Liability Insurance Coverage for Construction Defect Claims[dagger]

Liability Insurance Coverage for Construction Defect Claims[dagger]

Shusterman, Richard M



Lawsuits over defects in new construction are very common. That litigation in turn frequently gives rise to disputes regarding liability insurance coverage for the alleged defects. The following scenario is typical. A builder/developer undertakes to build a condominium or housing development. The builder enters into a contract with a general contractor, which in turn enters into contracts with subcontractors. Very frequently, these contracts provide that the contractor in its contract with the builder, and the subcontractors in their contracts with the general contractor, will indemnify, respectively, the builder and general contractor for liability in connection with the project. Further, they provide that the subcontractors will also see to it that the builder and general are added to their liability insurance policies as additional insureds. In addition, the contractor or subcontractors may purchase products that are incorporated into the structure. Thus, another group involved in construction projects are the suppliers of those products. This article does not focus on coverage for products liability claims, which is a separate subject in its own right.1

Then, after the project is built, the new homeowners discover what they contend are defects or deficiencies in the construction of the buildings or in the products supplied and incorporated. Frequently they also allege that the defects have caused damage to the property. Sometimes these claims include allegations of bodily injury, for example, when construction defects allow intrusion of rain water into the homes, resulting in bodily injury from exposure to mold growth. As a result, the underlying litigation may include claims for both property damage and bodily injury.

Coverage disputes arising out of these underlying claims and suits raise a number of issues. The type of policy most often involved in these claims is the commercial general liability (“CGL”) policy.2 This article focuses on the issues most typically associated with coverage for construction defect claims under CGL policies. They are the following:

(1) CGL policies cover bodily injury or property damage caused by an “occurrence.” An issue that often arises is whether faulty or defective construction constitutes an occurrence.

(2) CGL policies provide coverage for property damage that occurs during the policy period. Construction defect claims often involve questions of when the damage took place and therefore which policies may potentially have to respond. Thus, trigger of coverage can be an important issue in construction defect coverage disputes.

(3) A number of exclusions, including especially the “business risk” exclusions, give effect to the general purpose of CGL policies to provide coverage only for damage to property other than the insured’s work or product and to bar coverage for purely economic damages resulting from the insured’s breach of contract, or failure to perform the contract as specified. These “business risk” exclusions are an important part of any construction defect coverage analysis.

(4) Construction defect claims sometimes involve bodily injury claims arising from exposure to mold and allegedly toxic building materials. An issue that arises in coverage disputes is the applicability of the pollution exclusion to such claims.

(5) CGL policies impose a number of obligations on insureds as conditions to coverage. These include the obligation to give notice of claims and suits and to cooperate. In addition, the policies do not provide coverage for the insured’s voluntary payments or payments made prior to notice of the claim being given to the insurer.

(6) As noted above, construction contracts often require inclusion of the builder or general contractor as an additional insured to the other party’s CGL policy. The inclusion of a party as an additional insured in another party’s liability policy raises issues under the “other insurance” clauses of the parties’ policies.

(7) A related issue arises when an occurrence causes injury or damage during a number of years and, as a result, the court rules that the policies in effect throughout such period of time are triggered. This gives rise to issues of “allocation,” i.e., which policies will in fact be required to respond.

(8) An issue of potential significance in many cases is allocation of defense costs to covered and non-covered claims. When an insurer defends a suit under a reservation of rights and it later turns out that there was no coverage for some of the claims being defended, the insurer may have the right to be reimbursed for the cost of defending the non-covered claims.

These issues are discussed in the following sections.



A. The Insurer’s Obligations Under The Insuring Agreement

The first step in any coverage analysis is determining whether the underlying claim or suit comes within the terms of the insuring agreement of the policy. Modern CGL policies typically provide the following insuring agreement:

We [the insurer] will pay those sums that the insured becomes legally obligated to pay as damages because of “bodily injury” or “property damage” to which this insurance applies. We will have the right and duty to defend the insured against any “suit” seeking those damages. However, we will have no duty to defend the insured against any “suit” seeking damages for “bodily injury” or “property damage” to which this insurance does not apply. We may, at our discretion, investigate any “occurrence” and settle any claim or “suit” that may result.3

The insuring agreement then provides that the policy applies to bodily injury and property damage only if

(1) The “bodily injury” or “property damage” is caused by an “occurrence” that takes place in the “coverage territory”; and

(2) The “bodily injury” or “property damage” occurs during the policy period.4

The standard occurrence-based policy thus obligates the insurer to defend and indemnify the insured only if there was bodily injury or property damage, as defined, caused by an “occurrence.” These terms are analyzed below.

B. Property Damage (

CGL policies generally define “property damage” as

a. Physical injury to tangible property, including all resulting loss of use of that property. All such loss of use shall be deemed to occur at the time of the physical injury that caused it; or

b. Loss of use of tangible property that is not physically injured. All such loss of use shall be deemed to occur at the time of the “occurrence” that caused it.5

“Bodily injury” is defined as “bodily injury, sickness or disease sustained by a person, ‘ including death resulting from any of these at any time.”6

The definition of “property damage” is limited to physical injury to tangible property and loss of use of tangible property. Not included are losses to intangible property interests, such as economic losses. As we discuss below, in the construction defect context, while the insured builder or contractor may be liable for a broad array of defective conditions, including “property damage” as defined, their insurers are not responsible for loss that is purely economic resulting from a breach of contract or warranty that does not also give rise to “property damage.” As some have noted, “[g]eneral liability policies … are not designed to provide contractors and developers with coverage against claims their work is inferior or defective.”7

C. Faulty or Defective Construction, By Itself, Not an Occurrence

The term “occurrence” is defined in a CGL policy as “an accident, including continuous or repeated exposure to substantially the same general harmful conditions.”8 cases from numerous jurisdictions state the general proposition that faulty construction, by itself, which does not cause at least some “property damage,” is not an “occurrence” and hence is not covered.9 This rule is based on the idea that the purpose of liability insurance is to provide protection for bodily injury or property damage caused by the insured’s product or work, but not for the repair or replacement of defective or flawed products or work that do not comply with contractual requirements.10 As stated by Dean Henderson, “[t]he coverage is for tort liability for physical damages to others and not for contractual liability of the insured for economic loss because the product or completed work is not that for which the damaged person bargained.”11 Generally, however, an allegation of unintended damage to property other than the insured’s work or product has been held sufficient to constitute an alleged “occurrence.”12

Three recent cases from state supreme courts illustrate the different (and conflicting) approaches that courts have taken in analyzing whether faulty construction can be an occurrence.13 In American Family Mutual Insurance Co. v. American Girl, Inc.,14 a soil engineering subcontractor gave faulty site-preparation advice to the general contractor in connection with the construction of a warehouse. “As a result, there was excessive settlement of the soil after the building was completed, causing the building’s foundation to sink. This caused the rest of the structure to buckle and crack. Ultimately, the building was declared unsafe and had to be torn down.”15 In the resulting arbitration between the owner and the insured contractor, the owner asserted solely breach of contract and breach of warranty claims based on the contractor’s express warranty relating to the settling of the structure caused by subsoil conditions that the parties knew beforehand were unstable and that had to be corrected as part of the construction.16

The court first held that the sinking, buckling and cracking of the building caused by the soil settlement was property damage, defined in the policy as “physical injury to tangible property.”17 It then rejected the insurer’s argument that the owner’s claim for breach of contract and warranty was only for economic loss rather than for property damage and was therefore not covered.18 The court characterized the economic loss doctrine as precluding recovery in tort for economic losses resulting from the failure of a product to live up to contractual expectations:

The economic loss doctrine is based on an understanding that contract law and the law of warranty, in particular, is better suited than tort law for dealing with purely economic loss in the commercial arena…. The economic loss doctrine operates to restrict contracting parties to contract rather than tort remedies for recovery of economic losses associated with the contract relationship. . . . The economic loss doctrine is a remedies principle. It determines how a loss can be recovered – in tort or in contract/warranty law. It does not determine whether an insurance policy covers a claim, which depends instead upon the policy language.19

As a result, the court reasoned that in some circumstances, a CGL policy may provide coverage for breach of contract or warranty claims. According to the court, the present case was such a circumstance.

