchard arose in the context of a liability policy, the narrow issue was whether the items being claimed were ones “arising directly or indirectly out of injury to or destruction of property, caused by accident, and arising out of the operations…” The case is instructive because the policy excluded the cost of making good faulty workmanship but not loss or damage arising as a consequence of the faulty workmanship. The analysis is not unlike that which a court have had to undertake in considering Builders’ Risk policies.
Other cases have interpreted the “resulting damage” exception as potentially allowing coverage for damage to property owned by third parties, i.e., treating damage to the insured’s property as “original”. In Golden Eagle Canada Ltd. v. The American Home Assurance Co., the insured sought recovery on a Builders’ Risk policy after the roof over two storage tanks collapsed. The roofs over other tanks had to be replaced under threat of imminent collapse. While the initial loss was triggered by winter temperatures, the evidence at trial made it clear that the problems encountered would have occurred under even normal climatic conditions. The court concluded that the loss was excluded by reason of the “latent defect” exclusion, but the insured argued that the cost of replacing the tanks under threat of imminent collapse was recoverable by reason of the exception to the “faulty workmanship” exclusion. Expressing the opinion that it would “be very surprising” that an exception clause would cause an insurer to be liable for a loss not covered in the first place by the insurance contract, the Court stated:
Obviously, what these words mean is that when the collapse of a property, ascribable to an error in design or an inherent vice, causes damages to other properties covered by the insurance, the insured will be entitled to recover for the losses thus sustained despite the fact that the original loss or damage, not being from an external cause, would not be covered. In other words, the original collapse resulting from an internal cause could become an external cause concerning the damages caused by such collapse to other insured properties.
Golden Eagle Canada Ltd. narrowed the ambit of the “resulting damage” exception by requiring that the loss consist of damage caused to a third party who provided materials or tangible property to the insured property. An example might include the sub-contractor that supplies equipment to the construction site and sustains a loss of equipment as a result of the general contractor’s faulty workmanship. The equipment supplier can obtain indemnity if the supplier can demonstrate that its property is not functionally integrated into the entire work or structure that is the subject matter of the Builders’ Risk policy.
The cases discussed above demonstrated that the “resulting damage” exception would only allow the insured to recover its damages, firstly, if the project or property is capable of being functionally divided into separate units and, secondly, if loss or damage was caused to property functionally separate from the insured property.
To summarize: many of the difficulties surrounding the interpretation of the “faulty workmanship” and “defective design” exclusions arise from the terminological problems.
It has been difficult for courts to determine the meaning of “fault”, and whether it is different than “negligence”, and to provide clear guidance on the distinction between “workmanship” and “design”. Attempts to distinguish between fault arising on- or off-site were not successful. It is not surprising, then, that adding the issue of “resulting damage” produced even more confusion. The result has been a complex set of exclusions that limit coverage beyond the original intention of the Builders’ Risk policy.
VII. THE “MODERN APPROACH” – FORTUITY TO THE RESCUE:
In Sections III, IV, and V, above, we explained how recent decisions have revised our understanding of several of the foundational principles of “all risk” insurance, such as fortuity, proximate cause, and externality. That process of reinterpretation is also helping to resolve some of the tangles that have arisen in interpreting the “faulty workmanship” and “defective design” exclusions.
A significant milestone in the courts’ new approach to these issues is the 2008 decision of the Supreme Court of Canada in Canadian National Railway, supra, a case that deserves close attention.
In that case, the insured developed a process for boring tunnels using a “tunnel boring machine” (TBM) of its own design. The insured’s TBM was at that time the largest example of its kind in the world. The insured was hired to construct a tunnel under the St. Clair River. Unfortunately, dirt was able to penetrate the TBM’s central bearings, forcing repairs and causing lengthy delays.
The insured had obtained a Builders’ Risk policy for the St. Clair project. It contained a typical “defective design” exclusion. The trial judge concluded that the exclusion did not apply because the design, though defective, had taken into account all foreseeable risks, including the necessity of sealing the TBM against the intrusion of dirt. The Court of Appeal reversed, holding that the design was “faulty” in that it had failed to keep out the dirt; the delay claims were therefore excluded from coverage.