It then held that the property damage was the result of an occurrence:

No one seriously contends that the property damage to the [warehouse] was anything but accidental (it was clearly not intentional), nor does anyone argue that it was anticipated by the parties. The damage to the [warehouse] occurred as a result of the continuous, substantial, and harmful settlement of the soil underneath the building. Lawson’s inadequate site-preparation advice was a cause of this exposure to harm. Neither the cause nor the harm was intended, anticipated, or expected.20

The court then rejected the insurer’s argument that because the building owner’s claim was for breach of contract and breach of warranty, the loss could not be an occurrence, because the CGL policy is not intended to cover contract claims arising out of the insured’s defective work or product. It said that while it was true that CGL policies generally do not cover contract claims arising out of the insured’s defective work or product, that was because of the policies’ business risk exclusions.21 It was not, according to the court, because a loss actionable only in contract could never be the result of an occurrence within the meaning of the CGL policy’s grant of coverage.22 The court therefore concluded that, because there was nothing in the policy’s grant of coverage that barred coverage for a claim only because it was in contract rather than tort, the property damage to the building was an occurrence.23

The court then went on to rule that, although the “your work” exclusion would bar coverage, the exception for work by a subcontractor on behalf of the insured prevented application of the exclusion.24 In sum, American Girl concludes that a breach of contract claim may qualify as an occurrence causing property damage, even if only to the insured’s own work and that coverage under these circumstances will be barred, if at all, only by the “your work” exclusion.25

Auto-Owners Insurance Co. v. Home Pride Cos.26 took a different approach. There, the owner of apartment buildings entered into a contract with JT Builders to install new shingles on a number of the owner’s buildings. JT subcontracted the work to Home Pride, which in turn subcontracted the work to Hansen. After the work was done, the owner noticed problems with the roof. In the ensuing litigation, the owner alleged that the shingles had not been installed in a workmanlike manner and that the faulty workmanship had caused damage to the roof structures and building. In analyzing the issue of whether faulty workmanship constitutes an occurrence or accident under CGL policies, the court stated that a “majority of courts have determined that faulty workmanship is not an accident and, therefore, not an occurrence.”27 It also noted the “relatively small number of courts” that have determined the damage that occurs as a result of faulty or negligent workmanship constitutes an accident, so long as the insured did not intend for the damage to occur.28

The court then went on to analyze the justifications for the general rule that faulty workmanship, standing alone, is not covered under a CGL policy. One justification has been “public policy,” on the notion that the cost to repair and replace damage caused by faulty workmanship is a business risk not covered under a CGL policy. As noted by the court, the business risk rule is part of CGL policies in the form of the “your work” exclusion. Therefore, the business risk rule does not serve as an initial bar to coverage, but as a potential exclusion.

The court then described the second justification for the rule barring coverage for faulty workmanship: what it described as the “majority rule” that faulty workmanship, standing alone, is not covered because, as a matter of policy interpretation, the fortuity implied by reference to accident or exposure to conditions is not what is commonly meant by a failure of workmanship.29 The court went on to hold that because the “majority rule” is based on an actual interpretation of the policy language, as opposed to a “mere exposition of policy,” which comported with the court’s prior definitions of the term “accident,” it concluded that it represented the better rule.30 Consequently, faulty workmanship, standing alone, is not covered under a standard CGL policy because it is not a fortuitous event. The court then went on to rule that because the apartment building owner had alleged damage to the buildings as a result of the faulty roofing work, the owner had alleged an occurrence.

In L-J, Inc. v. Bituminous Fire & Marine Insurance Co.,31 the court on very similar facts took an approach directly opposite to that of American Girl. In L-J, the insured contracted to construct roads in a residential development. The roads deteriorated and failed because the underlying soil had not been properly prepared, including failure to remove tree stumps from the subgrade and insufficient compaction of the soft, wet clay soils. These conditions caused drainage problems, which in turn caused seepage of rain and other surface water, resulting in deterioration of the pavement.

While conceding that the cracking of the road surface may have constituted property damage, the court held that there was no occurrence. The only “occurrences” were the negligent acts of the contractor during road design, preparation and construction that led to the premature deterioration of the road. These negligent acts were “faulty workmanship causing damage to the roadway system only, which does not fall within the contractual definition of Occurrence’ under [the]. . . policy.”32 The court, in rejecting the argument of the court of appeals that the damage to the road caused by continuous exposure to surface water runoff was an occurrence (and reversing), noted that “the sole cause of the deterioration was the Contractor’s faulty workmanship in designing and constructing the road system. That the roads were subject to surface water damage was not an ‘accident’ as the court of appeals held. Rather, the damage was caused by the Contractor’s negligently designed drainage system to handle the water runoff and failure to properly compact the road’s subgrade.”33

An interesting illustration of how a court dealt with the distinction between damage that resulted from an occurrence and damage that did not is Auto Owners Insurance Co. v. Tripp Construction,34 in which the insured, Tripp, was a general contractor sued in a class action by numerous homeowners for construction defects in their homes. The trial court had bifurcated the class action into two phases. The first phase dealt with claims for damages for repairs of the actual defects in the construction of the homes. The second phase dealt with claims for damages to the property caused by the construction defects. The appellate court held that the insurer had no duty to defend as to phase one, but did owe a defense as to phase two.

D. Number of Occurrences

Once a court determines that there is an occurrence, it is often asked to consider how many occurrences may be involved. The issue might arise in the context of a self-insured retention or deductible that applies on a per occurrence basis or where a policy has a per occurrence limit but no aggregate limit (or an aggregate limit that is greater than the per occurrence limit). The overwhelming majority rule is that the number of occurrences is determined by focusing on the cause or causes of injury and not the effects, such as the claims or injuries.35 However, application of this rule can produce drastically different resuits, depending on what the court finds is the underlying cause.36 Moreover, notwithstanding the general rule, some courts appear to decide their cases in such a way as to maximize coverage. In such instances, where there is a large number of relatively small individual claims and the policy contains a per occurrence deductible that exceeds the amount of the claim, some courts tend to find a single occurrence, to avoid the limiting effect of the deductible. On the other hand, if the case involves per occurrence liability limits, some courts find multiple occurrences. Two frequently cited cases illustrate how application of the same general rule can give different results concerning coverage.

In Champion International Corp. v. Continental Casualty Co.,” Champion sold vinylcovered paneling to numerous manufacturers of house boats, house trailers, motor homes and campers. Not long after installation, the panels began to split, causing damage to the vehicles in which they had been installed. Champion’s policies had a $5,000 per occurrence deductible. Continental Casualty argued that there were 1,400 separate occurrences corresponding to each of the damaged vehicles in which the panels had been installed. If this argument had been accepted, there would have been no coverage, as the cost of repair in each case was less than the $5,000 deductible. The court held that there was one occurrence, Champion’s delivery of the defective panels. As a result, there was insurance coverage in the amount of $1.1 million to be applied against Champion’s $1.6 million total loss.

On the other hand, in Maurice Pincoffs Co. v. St. Paul Fire & Marine Insurance Co.,38 the insured, Pincoffs, was a wholesaler of birdseed. It imported canary seed that was sold, in its original bags, to eight different dealers. The seed was contaminated with a chemical insecticide toxic to birds and many birds were killed. The owners of the poisoned birds made claims against the dealers who in turn made claims against Pincoffs. Pincoffs had two liability insurance policies. The primary policy had a single occurrence limit of $50,000 and an aggregate limit of $100,000. There was also an umbrella liability policy above that. The primary carrier, St. Paul, settled with claimants for a total of $50,000. There remained outstanding claims in excess of another $50,000. St. Paul took the position that its policy limits were exhausted, as there was only a single occurrence, based on the contamination of the seed. In contrast, the umbrella carrier took the position that there were multiple occurrences and that its policy did not take over until St. Paul had paid a total of $100,000. The court of appeals held that the number of occurrences should be determined by the insured’s action that gave rise to its liability. In this case, the court reasoned that it was the sale of the contaminated seed for which Pincoffs was liable. Pincoffs received the seed in a contaminated condition and did not itself contaminate the seed. Therefore, the court concluded that it was not the act of contaminating the seed that subjected Pincoffs to liability but rather its sale. According to the court of appeals, each of the eight sales made by Pincoffs to eight different dealers was a separate occurrence.



The phrase “trigger of coverage” is a shorthand expression for identifying what must occur during a policy period in order to implicate a particular policy. As noted above, CGL policies state that the property damage or bodily injury must occur during the policy period in order for coverage to be implicated. Courts generally hold this to mean that the time injury or damage occurs is not the time the wrongful act is committed but the time the complaining party is actually damaged. This rule has been relatively easy to apply in cases of abrupt or discrete damage-causing events. For example, in Millers Mutual Fire Insurance Co. v. Ed Bailey, Inc. ,39 the policyholder sold and installed foam insulation. During the policy period, the policyholder installed insulation at a building site. After the policy had expired, there was a fire at the site. The owners sued the policyholder based on the allegedly defective insulation and the policyholder sought coverage. The court held that there was no coverage, as the time of occurrence of an accident is not the time the wrongful act was committed but the time the complaining party was actually damaged.40

Courts have adopted a number of tests for determining when a coverage-triggering event occurred, and hence, which policies on the risk may have to respond. This is because the definition of an occurrence as “an accident, including continuous or repeated exposure to substantially the same general harmful conditions,”41 allows for coverage for damage resulting from events or conditions unfolding over time. In addition, it is because of the difficulty in some situations of determining when injury or damage took place in cases of long-term latent exposure or damage.

(1) One such approach is the “injury-in-fact” theory of coverage. This approach is the most true to the literal policy language, as it mandates that only those policies in effect when the damage occurs are triggered. This theory can result in the triggering of more than one year, but based on proof of the actual injury or damage sustained during each period.

(2) A second alternative is the “first discovery” or “manifestation” trigger, under which the policy in effect at the time the property damage manifests itself or is discovered provides coverage.

(3) A third approach is the “exposure” theory. Thereunder, only those policies in effect at the time of actual exposure to the damage-causing substance or event would provide coverage. It should be noted, however, that exposure here really means “injurious exposure,” as it is still necessary that there have been injury or damage during the policy period.

(4) Some courts employ a “continuous” trigger approach, under which all policies in effect over a span of time, beginning from the first exposure to injurious conditions, continuing through any period of latency while the resulting damage remains undiscovered and is progressing, and ending at the time the injury manifests itself to the insured, are implicated.42

Much recent trigger law has developed in the context of long-term exposure cases, such as asbestos bodily injury and property damage and environmental coverage claims.

However, courts do not routinely apply a multiple trigger in construction defect and similar cases, as the problem of identifying the time of damage is not necessarily present. If the court can actually identify when the injury or damage occurred, it is more likely to apply an injury-in-fact trigger.43

Also, progressive damage will not necessarily result in application of a multiple-year trigger if that damage is not the subject of the claim. Thus, in Greenlee v. Sherman,44 a fire in a home in 1984 was caused by improper installation in 1980 of a flue pipe from a furnace, which caused the exposure of a wooden joist to intense heat from the furnace.45 This exposure progressively lowered the ignition temperature of the joist to the point where it was ignited. The claim was for the damage to the house, which was immediate, and not for damage to the joist, which had deteriorated slowly. The court held that only the policy in effect at the time of the fire was triggered and not the policies in effect during the progressive weakening of the joist.

Where, on the other hand, the date of the injury is unknown or difficult to identify, or the injury can be said to have continued over multiple policy periods, courts are more likely to apply a manifestation or a continuous trigger of coverage. West American Insurance Co. v. Lindepuu46 involved defective doors and windows in a residential development installed between 1987 and 1989. The homeowners alleged that they did not discover the defects until 1993. The policyholder argued that the relevant policies were those in effect when the windows and doors were installed. The insurer countered that the policy in effect when the defects first became apparent should respond. Rejecting the “continuous trigger” adopted in J.H. France,41 the court concluded that the damage allegedly caused by the defective doors and windows presumably produced injury when the homeowners discovered the damage in 1993.48

In United States Fidelity & Guaranty Co. v. American Insurance Co.,49 the insured was sued for damage resulting from the installation of defective bricks which spalled, or chipped and flaked, causing the brick surface to break off. This spalling took place during the periods of three insurers’ policies over a three-year period. In ruling on the trigger of coverage issue, the court applied the general rule that “[t]he time of the occurrence of an accident within the meaning of an indemnity policy is not the time the wrongful act was committed but the time when the complaining party was actually damaged.”50 It went on to hold that, if the initial spalling was notice of possible defectiveness of the entire structure, then all damage would accrue in the policy period when the spalling first becomes apparent.

Even under a continuous trigger, it is still necessary to determine when the damage actually began. Aetna Casualty & Surety Co. v. Ply Gem Industries,^ involved claims for property damage to homes resulting from the use of the insured’s fire retardant plywood, which, because of the fire retardant chemicals, deteriorated over time and lost structural integrity. The court held that the continuous trigger would not automatically apply to trigger the policies in force at the time of installation of the defective plywood, as there was no evidence of when the deterioration actually began.

Also, Century Indemnity Co. v. Golden Hills Builders52 applied a “modified continuous trigger” to hold that “coverage is triggered whenever the damage can be shown in fact to have first occurred, even if it is before the damage became apparent, and the policy in effect at the time of the injury-in-fact covers all the ensuing damages. . . . Coverage is also triggered under every policy applicable thereafter.”53



If it is determined that a claim involves property damage or bodily injury caused by an occurrence during the policy period, so that the claim comes within the terms of the policy’s insuring agreement, the next step is to determine whether any policy exclusions apply to bar coverage for injury or damage that would otherwise be covered. CGL policies contain a number of exclusions that often apply to construction defect claims.

A. Expected or Intended Injury Exclusion

CGL policies typically contain the following provision:

This insurance does not apply to … “[b]odily injury” or “property damage” expected or intended from the standpoint of the insured. This exclusion does not apply to “bodily injury” resulting from the use of reasonable force to protect persons or property.54

Courts have often limited this exclusion to cases in which the insured not only expected or intended his conduct, but also expected or intended the resulting harm. In other words, it is not the defective condition or workmanship that must be expected or intended for coverage to be barred, but instead the resulting consequential bodily injury or damage to property other than the insured’s work or product.55

B. Contractual Liability Exclusion

A contractual liability exclusion bars coverage for liability assumed by the insured under an agreement, typically stating that coverage is excluded for

“[bjodily injury” or “property damage” for which the insured is obligated to pay damages by reason of the assumption of liability in a contract or agreement. This exclusion does not apply to liability to damages:

(1) That the insured would have in the absence of the contract or agreement; or

(2) Assumed in a contract or agreement that is an “insured contract,” provided the “bodily injury” or “property damage” occurs subsequent to the execution of the contract or agreement.56

The term “insured contract” is defined to include

[t]hat part of any other contract or agreement pertaining to your business (including an indemnification of a municipality in connection with work performed for a municipality) under which you assume the tort liability of another party to pay for “bodily injury” or “property damage” to a third person or organization. Tort liability means a liability that would be imposed by law in the absence of any contract or agreement.57

In the construction defect context, often this exclusion will not apply because of the limitations in the two exceptions. As explained in Daily Express, Inc. v. Northern Neck Transfer Corp.:58

[A] general principle may be deduced that a provision . . . specifically excluding from coverage liability assumed by the insured under a contract not defined in the policy is operative in the sense that it relieves the insurer of liability otherwise existing under the policy only in situations where the insured would not be liable to a third party except for the fact that he assumed liability under an express agreement with such party … In other words, where the express contract actually adds nothing to the insured’s liability, the contractual liability exclusion clause is not applicable, but where the insured’s liability would not exist except for the express contract, the contractual liability clause relieves the insurer of liability.59

In construction defect cases, parties such as subcontractors usually would have potential liability in tort to the plaintiff homeowners based on their own alleged negligence. Also, their indemnification agreements will often come within the exception for insured contracts. As a result, this exclusion may not be an important factor in many construction defect coverage cases.

C. Damage to Property Exclusion

This exclusion (identified as Exclusion j. in recent policies) bars coverage for property damage to:

1. Property you own, rent, or occupy;

2. Premises you sell, give away, or abandon, if the “property damage” arises out of any part of those premises;

3. Property loaned to you;

4. Personal property in the care, custody, or control of the insured;

5. That particular part of real property on which you or any contractors or subcontractors working directly or indirectly on your behalf are performing operations, if the “property damage” arises out of those operations; or

6. That particular part of any property that must be restored, repaired, or replaced because “your work” was incorrectly performed on it.

Paragraph (2) of this exclusion does not apply if the premises are “your work” and were never occupied, rented, or held for rental by you.

Paragraphs (3), (4),. .. of this exclusion do not apply to liability assumed under a sidetrack agreement.

Paragraphs (3), (4), (5), and (6) of this exclusion do not apply to liability assumed under a sidetrack agreement.

Paragraph (6) of this exclusion does not apply to “property damage” included in the “products – completed operations hazard.”60

The “products-completed operations hazard” is defined as:

a. … all “bodily injury” and “property damage” occurring away from premises you own or rent and arising out of “your product” or “your work” except:

(1) Products that are still in your physical possession; or

(2) Work that has not yet been completed or abandoned. However, “your work” will be deemed completed at the earliest of the following times:

(a) When all of the work called for in your contract has been completed.

(b) When all of the work to be done at the job site has been completed if your contract calls for work at more than one job site.

(c) When that part of the work done at a job site has been put to its intended use by any person or organization other than another contractor or subcontractor working on the same project.

Work that may need service, maintenance, correction, repair or replacement, but which is otherwise complete, will be treated as completed.61

CGL policies provide coverage for liability for damage to property of a third party. They do not provide first-party coverage for damage to the insured’s property. Paragraphs 1 – 4 give effect to this distinction. Under these paragraphs, damage to property owned or over which the insured exercised care, custody or control is not covered.

As noted above, paragraph 5 of Exclusion j bars coverage for “[tjhat particular part of real property on which you or any contractors or subcontractors working directly or indirectly on your behalf are performing operations, if the ‘property damage’ arises out of those operations.” This exclusion applies only to damage that occurs during the conduct of operations.62

A number of cases analyze the limitation of paragraph 5 to “that particular part” of real property. In Columbia Mutual Insurance Co. v. Schauf,63 the insured was a contractor who was painting kitchen cabinets in a homeowner’s house. While he was cleaning his equipment, the pump generator used for applying the paint started a fire that caused extensive damage throughout the house. The court held that exclusion j.5 precluded coverage solely for damage to the cabinets but not damage to the remainder of the house.