The Supreme Court restored the trial judgment. It began by observing that the loss had not been caused by “unforeseen operating conditions ‘external’” to the TBM; the central issue was whether the failure of the design necessarily meant that the design was “faulty”, for the purpose of the exclusion. The Court described the Australian decision in Queensland, supra, at length, concluding that it stood for the principle that “faulty” design is one that fails to stand up to foreseeable risks – a principle that the Court roundly rejected, as it could mean that even the best design available (the “state of the art”) could be considered faulty. On the evidence before it, the Court found that the insured’s design for the TBM was not negligent, and should not be treated as faulty simply because it failed to operate as intended:
 I do not believe that where, as here, the risk is broadly defined (“metal deflects under stress”), and the design addresses that risk with state of the art diligence and expertise (as here), the insurers are entitled to the exclusion just because, with the benefit of hindsight, it turns out that “engineering knowledge and practice lacked a proper appreciation” (to quote Queensland again) of the design problem. A narrower interpretation of the exclusion, it seems to me, best accords with the intentions of the parties based on the plain meaning of the words used, namely “faulty or improper”. If the insurers wished to negotiate an exclusion of costs associated with simple “design failure” or “design failure in conditions of foreseeable risk”, it was open to them to have tried to do so but that is not the wording of the policy and this exclusion clause should not, in my opinion, be given that effect.
 The insurers themselves acknowledged in their factum (at para. 58) that while a loss must arise out of a fortuitous event, “the fortuity principle has subsumed the requirement that a risk must be ‘external’ to the property [insured]”. Accordingly, the absence of unforeseeable “external” conditions does not establish that the design was “faulty or improper”. The “internal” elements of the design will always operate in “external” conditions. They are inextricably linked. All of the cases referred to above involved the “internal” failure of a design to accommodate “external” conditions, some of which were unforeseeable and some of which were foreseeable.
In these passages, the Supreme Court discards the approach to “fault” followed in Queensland, and by extension in Pentagon Construction, Willowbrook, and other Canadian decisions. The “fault” exclusions may be applied where a design fails to address foreseeable risks that could reasonably have been addressed. The court in Queensland imposed a standard in which failure of a design spoke for itself; the Supreme Court’s test requires that the design be tested against the “state of the art” rather than perfection.
The Supreme Court’s analysis in Canadian National Railway disposes of externality as a measure of fortuity, and instead treats fortuity as a question of foreseeability. The “faulty design” exclusion cannot be applied with hindsight, reasoning backwards from failure, but must be applied in light of what the designer knew – or should have known – before the loss.
In 2015, the Alberta Court of Appeal applied similar reasoning to the interpretation of the “faulty workmanship” exclusion, in Ledcor Construction Limited, supra. In that case, a sub-contractor was hired to clean paint, concrete and dirt from the windows of a partially-built high-rise tower. The method chosen by the sub-contractor scratched many of the windows, which had to be replaced at considerable expense.
The insurer denied coverage on the basis that the damage resulted from “faulty workmanship”. The trial judge rejected the sub-contractor’s contention that the exclusion could only apply where the workmanship created something, i.e., that cleaning was not “workmanship”, but agreed that the policy was ambiguous as to whether re-installing the windows was “resulting damage”. The trial judge then applied the principle of contra proferentem, and construed the policy against the insurers; the glass replacement was covered.
The Court of Appeal sided with the insurers, and concluded that the loss was excluded from coverage.
- First, the Court observed that the fundamental intent of Builders’ Risk policies is to cover “unexpected events and occurrences”; thus, if the builders used poor quality glass that wore out quickly, the cost of replacing the glass would not be covered. Similarly, the policy would not cover the cost of re-cleaning the windows, if the sub-contractor had simply done a shabby job. In short – the policy only covered fortuitous losses.
- Second, the cleaning of the windows constituted the sub-contractor’s workmanship, i.e., the “application of skill or effort to a task”. Nothing in the policy suggested that workmanship was limited to the creation of a product.
- Third, the sub-contractor could foresee that poor workmanship on its part could damage the glass. The loss was therefore not fortuitous, and was not covered under the policy.
What is particularly interesting about Ledcor is its treatment of the “resulting damage” exception. The sub-contractor had argued that the court should distinguish between loss caused by a contractor in the scope of performing its own contract, and loss caused to work performed by other contractors. On that theory, the cost of re-cleaning the glass would be “original” damage, but the cost of replacing the windows, which had been installed by others, would be “resulting damage”. The Court rejected that distinction as artificial, since multiple contractors could “work” upon the same portion of a building. Further, the sub-contractor’s proposal would result in all damage caused by a service contractor to be resulting damage, since they would only ever work upon property installed by others. Finally, coverage under a Builders’ Risk policy would turn upon whether work was done by a single contractor or several, which would encourage insureds to slice work up into the smallest units possible and so maximize coverage.