In Unionamerica Insurance Co. v. Johnson,6″1 the insured was hired to repair and replace the homeowner’s roof. Rainwater entered the home during a storm and damaged the homeowner’s property and equipment. The court held that exclusion j.5 did not bar coverage for the damage to the remainder of the house, as the insured was performing operations solely to the roof and not to the remainder of the house.

Paragraph 6 of Exclusion j bars coverage for damage to property that must be restored, repaired, or replaced because “your work” was incorrectly performed on it. However, the exception to paragraph 6 provides that it does not apply to property damage included in the products – completed operations hazard. Therefore, to the extent the alleged damage took place after the contractor had completed operations at the site, this exclusion would not apply. In sum, paragraph 6 excludes coverage for damage to property that occurs during construction operations.

D. Business Risk Exclusions

The so-called “business risk” exclusions (such as the your work, your product and impaired property exclusions) give effect to the basic purpose of liability policies of protecting against the risk of injury to persons and damage to property caused by faulty workmanship, but not protecting the contractor against the economic consequences of the faulty workmanship itself.65 If the work performed is faulty, then a breach of contract or warranty occurs, as the customer did not obtain the benefit of his bargain. This gives rise to a claim for breach in which damages are measured by the lost expectation. Liability insurance does not protect against such risks. This is primarily on the theory that the contractor has it within his power to control the quality of his work and thus the scope of his liability, which is measured by the contract. On the other hand, injury to persons or damage to other property resulting from faulty work exposes the contractor to almost unlimited liabilities. It is this risk that CGL policies are intended to protect against.66 The discussion of these issues in the frequently cited case of Weedo v. Stone-E-Brick, Inc.61 includes the following illustration:

When a craftsman applies stucco to an exterior wall of a home in a faulty manner and discoloration, peeling and chipping result, the poorly-performed work will perforce have to be replaced or repaired by the tradesman or by a surety. On the other hand, should the stucco peel and fall from the wall, and thereby cause injury to the homeowner or his neighbor standing below or to a passing automobile, an occurrence of harm arises which is the proper subject of risk-sharing as provided by the type of policy before us in this case. The happenstance and extent of the latter liability is entirely unpredictable – the neighbor could suffer a scratched arm or a fatal blow to the skull from the peeling stonework. Whether the liability of the businessman is predicated upon warranty theory or, preferably and more accurately, upon tort concepts, injury to persons and damage to other property constitute the risks intended to be covered under the CGL.68

The basic distinction drawn in Weedo informs the following analysis, with the understanding that the wording and scope of these exclusions has changed over the years, so it is always necessary to pay close attention to the particular wording of the similar exclusions in older cases.

1. Damage to Your Product Exclusion

CGL policies often contain the following provision, “[t]his insurance does not apply to … ‘[p]roperty damage’ to ‘your product’ arising out of it or any part of it.”69 The term “your product” means “any goods or products, other than real property, manufactured, sold, handled, distributed or disposed of by … [y]ou.”70

This exclusion bars coverage for damage to an insured’s products, as opposed to damage to the property of third parties caused by the insured’s defective products.71 Numerous cases have found this exclusion (as well as the “your work” exclusion) clear and unambiguous and have applied it to bar coverage for the insured’s product or work.72

It is important in the present context, however, to note that in recent versions of the standard CGL form, the “your product” exclusion expressly excludes “real property” from the definition of “product.” As a result, this exclusion will not be applicable in those construction defect cases in which the insured’s product is deemed to be real estate.

2. Damage To Your Work Exclusion

CGL policies also often bar coverage for

“[p]roperty damage” to “your work” arising out of it or any part of it and included in the “products-completed operations hazard.”

This exclusion does not apply if the damaged work or the work out of which the damage arises was performed on your behalf by a subcontractor.73

The term “your work” is defined as

a. Work or operations performed by you or on your behalf; and

b. Materials, parts or equipment furnished in connection with such work or operations.

“Your work” includes:

a. Warranties or representations made at any time with respect to the fitness, quality, durability, performance or use of “your work”; and

b. The providing of or failure to provide warnings or instructions.74

This exclusion bars coverage for property damage to the insured’s own work, both work that fails and other work of the insured that is damaged by the work that fails. The intent of such an exclusion is to ensure that the insurer does not become a guarantor of the quality of the insured’s workmanship; in other words, the effect of the policy is to hold the contractor responsible for its own replacement and repair losses while the insurer takes the risk of injury to the property of others.

Significantly, this exclusion does not apply to work performed on the insured’s behalf by a subcontractor (although some courts have found that the exclusion is inapplicable when the subcontractor was acting under the control of the insured). Note that the term “subcontractor” is not defined in CGL policies. A number of cases support the common sense view that a subcontractor is one who performs a portion of an existing contract with a third party, i.e., takes a portion of an existing contractual obligation created with a third party. A case coming to that conclusion applying Exclusion 1. is National Union Fire Insurance Co. v. Structural Systems Technology, Inc.75

This limitation on the reach of the exclusion appears to dovetail with the very common practice of requiring the subcontractor to indemnify the general contractor and to add the general as an additional insured on the subcontractor’s policy. Thus, the general contractor’s CGL policy will not exclude coverage for the general’s work, but the general (and hence, its insurer) may be able to look to the subcontractor and its insurer for indemnification and coverage under the subcontractor’s policy.

A recent case in which this was not so is Pulte Home Corp. v. Fidelity & Guaranty Insurance Co.16 There Pulte sought reimbursement for its costs in defending and settling lawsuits arising out of the installation of Exterior Insulation Systems (“EIFS”) on homes it had constructed. EIFS has lately become a fertile source of litigation because of its alleged tendency to allow water infiltration. Pulte was seeking coverage as an additional insured under the liability policy issued to the subcontractor that had installed the EIFS. The insurers denied coverage for the portion of payments allocated to the repair and replacement of the EIFS, based in part on an exclusion that the insurance provided to an additional insured did not apply to property damage to the subcontractor’s work performed for the additional insured. In sum, this exclusion put the contractor and subcontractor in the same position regarding coverage for the subcontractor’s work, i.e., no coverage. Apart from that, the insurer argued that defective workmanship was not an occurrence. The court upheld the insurer on both arguments, ruling that the insurer had no obligation to indemnify Pulte for the portions of its settlements allocable to the repair and replacement of EIFS.

3. Impaired Property Exclusion

Many CGL policies do not apply to

“[p]roperty damage” to “impaired property” or property that has not been physically injured, arising out of:

(1) A defect, deficiency, inadequacy or dangerous condition in “your [the insured’s] product” or “your [the insured’s] work;” or

(2) A delay or failure by [the insured] or anyone acting on [the insured’s] behalf to perform a contract or agreement in accordance with its terms.77

This exclusion bars coverage for economic losses arising from damage to another’s property caused by the incorporation of an insured’s faulty workmanship or materials into a larger product, and is intended to apply when there is no actual physical injury to the property, but when property is impaired because of a defect, deficiency, inadequacy or dangerous condition in the insured’s product (“your product”) or the insured’s work (“your work”).78 Before this exclusion will operate to bar coverage, the impaired property must be capable of being restored by the repair or replacement of the insured’s work or product. The exclusion is thus limited to economic losses caused by the insured, so long as the damaged property can be repaired.79

4. The Sistership Exclusion

The “Recall of Products, Work or Impaired Property” Exclusion (or the “Sistership Exclusion”) bars coverage for

damages claimed for any loss, cost or expense incurred by [the insured] or others for the loss of use, withdrawal, recall, inspection, repair, replacement, adjustment, removal or disposal of “your [the insured’s] product”; . . . “your [the insured’s] work”;… or “impaired property”; if such product, work, or property is withdrawn or recalled from the market or from use by any person or organization because of a known or suspected defect, deficiency, inadequacy or dangerous condition in it.80

The exclusion gets its name from the aircraft industry’s practice of recalling all planes of a particular model when a plane of that model (a sister ship) crashed as a result of a suspected defect.81 The purpose of the provision is to exclude from coverage the cost of preventive or curative action undertaken through the withdrawal of a product that is potentially defective.82 The exclusion thus bars coverage for economic loss resulting from the withdrawal of property that has not yet been damaged. It does not apply, however, when property is not recalled, but is repaired or replaced because it has been damaged.83



Mold-related bodily injury and property damage claims have increasingly become a part of construction defect claims. An issue that arises is whether coverage for mold claims is barred by the pollution exclusion in CGL policies.84 These claims raise two basic issues: (1) whether mold is a “pollutant” as defined in the pollution exclusion; and (2) whether an infestation of mold constitutes the type of discharge, dispersal, or release of pollutants required under the exclusion. To date, there are only a few cases around the country dealing with this issue. Although a number of them involve claims by property owners under their “first-party” policies, the analysis in these cases is relevant in the liability context as they involve similar exclusion language and similar facts as may be found in construction defect liability cases.

In Leverence v. United States Fidelity & Guaranty,85 the court held that the sudden and accidental pollution exclusion did not apply where the alleged cause of bodily injury and property damage was exposure to water vapor trapped in the walls, which caused the growth of mold and mildew, as no contaminants were released, but rather formed over time as a result of environmental conditions.

In State Farm Fire & Casualty v. M.L. T. Construction Co. ,86 an employee in an office building sued a number of parties, including the roofing contractor, involved in the reroofing of the building, alleging that she sustained bodily injuries caused by mold that grew in the building as the result of massive amounts of water that entered due to rain storms during the re-roofing. The roofing contractor’s insurer argued that there was no coverage because a “total pollution exclusion” in the policy barred coverage for injury due to mold. The exclusion barred coverage for injury or damage that “would not have occurred in whole or in part but for the actual, alleged, or threatened discharge, dispersal, seepage, migration, release or escape of pollutants at any time.”87 The court held that the pollution exclusion did not apply for a number of reasons. First, and most important here, the court said that the plaintiff’s damages were not caused by the use or dispersal of any dangerous materials, but rather by the failure to prevent the intrusion into the building of rainwater, which is not a pollutant. second, the court declined to apply the pollution exclusion in a construction context, as pollution exclusions are intended to apply only to active industrial polluters who knowingly emit pollutants over extended periods. As a result, for courts that take this view, the pollution exclusion will not apply to the accidental and occasional discharge of pollutants at job sites during operations or from chemicals and materials used in construction materials.

In Lexington Insurance Co. v. Unity/Waterford-Fair Oaks, Ltd. ,88 a building owner sought first-party coverage for mold damage to its building caused by a severe rainstorm and flooding. The policy contained a pollution and contamination exclusion that barred coverage for damage caused by the “release, discharge, dispersal, escape or dispersal of contaminants or pollutants,” and defined “contaminants or pollutants” as including “bacteria, fungi, virus, or hazardous substances as listed in [various federal statutes].”89 The policyholder did not dispute the validity of the exclusion or argue that mold was not included within it, but argued that the exclusion did not apply because the mold in the building had not been released, discharged or dispersed. The policyholder pointed to the testimony of the insurer’s expert that mold and mold spores exist at low levels in all environments and that the mold causing the damage was already present and thrived because of the moisture. As a result, there was no release of a pollutant into the buildings that caused the damage. The court disagreed, pointing to the evidence that mold proliferates when exposed to unusual amounts of water. Under normal conditions, mold spores exist at safe levels, but when conditions for the mold are enhanced, the spores proliferate, giving off reproductive spores that are dispersed into the air to the point that they can become unhealthy. The result is that there is a release or dispersal in the case of mold contamination, thus satisfying the requirements of the pollution exclusion.

It is worth mentioning some other representative cases dealing with the application of the pollution exclusion to toxic materials used in construction or other operations (i.e., situations involving no disposal of wastes but the use of products). In Madison Construction Co. v. Harleysville Mutual Insurance Co.,90 the court held that the absolute pollution exclusion was clear and unambiguous in denying a claim for damages caused by vapors emanating from a curing agent applied to a concrete trench. The court, in rejecting the insured’s contention, held that no ambiguity existed in the exclusion, as it clearly precluded coverage for the discharge, dispersal, release or escape of smoke, vapors, soot, fumes.

In another case, the court ruled that the absolute pollution exclusion barred coverage for claims alleging personal injury and wrongful death caused by carcinogenic emissions from a defectively manufactured furnace.91



CGLpolicies also contain “conditions” detailing the insured’s obligations. Among other things, these provisions require the insured to give the insurer timely notice of an occurrence, claim or suit. A typical policy may contain the following:

Duties in the Event of Occurrence, Offense, Claim or Suit

a. You must see to it that we are notified as soon as practicable of an “occurrence” or an offense which may result in a claim. . . .

b. If a claim is made or “suit” is brought against any insured, you must:

(1) Immediately record the specifics of the claim or “suit” and the date received; and

(2) Notify us as soon as practicable.

You must see to it that we receive written notice of the claim or “suit” as soon as practicable.92

As noted above, CGL policies typically require that the insured give certain notice in the event of an occurrence, offense, claim or suit. Even though an insured breaches its duty to give timely notice, courts in many jurisdictions (with some significant exceptions, such as New York) do not allow an insurer to deny coverage on account of the insured’s breach of the policy notice conditions unless the insurer has been prejudiced by the insured’s breach. In most jurisdictions the insurer bears the burden of proving that the notice was late and that the late notice resulted in “appreciable prejudice” to the insurer.93 In Morales v. National Grange Mutual Insurance Co. ,94 the court explained that the phrase “appreciable prejudice” consists of two elements: (1) substantial rights “irretrievably lost by virtue of the failure of the insured to notify the insurer in a timely fashion”; and (2) likelihood that the insurer would have had a meritorious defense had it been informed of the accident in a timely fashion.95

Prejudice as a matter of law has been found when settlement is “a fait accompli” and when the insurer had no opportunity to control the proceedings or in any way protect itself.96

In some jurisdictions prejudice will be presumed from an unreasonable delay by the insured in giving notice and the insured then has the burden to show an absence of prejudice.97



Analysis of an insurer’s duty to defend and indemnify a construction defect action is complicated by the fact that construction defect claims involve multiple policies in effect for the same triggered period, as well as triggered policy in effect in different periods.98 Each of these situations raises questions concerning the extent of an insurer’s duty to defend and indemnify, versus the obligations of other insurers and the insured. This section deals with issues of multiple policies in effect for the same period, usually analyzed under CGL policies’ “other insurance” provisions.” The next section discusses multiple-year issues, often described under the heading of “allocation.”

An example may help to explain the additional insured/other insurance problem. Assume that Developer hires General Contractor to construct a development of single family homes. General Contractor subcontracts with Window Installer to install the windows and sliding glass doors at the homes. After the windows and doors are installed, homeowners file suit against the Developer and the General Contractor alleging that the windows and doors are defective, resulting in water leaks into their homes, causing both damage to the structure and interior of the homes and mold-related bodily injury claims.

General Contractor tenders the defense of the homeowners’ action to its CGL insurer for coverage under the policy in which it is the named insured. The General Contractor, however, may have other coverage available as well. The contracts entered into between the General Contractor and the subcontractors, including Window Installer, may contain both (1) an indemnification provision calling for the subcontractor to hold harmless and indemnify the General Contractor from any loss arising out of the subcontractor’s negligence, and (2) an insurance provision obligating the subcontractor to add the General Contractor as an additional insured to the subcontractor’s CGL policy.100

If General Contractor’s CGL insurer is presented with the homeowners’ suit described above, that insurer must ( 1 ) determine whether General Contractor has a contractual indemnification provision from Window Installer, or any other potentially liable party; if so, the insurer should tender the claim under that provision; and (2) assess whether General Contractor is an additional insured under Window Installer’s policy, and, if so, then the claim should be tendered to that insurer as well.

A party may be added to a policy as an additional insured in a number of ways.101 Sometimes this is done by an endorsement that adds an identified entity as an additional insured for all purposes. Other policies add specific entities as additional insureds only for limited purposes, such as for a particular project or location. Other endorsements add additional insureds by general description, by modifying the “Who Is an Insured” provision of the policy to include as an insured

the person or organization shown in the Schedule, but only with respect to liability arising out of your ongoing operations performed for that insured.102

Another form provides:

This endorsement includes as an insured any person or organization (called additional insured) whom you [the named insured] are required to add as an additional insured on this policy under a written contract. . . .103

The question then becomes: what is the relationship between the different coverages that may apply when the developer or general contractor is an additional insured under one or more of its subcontractors’ insurance policies while at the same time it has purchased its own general liability insurance? All of these policies may provide coverage for the same construction defect claims. Most policies contain an “other insurance” clause that defines the extent to which the insurer will respond if some other policy covers the same claim or suit. Courts have applied such provisions both to concurrent as well as consecutive policies, although usually “other insurance” clauses are applied to concurrent policies, that is, policies in force during the same period.104 There are three main kinds of “other insurance” clauses: (1) “pro rata” (insurer shall not be liable for more than its pro rata share of the loss), (2) “excess” (insurer shall be liable only to the extent the loss exceeds other valid and collectible insurance), and (3) “escape” (if other valid and collectible insurance exists, insurer shall have no further liability).105

Given these different “other insurance” provisions and the possibility of a number of policies covering a claim, there are a large number of possible combinations of potentially conflicting clauses in any given case. Thus, one policy could have a “pro rata” clause, the second an “excess” clause, the third an “escape” clause, and the fourth none. In general, there are three possible ways that courts have applied or reconciled additional insured and direct coverage, depending upon the language of the “other insurance” provisions in the two policies: (1) where the insured’s own policy contains an excess “other insurance” provision, a number of cases hold that the policy affording additional insured coverage is primary and the additional insured’s own policy is excess or not applicable; (2) in cases where the subcontractor’s policy contains an excess “other insurance” provision, several cases hold that the policy affording additional insured coverage is excess and the additional insured’s own policy is primary; (3) finally, there are cases in which both the additional insured coverage and the additional insured’s own policy are found to be co-primary and obligated to share pro rata in the additional insured’s defense. Representative cases are discussed below.