The Court of Appeal also rejected the insurers’ initial argument at defining the scope of the “resulting damage” exception. The insurers advanced what they called the “geographic theory” – any portion of the project on which the sub-contractor was working would be excluded from coverage as “original” damage. The Court accepted that the theory was attractive in its simplicity, certainty, and predictability; however, the theory was circular, in that it failed to distinguish between original and “resulting” damage. The Court instead preferred a refinement of the “geographic” theory, which it described as the “degree of physical or systemic connectedness”. This test required consideration of:
- the extent to which damage was to a portion of the building being worked upon, including to other portions of the building that might be physically or systemically connected to it;
- the nature of the work being done, and the extent to which the damage is a “natural or foreseeable consequence” of the work; and
- the degree to which the damage was “within the purview of normal risks of poor workmanship”, or was “unexpected and fortuitous”.
Using the “degree of physical or systemic connectedness” test described in Ledcor is revolutionary. It allows courts to take into account all of the circumstances surrounding a loss to determine whether it is covered under a Builders’ Risk policy. Losses will more likely be covered where the contractor, working within the scope of his contract, could not foresee the consequences of its chosen methods, particularly upon portions of the project remote from the area and purpose of the contractor’s work. Conversely, a contractor whose contract requires it to take particular risks into account, but fails to do so, and thereby damages the building in foreseeable ways will likely not be covered.
It is striking that the Court in Ledcor acknowledges that the test it proposes imports into insurance law familiar tort principles of causation and foreseeability:
 The principle just stated reflects the proper interpretation of this wording of the insurance policy. The presumptive test is that damage which is physically or systemically connected to the very work being carried on is not covered. Whether coverage is nevertheless extended under that test in the factual context of any particular case will depend on the consideration of the factors listed above (supra, para. 50). Those factors all engage elements of “causation” and “foreseeability”, concepts which are well known in the common law, when applying the policy wording to particular factual situations. The presumptive test stated above reflects the proper interpretation of the policy, but these collateral factors will come into play in applying the policy wording to particular factual situations, especially in extreme cases.
In summary, emphasizing fortuity, as in both Canadian National Railway and Ledcor, allows the courts to escape the arid and often unworkable distinctions between external and internal defects; between design and workmanship (i.e., where such functions are inherently blended within one party’s work, or turn upon the contributions of multiple parties on a project); and between on-site and off-site conduct. A sub-contractor that relies on others’ instructions likely would not be expected to anticipate that work performed in accordance with those instructions would result in damage, so the loss will more likely be fortuitous, regardless of whether the instructions came from a head contractor or an engineer, or whether the instructions were prepared off-site or delivered verbally on-site. The source of the loss is “external” to the sub-contractor. Conversely, a contractor that selects its own materials and methods may be expected to anticipate the possible risks arising from those choices, and the resulting loss may not be fortuitous.
VIII. NON-FORTUITOUS LOSS EXCLUSIONS:
The renewed emphasis on fortuity has naturally had less of an impact on a second group of exclusions, all of which address non-fortuitous losses, i.e., losses caused by factors that are inherent in the insured property or will inevitably occur. We review several of the most notable examples below.
A. THE “WEAR AND TEAR” EXCLUSION:
IBC Form 51208 (Builders’ Risk Broad Form) stipulates, in exclusion (1), that there is no indemnity for loss attributable to:
…wear and tear, gradual deterioration, normal upkeep, latent defect, or normal making good.
Although an integral part of the All Risk policy for many years, few cases have considered the scope of the “wear and tear” exclusion. “Wear and tear” is actually a group of different exclusions, each of which is characterized by one factor: inevitability.
The Builders’ Risk policy, as with most types of insurance, requires fortuity in the occurrence of the loss. In Glassner et al v. Detroit Fire and Marine Insurance Company the Wisconsin Supreme Court stated:
An `all risk’ policy is a promise to pay for loss caused by fortuitous and extraneous happening, but it is not a promise to pay for loss or damage which is almost certain to happen because of the nature and inherent qualities of the property Insured.
Viewed in that light, losses falling within the scope of the exclusion for “wear and tear” are not risks at all, but rather inevitable consequences. Consensus that these types of losses are not to be covered explains why there are so very few decided cases concerning the precise scope of the exclusion.