A. Additional Insured Coverage Found Primary and Additional Insured s Policy Excess

In Northbrook Property & Casualty Insurance Co. v. United States Fidelity & Guaranty Co.,106 the court enforced the “other insurance” provisions as written to require the policy providing additional insured coverage to a construction manager to be primary and the construction manager’s own coverage to be excess. Northbrook, the construction manager’s own insurer, issued a policy that included an excess “other insurance” endorsement. A subcontractor had the construction manager added as an additional insured under its policy, issued by USF&G, which apparently had a standard primary “other insurance” provision. The court held that USF&G was obligated to defend the construction manager on a sole primary basis.

In Transamerica Insurance Group v. Turner Construction Co.,101 the court enforced the “other insurance” provisions as written to require the policy affording additional insured coverage to be primary and the additional insured’s own policy to be inapplicable. Turner was a general contractor that required a subcontractor to procure coverage for it as an additional insured under the subcontractor’s policy. Turner’s policy, issued by Liberty Mutual Insurance Company, contained an “other insurance” clause stating “it did not apply to that portion of the loss for which the Insured has other valid and collectible insurance as an Additional Insured on a Liability Insurance policy issued to a subcontractor of the Named Insured.”108 Transamerica, which insured the subcontractor, issued a policy with a primary “other insurance” clause. The court concluded that Liberty Mutual’s “other insurance” provision meant that Transamerica was obligated to defend and indemnify Turner on a sole primary basis.

B. Policy Affording Additional Insured Coverage Excess and Insured’s Own Primary

In Honeywell, Inc. v. American Motorists Insurance Co.,109 the court enforced the “other insurance” provisions as written to require the additional insured’s own policy to be primary and the additional insured coverage to be excess. The additional insured was Honeywell, the owner of premises on which a vending company’s worker was injured. Honeywell was added as an additional insured under the vending company’s insurer’s policy. Hartford insured Honeywell and American Motorists insured the vending company. Hartford’s policy contained a standard primary “other insurance” clause. American Motorists’ policy included an excess “other insurance” clause. Contending that the parties intended the additional insured coverage to be primary, Hartford argued that “the intentions of the parties should govern” and that “the language of the two Other insurance’ clauses should not control over that intention.”110 In rejecting this argument, the court explained that “it is not necessary to disregard the Other insurance’ clauses and to resort to the intent of the underlying agreement between the two parties to determine whose carrier is primary.”111

Honeywell was followed in Deerfield Management Co. v. Ohio Farmers Insurance Co.112 Deerfield was a hotel management company that leased hotel space to Georgia’s Fashions Cleaners and was added as an additional insured under Georgia’s policy issued by Ohio Farmers. Deerfield’s policy contained a standard primary “other insurance” provision. Ohio Farmers’ policy included an excess “other insurance” clause. Interpreting these “other insurance” clauses as written, the court held that Deerfield’s own insurance applied on a sole primary basis and that Ohio Farmers’ policy provided excess coverage to Deerfield. The court rejected Deerfield’s arguments that the terms of the lease agreement between it and Georgia’s “indicate that the intent of the parties” was that Ohio Farmers’ policy provide primary coverage for Deerfield.113 Deerfield unsuccessfully relied on both the insurance procurement provisions of the lease and the indemnity agreement in the lease in an attempt to avoid the application of the “other insurance” clauses.

C. Both Policies Co-Primary and Obligated to Share in Additional Insured’s Defense

In Sacharko v. Center Equities Ltd. ,”4 the appellate court reversed the trial court’s ruling that required a tenant’s insurer to defend and indemnity a landlord against a premises injury claim on a sole primary basis. The court remanded the case to the trial court to reapportion defense costs and damages between the two insurers based on the policies’ “other insurance” clauses. The tenant had added the landlord as an additional insured under its general liability policy, which was issued by Insurance Company of North America (“INA”). The landlord had its own general liability coverage under a policy issued to it by Home Indemnity Company. The INA policy had a standard primary “other insurance” clause. The appellate court stated that the INA policy was not in the record before it. Citing the general rule that all primary insurance contributes pro-rata to defense and indemnity, the court remanded the case to the trial court to determine a proper allocation of the defense costs and damages between the two insurers.

Federal Insurance Co. v. Insurance Co. of North America115 apportioned liability for defense costs equally between the two policies at issue for a series of underlying actions against the insureds. Federal issued a general liability policy to the subcontractor, which added the general contractor and the project owner as additional insureds. INA’s policy was issued to the general contractor and added the project owner as an additional insured. Both policies apparently contained primary “other insurance” clauses, although the court did not specify the type of “other insurance” clauses involved. The court held that because the policy limits of both policies were the same and both contained the same “other insurance” clause, and also because all liability actions were settled, the defense costs should be apportioned equally.

Courts will generally honor excess “other insurance” clauses when no prejudice to the interests of the insured will result.”6 However, what happens if the “other insurance” clause in the subcontractor’s policy also provides that it is excess over any other insurance? General rules of construction dictate that when two policies both provide excess insurance, then the two excess clauses cancel each other and the insurers share pro rata.117

An exception to the rule prorating co-primary policies is the rule in Illinois. There the insured has the right to designate which of the policies will be required to defend and indemnify, and that the insurer so chosen does not have the right to seek contribution from the other.118 The rationale for this rule is to “protect the insured’s right to knowingly forgo an insurer’s involvement.”119 In John Bums the general contractor tendered the case to the subcontractor’s insurer for defense and indemnity under the policy under which it was an additional insured. The court held that the subcontractor’s insurer did not have a right of contribution or pro rata sharing against the other insurer, even though both policies had the same standard “other insurance” provision.

Although these different groups of cases have different results, it seems clear that the outcome in each case generally depends on the specific language of the “other insurance” clauses. When the policies contain language that clearly makes one primary and another excess, courts will generally enforce those provisions.

D. The Special Conditions Endorsement

Some policies include an endorsement requiring the insured builder or contractor to obtain coverage as an additional insured under their subcontractors’ policies. A California court found such an endorsement valid and enforceable in Scottsdale Insurance Co. v. Essex Insurance Co.120 Scottsdale and Essex provided primary CGL coverage for successive periods to a builder. When the homeowner sued the builder for damages resulting from a progressive buildup of water infiltration damage, Scottsdale defended and settled the case, while Essex denied the claim. Scottsdale then sued for equitable contribution.

Essex relied in part on a provision of its policy that made it a condition of coverage that the builder would obtain certificates of insurance and hold harmless agreements from all his subcontractors and that he would be named an additional insured on all subcontractor general liability policies. In fact, the builder failed to meet these conditions, so that Essex claimed it was not obligated to defend the builder, as the condition had not been satisfied.

The court held that the endorsement (a) was not an illusory contract, as it provided that the Essex policy and the subcontractor’s policy would participate with one another in providing coverage; (b) was not merely an escape clause; (c) was not ambiguous; (d) was not impossible of performance; and (e) was enforceable. Further, although prejudice was not required for Essex to be able to deny coverage, in fact Essex had been prejudiced by the builder’s failure to obtain insurance coverage from its subcontractors, as it had lost the ability to participate with another insurer in the coverage. Therefore, Essex was justified in denying coverage and would not be required to contribute.



In long-term exposure cases in which injury or damage has occurred in more than one policy period, so that more than one policy has been triggered, the issue arises as to how the insurers’ liability will be allocated among the policies in the different years.121 In selecting an allocation method courts will usually be strongly influenced by the rationale or logic of the trigger of coverage theory they have adopted, as the determination of what policies are triggered will in the first instance dictate whether allocation is necessary at all and will also play a significant role in the selection of an allocation methodology. Thus, as discussed below, what an injury-in-fact trigger is used because the injury is found to be divisible among policy periods, then allocation is not necessary, as the known or proven injury will trigger only the policies in effect when the injury occurred. Similarly, if the court adopts a manifestation trigger, then no allocation is necessary, as the court has already determined which specific policies must respond.