One decision that does provide some guidance in this area is Cyclops Corp. v. The Home Insurance Company, where the Court stated:
[C]onstruing the words ‘wear and tear’ in their everyday common usage, we are convinced that the words ‘wear and tear’ mean simply and solely that ordinary and natural deterioration or abrasion which an object experiences by its expected contact between its component parts and outside objects during the period of its natural life expectancy.
When the insurer does rely upon the “wear and tear” exclusion, the insured will argue that the deterioration is remarkable or extraordinary, thereby removing the loss from the intended scope of the exclusion. Illustrative of this argument is the decision in Potomac Power Co. v. Arkwright-Boston Manufacturers Mutual Insurance Co., albeit in connection with a boiler and machinery policy. The insured sought indemnity for costs in correcting damage which it alleged was caused by excessive and abnormal abrasion of a generator coil. On a preliminary application the Court stated:
An excessive degree of abrasion does not necessarily preclude a finding of wear and tear. The magnitude of the abrasion found merely manifests the effect of a process – perhaps beginning as `ordinary’ abrasion – that intensified over time. The fact that the extent of abrasion was unusually severe does not demonstrate that the abrasion process itself was extraordinary. It might well have been the culmination of a process which commenced long before.
Other exclusions related to “wear and tear” include exclusions for losses arising from rust and corrosion, which are held to be natural processes, rather than fortuitous losses.
B. THE “INHERENT VICE” EXCLUSION
“Inherent vice” generally involves internal decomposition or some quality which brings about an object’s own damage or destruction. One of the few Canadian cases to consider the term in the context of an “all risk” policy was Brown Fraser and Company Limited v. Indemnity Marine Assurance Company, a decision of the British Columbia Court of Appeal. The collapse of a boom crane was found to be attributable to the failure of the insured’s employees to install certain steel pins. The Court rejected the application of the “inherent vice” exclusion, indicating that it entailed “that which is necessarily incidental to the property rather than occasioned by an adventitious cause. The negligence of the insured’s employees constituted an “adventitious cause” sufficient to oust the operation of the exclusion.
The American case of State Farm Fire & Casualty Co. v. Volding also illustrates the scope of the exclusion. The insured incurred a loss when rainwater entered pores in a chimney’s bricks, causing the bricks to crack and ultimately separate from one another. The evidence at trial indicated that the bricks were porous and not suitable for use on an exterior wall or when exposed to the elements. That failing was characterized as “inherent vice” and indemnity was refused.
More recently, in Dawson Creek (City) v. Zurich Insurance Co., the courts concluded that the failure of an arena’s roof after a heavy snowfall was not the result of inherent vice or a latent defect. The roof’s original construction was held to have been faulty, with inadequate bracing, and the roof had been weakened by a previous snowfall several years earlier, but the ultimate collapse was not inevitable, and exclusions for inherent vice and latent defect did not apply.
The “inherent vice” exclusion simply restates again the need to demonstrate that a loss is truly fortuitous. Note that the loss in Volding resulted from a combination of an “external” factor (the rainwater) and an “internal” condition (large pores). The loss was not covered, not because of a purely “internal” defect, but because it should likely have been appreciated by the insured that the bricks were not appropriate for installation in that location.
C. THE “LATENT DEFECT” EXCLUSION:
The “latent defect” exclusion is intended to preclude indemnity if the loss or damage to the insurable property is attributable to a defect which could not be discovered by known or customary tests. Not unlike the exclusion for “inherent vice”, the purpose of the exclusion is to ensure that the insurer does not become a guarantor of the quality or fitness of the insured’s property. That function is performed through enforcement of the contractual obligations which the supplier of the property owes to the insured, including those obligations pursuant to the Sale of Goods Act. Usually the latent defect will be a result of a fault in the manufacture of the product, in which case the insured can claim against the manufacturer or supplier.
The courts have always treated this exclusion narrowly by taking a more stringent approach to the question of reasonable diligence. Latent defect is treated as that which could not have been detected through reasonable diligence. What the courts consider to be “reasonable diligence” has changed over the decades, with advances in technology and the development of new building materials, but an insurer relying upon the exclusion must show that proper scientific inquiries would not have detected the defect before the loss. That test is extremely onerous.
Illustrative of this trend is General American Transportation Corp. v. Sun Insurance Office Ltd. The insured was building an underground silo in which to test jet engines. A temporary platform, used to construct the silo, collapsed as the result of a defective weld. The insurer relied upon the latent defect exclusion. That argument was rejected since evidence led at trial suggested that the defect in the weld could have been detected by means of proper inspection, including radiography; if the defect could have been detected, it could have been remedied, and thus was not inevitable or inherent.
This approach is used even if, in fact, scientific inquiries would have been improbable and unrealistic in the circumstances. For example, in Mattis v. State Farm Fire & Casualty Co. an insured homeowner commenced action on a All Risk policy when cracks appeared in his basement walls. The Court rejected the insurer’s reliance on the latent defect exclusion by concluding that scientific inquiry could have detected the deficiency before the damage became manifest. It did not matter that the insured, in all likelihood, would never have undertaken any such inquiry.
It is generally recognized that “latent defect” does not embrace “design defect”. In Mattis, it was indicated that “latent defect” was limited in its scope to any defects in the materials used for construction and did not include matters of design or construction. What limited authority exists within Canada would support that view. In Maclab Enterprises Ltd. (supra), the architect recommended the implementation of an improper fastening mechanism for the installation of a brick wall and suggested the use of inadequate mortar. That case involved an All Risk policy that did not include a “design defect” but did include a “latent defect” exclusion. It was argued by the insurer that what had originated as a “design defect” was transformed into a “latent defect” upon completion of construction. In rejecting that submission the Alberta court adopted with approval comments, from an English case, Jackson v. Mumford, which held that:
… latent defects in the machinery [did not] cover the erroneous judgment of the designer as to the effect of the strain which his machinery would have to resist – the machinery itself being faultless, the workmanship faultless and the construction precisely that which the designer intended it to be.
As a general principle, therefore, “design defects” do not constitute “latent defects” as those terms are utilized in an Builders’ Risk policy.
D. THE “MECHANICAL BREAKDOWN” EXCLUSION:
IBC Form 51208, by reason of exclusion (d), bars indemnity for:
…loss or damage caused directly or indirectly by mechanical or electrical breakdown or derangement provided however, to the extent otherwise insured and not otherwise excluded under this Policy, resultant damage to the property shall be insured.
Like the “latent defect” exclusion, the “mechanical breakdown” exclusion is intended by the insurer to avoid coverage for property damage that an insured may claim in contract and/or negligence against the manufacturer or supplier of the product.
The exclusion for “mechanical breakdown” is not always present in “all risk” policies and few cases provide any meaningful analysis of its scope. Most of the cases conclude that “mechanical breakdown” has occurred without much analysis as to the scope of the exclusion: see Sterling Crane, A Division of Procor Ltd. v. Penner Bros. Utilities Ltd. et al, and Hard Rock Construction Limited v. Sun Alliance Insurance Company Limited.
However, there is one instructive authority. In Brown Fraser and Company Limited (supra) a crane had been damaged as a consequence of the failure of the insured’s employees to properly install metal pins. The Court concluded that the “mechanical breakdown” exclusion did not apply because the error resulted from the assembly of the crane rather than its original manufacture. The Court stated:
Mechanical breakdown, it seems to me, must be interpreted in the circumstances of this case and in its context as a failure in operation due to some mechanical defect in some part or parts of the equipment when properly assembled to constitute a crane. Here the failure to operate was due to negligence in assembling the machine so that it could function as an operating unit. In other words there was a failure to function, not due to any mechanical defect, but due to a failure to insert a part or parts in the machine which ought to have been inserted to make it a complete operating unit. Without these parts the machine could not function. It was the absence of these necessary parts that caused the boom to fall. There was an operating failure not due to any mechanical defect but to negligence.
This analysis, which limits the scope of the “mechanical breakdown” exclusion, was affirmed in MacLab Enterprises Ltd. (supra).
E. THE “MYSTERIOUS DISAPPEARANCE” EXCLUSION:
The rationale of the insurance industry in issuing “all risks” coverage is to provide indemnity in cases where there may be some uncertainty as to the manner in which the loss arose and wherein the insured would have difficulty demonstrating that the loss fell within any specified peril. In light of the industry’s rationale in issuing such coverage, it is somewhat surprising that policies would exclude coverage in circumstances involving a “mysterious disappearance”.
There are variations in the exclusion for “mysterious disappearance”. For Example, IBC Form 51208 excludes indemnity for:
… any loss or shortage disclosed on taking inventory or making appraisal, or any mysterious disappearance.