However, when a continuous trigger is employed and the court determines that factual apportionment is not possible, some method of allocating the loss becomes necessary. At that point, the court will adopt a method that attempts to allocate based on its presumptions about the occurrence of the damage. Broadly speaking, under these circumstances courts have adopted two main allocation methods: (a) some form of proration among insurers and the insured, and (b) joint and several allocation, under which the insured selects which of the available policy years will be liable for the entire loss. These allocation methods are discussed below.122

A. Injury-in-Fact Allocation

Where an occurrence gives rise to damage occurring in a number of policy periods, but it is possible to quantify the damage occurring in each period, a court will likely employ an “injury-in-fact” trigger theory. As a result, damages will be apportioned based on the proof.123 Strictly speaking, this is not allocation at all, but simply injury-in-fact trigger applied to separable damage occurring in different years: In American Home Products Co. v. Liberty Mutual Insurance Co.,124 the court made a factual determination as to the time at which each claimant’s injury resulting from the ingestion of a drug took place, stating that “a real but undiscovered injury, proved in retrospect to have existed at the relevant time, would establish coverage, irrespective of the time the injury became [diagnosable].”125

An analogous situation in the construction defect area might be one in which, for example, faulty workmanship employed or defective products incorporated in different phases (and thus in separate buildings or units) of a condominium development resulted in divisible damage during successive policy periods. If an injury-in-fact trigger, rather than a manifestation trigger, were applied by the court, it would be appropriate in that situation to employ also an injury-in-fact trigger to allocate coverage according to when the damage actually occurred.

Injury-in-fact allocation is the method that is closest to the language of CGL policies, which provide that they cover damage that occurs “during the policy period.”126 Application of injury-in-fact apportionment ensures that an insurer will not be required to cover damage that occurred in periods for which it did not receive a premium. Also, because an insurer will be held liable only for damage that took place during its policy period, there will be no need for a later contribution action against other insurers. Further, because it is generally the insured’s burden to establish coverage, the insured must in the first instance prove the amount of injury or damage that took place in the policy period.

B. Pro Rata Allocation by Years on the Risk

As noted above, other allocation methods of sharing liability among insurers are employed in cases in which multiple years are triggered but it is not possible, or the court declines, to apportion liability based on the actual injury or damage in each period. One such approach is to allocate responsibility on a pro rata years-on-the-risk basis, under which each insurer is responsible only for the portion of the loss presumed to have occurred during its policy period and the insured is responsible for uninsured and self-insured periods. This proration may be based on the assumption that the damage or injury took place equally during each year of exposure up to manifestation.127

Courts that apply a pro rata by years-on-the-risk approach generally find that the even distribution of damages over the entire trigger period is most consistent with the concept of an injury-in-fact trigger.128 If the court accepts the assumption that damage occurred in an equal or linear fashion, then it will likely conclude that allocation based on time on the risk is fair to insurers, as it means, in effect, that no insurer is unfairly being required to pay for damage that took place outside its policy period.

C. Pro Rata by Limits

Another pro rata approach is based on limits of liability according to the amount of insurance purchased each year, on the theory that limits reflect premiums paid to each insurer, or a conscious decision to transfer risk to a particular policy period.129 Under this theory, the loss is allocated to the same policy periods that are implicated under a time-on-the-risk approach, but the amount allocated to each policy year depends on the degree of risk retained or transferred by the insured during that policy year, as compared with the other years during which the loss took place. As a result, under this theory as well, the policyholder is allocated a share of the loss for uninsured periods.

In Owens-Illinois, Inc. v. United Insurance Co.,130 the insured had insurance at different times and in greatly varying amounts over the 40-year span at issue; and was also self-insured, either partially or entirely, for many of those years. The court rejected both pro rata by years on the risk and “joint and several” allocation discussed below. It concluded that it was “unable to find the answer to allocation in the language of the policies.”131 In particular, the court found that “other insurance” provisions, which typically are designed to prevent double recovery from multiple concurrent policies, were not applicable in the continuous trigger context, where successive rather than concurrent policies were at issue.132

The court ruled that allocation should be “in proportion to the degree of the risks transferred or retained during the years of exposure.”133 Thus, the insured’s year-by-year increases in coverage “must have reflected an increasing awareness of the escalating nature of the risks sought to be transferred.”134 As a result, losses were to be allocated among the insurers on the basis of the extent of the risk assumed, so that proration would be on the basis of policy limits, as well as years on the risk. The court also said that “[w]hen periods of no insurance reflect a decision by an actor to assume or retain a risk, as opposed to periods when coverage for a risk is not available,” it would be reasonable to expect the insured to share in the allocation.135

By allocating based on the total level of risk assumed in each year, the pro rata by limits approach, unlike the pro rata by time on the risk approach, does not seek to address the issue of how much damage took place during a particular policy period. Rather, the focus is on the policyholder’s choice regarding how much coverage to purchase, which may not reflect the amount of damage that has already occurred. The effect of Owens-Illinois will often be to shift a disproportionate share of a loss to years in which policies happen to have higher limits, without regard to how much damage or injury occurred in any given year. Thus, for example, assuming that the insured over the years has purchased larger and larger amounts of excess coverage above approximately the same level of primary coverage, a primary policy in the first year may end up paying substantially less than the same primary policy in the last year, simply because of the increased amount of excess insurance sitting above it, and without regard to when the damage actually occurred. As a result, because this allocation formula does not attempt to allocate based on the percentage of injury during successive years, but instead focuses on the retention and transfer of risk, it does not follow the policy language.

Under Owens-Illinois, should an insured be responsible for amounts allocated to policies that contain applicable exclusions or were issued by companies that have become insolvent? Chemical Leaman Tank Lines v. Aetna Casualty & Surety Co.136 held that in a case in which the pollution exclusion barred coverage under a number of the policies at issue, a proper allocation should include an examination of whether insurance was available and the insured chose not to purchase it, in which case the insured would be required to assume some of the environmental liability. Also, the insurers had the burden of proving the availability of insurance.

D. Joint and Several Allocation

The next allocation method is “joint and several.” As articulated in Keene Corp. v. Insurance Co. of North America,137 any triggered policy, as selected by the insured, must respond for the entirety of a claim, subject to the effect of other insurance clauses and principles of equitable contribution. Under the joint and several rule, because a policy once triggered (as, presumably, at least some injury or damage took place during the policy period) is required to pay “all sums” for which the insured becomes liable arising out of the triggering occurrence, the insurer may allocate the entire loss to that year, regardless of the fact that other years have been triggered. In addition, because the designated insurer must pay “all sums,” the policyholder is not responsible for any portion of coverage during periods it was uninsured or self-insured. Under this rule, the policyholder does not have to deal with allocation issues. Instead, the burden shifts to the selected insurers, that may seek contribution from other triggered insurers under the “other insurance” clause of their policies or under the doctrine of equitable contribution. This approach has been adopted by several cases, including J.H. France Refractories Co. v. Allstate Insurance Co.138

A number of courts have criticized the joint and several approach, noting that policies do not simply include the “all sums” language cited by some joint and several courts, but also provide that the obligation to pay “all sums” is limited to liability for damage “during the policy period.”139 Thus, the insurer’s obligation is to pay “all sums” that the insured becomes liable to pay as the result of an occurrence that causes damage during the policy period. The term “all sums” does not apply to damage occurring outside the policy period. The result is that an insured by purchasing only one year of insurance may obtain full coverage for a loss that triggered multiple years as the result of continuous damage. This means that, in effect, a policyholder purchasing only one year of insurance could get as much coverage as an insured that purchased policies throughout the entire period of exposure and damage.

One of the most significant features of joint and several allocation is that in a sense it does not call for allocation at all. That is, under joint and several allocation the court simply gives the policyholder the right to select which triggered policy will respond to the loss. No allocation by the court among the insurers is done. Instead, it is the obligation of the selected insurer to seek contribution from other insurers.

E. Excess Insurance Under the Allocation Theories

An issue that arises in connection with allocation is the treatment of excess insurance, in particular, how the policies are to be exhausted in relation to one another. Different allocation methods will give rise to different treatment of excess policies. Thus, under either a pro-rata allocation or an injury in fact approach, a specific portion of damages is allocated to a particular policy. The question then arises as to how the allocated amounts are to be exhausted. Some courts have held that once the loss is spread pro-rata by the years of the triggered policies and the amount allocated to a particular year is determined, the policies within that year, both primary and excess, pay as they normally would, beginning with the primary policies and proceeding vertically through each succeeding excess layer.140 Under this approach liability could attach to an excess policy only if the portion of the loss allocated to that excess policy was sufficient to exhaust all underlying coverage in that particular year.

A feature of this approach is that it is possible that some excess policies will be required to pay prior to exhaustion of all triggered primary policies. Thus, if the loss is such that $500,000 is allocated to each policy year under a pro-ration by time on the risk approach, the excess policies that happen to be in effect during early years with, for example, $100,000 annual primary limits, would be required to pay $400,000 while later excess policies that sit above $1 million in annual primary limits would not be reached.