Another commonly utilized manuscript wording available in Canada contains an exclusion for:
…mysterious disappearance of property (except property in the custody of carriers or bailees for hire) or shortage disclosed by taking inventory.
While continually broadening the scope of most other exclusions, the courts have moved cautiously in allowing insurers to rely upon the “mysterious disappearance” exclusion. This deference to the insured’s plight recognizes that to do otherwise would defeat the broad protective rationale for “all risk” policies. Noteworthy in this respect is the case of Boston Insurance Company and Friedman, from the Supreme Court of Pennsylvania. The facts are novel. The insured was a jeweller who accepted a ring on consignment. He then consigned the ring to a second jeweller who, in turn, consigned it to a third jeweller who, ultimately, was found dead without the ring. The insurer relied upon a “mysterious disappearance” exclusion worded in the following terms:
… all risk of loss or damage … arising from any cause whatsoever except: …(M) Unexplained loss, mysterious disappearance or loss or damage disclosed on taking inventory.
The insured said that the loss of the ring was obviously connected with the jeweller’s death. The insurer denied coverage, contending that the insured had to provide a more precise explanation of the circumstances of the loss. The Court concluded, firstly, that the insured need only prove the fact of a loss and not more and, secondly, that the insurer had the burden of demonstrating that the loss fell squarely within the exclusion. The Court stated:
It is axiomatic that [the insured] must show that the loss falls within the risks insured against, but it is also axiomatic, that it is for [the insurer] to show that the loss was not due to one of the risks insured against but rather to an excepted cause. It would seem that all [the insured] need show in such a case is a loss, since losses from all causes are covered. [The insurer], arguing that a mysterious disappearance is “any disappearance the circumstances of which excite – and at the same time baffle – wonder or curiosity’ “attempts to distinguish between the classic cases of loss or misplaced property, and a case which is baffling and therefore a mysterious disappearance. Assuming that it has proved its point, at least in the first instance, [the insurer] argues that [the insured] was therefore under the obligation to go forward and prove a theft, and, having failed to do so, cannot recover. As can be seen, [the insurer] relies to a large extent on semantics. Under this theory, any loss, the exact cause of which could not be proved by at least a preponderance of the evidence, would automatically be classed as a mysterious disappearance, and recovery would be defeated unless [the insured] could prove a theft, embezzlement, or some other specific cause.
The Court adopted the views of an earlier decision, Chase Rand Corporation v. Central Insurance Co. of Baltimore, which held:
The insured’s sole obligation was to furnish the insurer with such explanation, as it, in good faith, received and accepted concerning the time and cause of the loss, and this it has done. If plaintiff were required to go further the inclusive character of the coverage of the insurance policy would be a delusion and a snare.
It can therefore be said that the “mysterious disappearance” exclusion, being antithetical to the concept of the “all risk” policy, has been severely limited in its scope. This result has been achieved by shifting the evidentiary burden onto the shoulders of the insurer. The insured’s obligation is limited solely to proffering the explanation which it, in good faith, received and accepted.
It has long been recognized that the Builders’ Risk wording neither covers all losses nor responds to all risks. Several decades of judicial history further and systematically narrowed the true scope of the Builders’ Risk policy, undermining its role in the marketplace, all because of judicial attempts to resolve the conceptual challenges arising from the application of the standard policy wording to complex construction projects.
So, for example, the “faulty workmanship” exclusion was interpreted to apply in cases involving on-site carelessness. Further uncertainty arose from the courts’ willingness to treat design consultants’ activities on the site as being within the scope of the “faulty workmanship” exclusion, and to exclude losses for “defective design” merely because the design failed (despite being “state of the art”). The “resulting damage” exception was confined to loss incurred by parties other than the insured and only if the loss does not entail property which is functionally integrated into the on-site improvement.
However, in more recent years the pendulum appears to be swinging back, towards interpretations of the Builders’ Risk coverage that properly account for its purpose and original intent. By emphasizing that losses must be fortuitous, the courts are escaping conceptual traps and arid distinctions created by previous interpretations of such policies. Canadian courts are explicitly drawing upon tort principles of foreseeability and causation to assist them in analyzing the circumstances of specific cases. The irony is that the retreat from arbitrary rules in favour of more nuanced case-specific analyses of fortuity will likely produce greater predictability of outcomes, consistent with the purpose of the coverage in the first place.