In contrast, some jurisdictions adopt an “exhaustion by layers” approach under which an excess carrier is not implicated until all applicable primary policies have been exhausted.141 They rely on the “other insurance” clause and the distinction between primary and excess insurance. These cases also rely on the provision in excess policies that they do not contribute “if other valid and collectible insurance with any other insurer is available to the insured.”142 This allows the insured to stack primary limits, and then, once exhausted, seek coverage under excess policies. The result is that primary policies in effect during different periods become coinsurance rather than concurrent insurance.

Owens-Illinois,143 discussed above, did not discuss treatment of excess policies under the pro rata by limits approach. Carter-Wallace, Inc. v. Admiral Insurance Co.144 took up that issue. There, the court allocated liability to an excess layer in any given year based on the relative liability of all coverage in that year to the total of all coverage liability in all years at issue. The court used the following example. Assume a nine-year period at issue, there are policies in force in each of the first three years of $2 million per year, of $3 million per year in years four through six, and the insured was self-insured in years seven through nine at a risk assessed at $4 million per year. This would result in a total available coverage and self-insurance of $27 million. The carriers in years one through three would bear 2/ 27ths in each year, those in years four through six would bear 3/27ths in each year and the insured would bear 4/27ths of the risk in each of the last three years.

Then, after the total liability is calculated, the appropriate share of that total is allocated to each year. For example, assume that the primary coverage for one year was $ 100,000, the first level excess was $200,000 and the second level excess was $450,000. If the loss allocated to that year was $325,000, then the primary insurer and the first layer excess would both pay their policy limits (totaling for both $300,000) and the second-level excess would pay $25,000. In sum, Owens-Illinois and Carter Wallace adopt an allocation scheme that combines both horizontal and vertical application into a single allocation covering all applicable years. As a result, under Carter Wallace an excess insurer can become liable only if the amount allocated to its policy period would exhaust the limits of liability in the primary and all other underlying policies.

Under joint and several liability it is usually held that an insured may select a single policy period to respond to a claim in the first instance and if the loss exhausts the primary limits in that period, an insured can require the excess policies in effect during that same period to pay up to their limits. The primary and excess insurers in that period may then seek contribution from other insurers. Under that scheme the insured may not stack limits of consecutive policy periods, so that for any given loss the insured is entitled to only one year’s limits.145 In J.H. France Refectories Co. v. Allstate Insurance Co.,146 however, the court said that when the policy limits of a given insurer are exhausted, the insured is entitled to seek indemnification from any of the remaining insurers.

Because the courts that apply a joint and several approach hold that each triggered insurer is liable for the entire loss (subject to its policy limits), they treat the various triggered policies as co-insurance, even though those policies provide coverage for different periods. Thus, those courts rely on the “other insurance” clause as a basis for holding that a “selected” insurer can seek contribution from other triggered policies. This suggests that all primary policies are available insurance for the loss and an excess policy that is in excess of all “other available insurance,” should respond only after all available primary insurance has been exhausted. In Stonewall Insurance Co. v. CityofPalos Verdes Estates,147 the court stated that “all of the primary policies in force” during the continuous period of the occurrence are primary to each of the excess policies.148



The rule in most states is that if an insurer has a duty to defend with regard to any aspect of a lawsuit, it must defend the entire suit.149 Courts therefore require an insurer to pay all reasonable and necessary defense costs, so long as at least one claim implicates coverage. Some courts make an exception when defense costs can be readily apportioned between covered (and potentially covered) and non-covered claims.150 In the leading case of Buss v. Superior Court,151 the California Supreme Court adopted this rule relating to apportionment of defense costs, but held that the insurer should pay all of the defense costs and then seek reimbursement from the insured. Further, the insurer’s right to reimbursement will depend on the insurer having put the insured on notice of the intent to seek reimbursement. However, this may be done by a reservation of rights. Obviously, this should be done as early in the case as possible, so that the insurer is not deemed to have waived the right to seek reimbursement. In addition, the insurer has the burden of proving that it is possible to allocate the costs between covered and non-covered claims. This issue was address by the Court of Appeals in Buss.152

A related issue is what happens when an insurer has wrongfully refused to defend an action. SL Industries v. American Motorists Insurance Co.153 held that in such a situation the insurer is required to reimburse the insured for its defense costs, but its duty to reimburse is still limited to allegations covered under the language of the policy, provided that the defense costs can be apportioned between covered and non-covered claims.154 As a result, the insurer is obligated “to pay only those defense costs reasonably associated with claims covered under the policy.”155 The court further noted that when defense costs cannot be apportioned, the insurer must assume the costs of the defense for both covered and noncovered claims.156



Insurers and insureds have been litigating cases concerning the scope of insurance coverage for construction defect claims for decades. Nonetheless, courts today continue to grapple with these issues and struggle with the question of whether construction defect claims involve an “occurrence” covered under a typical CGL policy and whether any such claim is otherwise excluded by the policy.157 Given the dollars at stake and the numerous parties, and thus insurance policies, involved in many construction defect cases, we can expect insureds to continue to demand insurance coverage for construction defect claims and expect the courts to be repeatedly called upon to determine the extent, if any, of coverage for these claims.

Richard M. Shusterman is a partner in and past chair of the Commercial Litigation Department of White and Williams, LLP and the current chair of the ADR Practice Group. He founded and served as chair of both the Insurance Coverage and Reinsurance Practice Groups. He has over thirty-five years experience in representing and advising insurance, reinsurance and business clients in the resolution of complexfrequently multi-party-disputes through negotiation, mediation, arbitration, or litigation. Mr. Shusterman s national practice has included some of the insurance industries ‘ most visible litigation. In recent years, his practice has concentrated on the use of ADR techniques to resolve complex multiparty disputes. He has served as Umpire, Arbitrator, Mediator or party advocate in a variety ofADR proceedings. He is an active member of the Federation of Defense & Corporate Counsel and has served on its Board, as Vice President and as Chair of its Publication Committee and its Insurance Coverage and its Technology and ?-Commerce substantive law sections. He currently chairs the FDCC ADR section. He has been a frequent author and speaker on issues of concern to the insurance and reinsurance industries. Mr. Shusterman graduated in 1961 from Lafayette College, magna cum laude, with honors in history and in 1964 from the Law School of the University of Pennsylvania, cum laude. From 1965 to 1968 he served as an officer in the U.S. Army Judge Advocate General Corps. Mr. Shusterman is licensed to practice in Pennsylvania and New York.

Anthony L. Miscioscia is a partner in the Commercial Litigation Department of White and Williams, LLP and a member of the Business Insurance Practice Group and Environmental Coverage Practice Group. He has a broad range of experience in complex insurance coverage, bad faith and commercial litigation matters. Mr. Miscioscia represents major property and casualty insurers and smaller regional insurance companies on both a national and regional basis. His practice includes representation in state and federal court, as well as in private arbitrations and mediations. Mr. Miscioscia has presented lectures to the insurance industry on topics ranging from advertising injury/intellectual property coverage to successor liability. He is a co-author of Insurance Coverage for Cyberspace Claims, The Legal Intelligencer, August 28, 2000. Mr. Miscioscia received his B.A., magna cum laude and Phi Beta Kappa, in 1990 from Duke University. He received his J.D. in 1993 from the University of Virginia School of Law, where he served on the Editorial Board of the Virginia Law Review and graduated Order of the Coif. Licensed to practice law in Pennsylvania and New Jersey, Mr. Miscioscia is admitted to appear before the Supreme Court of Pennsylvania, the Supreme Court of New Jersey, the United States District Court for the Eastern District of Pennsylvania, the United States District Court for the District of New Jersey and the Third and Ninth Circuit Court of Appeals. He is a member of the Defense Research Institute and the Philadelphia Bar Association.

Peter F. Rosenthal is an associate in the Commercial Litigation Department and a member of the Business Insurance Practice Group at White and Williams, LLP. He has extensive experience in insurance coverage and bad faith litigation and also in complex commercial litigation. Mr. Rosenthal represents insurers in a broad range of insurance issues, including environmental and asbestos coverage litigation, and construction and product defect, educator s legal liability, and professional liability coverage disputes. Mr. Rosenthal s writings in the coverage area include, as coauthor, “Insurance Coverage For Year 2000 Claims – The Insurer s Perspective, ” Journal of Insurance Coverage, 1998; and “Liability Insurance Coverage For Products Liability Claims, ” Insurance Society of Philadelphia Seminar Course Materials, 1996. He has also lectured on insurance coverage issues for the Insurance Society of Philadelphia. Mr. Rosenthal received his B.A. in 1968 from St. John s College in Annapolis, Maryland and his J.D. in 1978 from Villanova University School of law. He practices in state and federal courts around the country and is experienced in alternative dispute resolution. He is licensed to practice in Pennsylvania and is also admitted before the United States District Court for the Eastern District of Pennsylvania.

Copyright Federation of Defense & Corporate Counsel, Inc. Summer 2005

Provided by ProQuest Information and Learning Company. All rights Reserved