When is a Mistake an ‘Accident’?

Jason R. Potter | Claims Magazine | June 2019

Commercial General Liability (CGL) policies, like all insurance products, are intended to protect the insured from unexpected claims or suits by third parties. A CGL policy covers bodily injury, property, personal and advertising liability, products and completed operations and fire liability unless they are excluded by a specific endorsement. CGL coverage can provide important protections to a business owner for a variety of losses, but they are most ubiquitous in the construction setting.

Perhaps no issue is as litigated and disputed within the CGL context as whether faulty construction work comes within a typical CGL policy’s initial grant of coverage. Pundits, professors and law professionals have all offered starkly differing views, but recent cases appear to be coalescing in favor of coverage, albeit for a specifically delineated subset of damages. This article will briefly summarize the positions and arguments in favor and in opposition to such coverage and offer a glimpse of what the future may hold. This article is limited to analyzing the CGL’s initial grant of coverage and does not delve into how, if at all, the policy’s 20 or so exclusions would affect that grant. The case for CGL coverage CGL insurance was first developed in 1941 as a way to purchase separately insured risks in one, unified policy. The Insurance Services Office, Inc. last modified the current standard form CGL policy in 1986. Its initial grant of coverage provides: “We will pay those sums that the insured becomes obligated to pay as damages because of ‘bodily injury’ or ‘property damages’ to which this insurance applies.” It clarifies that “This insurance applies to ‘bodily injury’ and ‘property damage’ only if: 1. The ‘property damage’ is caused by an ‘occurrence’ that takes place in the ‘coverage territory’

The case for CGL coverage

CGL insurance was first developed in 1941 as a way to purchase separately insured risks in one, unified policy. The Insurance Services Office, Inc. last modified the current standard form CGL policy in 1986. Its initial grant of coverage provides: “We will pay those sums that the insured becomes obligated to pay as damages because of ‘bodily injury’ or ‘property damages’ to which this insurance applies.” It clarifies that “This insurance applies to ‘bodily injury’ and ‘property damage’ only if: 1. The ‘property damage’ is caused by an ‘occurrence’ that takes place in the ‘coverage territory’ 2. The ‘bodily injury’ or ‘property damage’ occurs during the policy period.” It defines “occurrence” as “an accident, including continuous or repeated exposure to substantially the same general harmful conditions.” The term “accident,” however, is undefined. That lack of definition bears responsibility for much of the current dispute.

Consider the following example. A university contracted with a general contractor (GC) to build a new hotel on its campus. GC has a CGL policy. GC subcontracted most of the work to subcontractors. After construction was complete, the university discovered extensive water damage from hidden leaks, as well as other structural problems, all of which, it contended, arose from defective workmanship. The repair cost was estimated at $6 million.

The university sued the GC for the defective construction, and the GC turned to its CGL carrier for a defense of the case and indemnity for any judgment rendered against it. The CGL carrier argued there was no coverage because no accident (and thus no occurrence) had taken place. The court agreed with the insurer, leaving the GC with the $6 million repair bill.

In this scenario, if the GC is relatively solvent, it will file suit against its subcontractor or subcontractors seeking some or all of those repair costs. If the subcontractor has no insurance coverage and no surety performance bond, the GC and/ or the subcontractors will bear the risk of loss. If, however, the GC and/or the subcontractors has insufficient capital or credit, then some or all of the repair costs will be passed onto the owner. Regardless of who bears that cost, all of the parties are likely to complain that they believe the GC’s CGL carrier should bear such a burden because, for contractors, defective workmanship is by far the largest claim risk they face. In fact, for that reason, they are likely to say that CGL policies are purchased specifically to protect against such risks. Who is right? It will come as no surprise that the answer depends on whom you ask, or, in this case, what court you ask.

Courts’ treatment of CGL coverage for faulty workmanship

Courts throughout the country hold disparate views regarding whether defective construction is an occurrence within the confines of a CGL policy’s initial grant of coverage. By definition, an occurrence must be an accident; thereby raising the question as to whether the failure to properly construct something is an “accident.” As courts have repeatedly demonstrated, the answer to that question is not an easy one.

A minority of jurisdictions holds that defective construction is not an accident. Pointing to the everyday meaning of the term accident, these courts say that construction defects are not unexpected, because, they argue, it is foreseeable that a contractor that does faulty work would cause damages for which it would be obligated to pay the costs of repair. Because an accident must be, by its nature, unexpected, defective construction is no accident.

These jurisdictions often state that if a CGL policy provided coverage for defective workmanship, it would be converted into a warranty or guaranty of the contractor’s performance, which is more properly the province of a surety performance bond, whose sole purpose is to guaranty the contractor’s performance on the construction project. Courts in Ohio, Arkansas, Kentucky, and Pennsylvania adhere to this view and typically deny CGL coverage for construction defects, regardless of the types of damages caused by the defective work.

The majority position says that construction defects may be accidents, and thus occurrences that are eligible for CGL coverage. They are accidental because the contractor performing the work does not intend for the resulting damages to occur. Further, they argue, if CGL policies did not cover construction defects, no contractor would ever purchase them because they would provide no coverage for their single largest category of claim risk – construction defects. These majority -view jurisdictions often look at whether the damages were intended, expected or foreseeable. The focus, therefore, is on whether the results (the damages resulting from the defective work) would have been expected or foreseeable had the contractor performed correctly. If so, then it was not an accident for purposes of the CGL policy; if not, then it was.

Granting coverage for construction defects does not render as insured all such damages, however. Within these majority -view jurisdictions, there is a further split of authority as to whether the CGL policy covers only damages to a third party’s work or whether it also covers the repair and/or replacement cost of the defective work itself.

Most hold that a construction defect may be an occurrence (i.e., an “accident”) but only to the extent that the faulty work damaged property other than the insured’s defective work. These jurisdictions, which include Oregon, Illinois, Iowa, South Carolina, and Nebraska, believe that faulty workmanship itself is not an accident because the damages (the defective work) were within the insured’s control. They further justify their position on policy grounds, believing that the ultimate liability for such defective work should fall on the party that performed it, not the insurance carrier.

Other courts say that there is no basis for distinguishing between damages to the work itself and damages to other property. As the Supreme Court of Florida stated, “the definition of ‘property damage’ in the CGL policies does not differentiate between damage to the [insured] contractor’s work and damage to other property. [W]e reject a definition of occurrence that renders damage to the insured’s own work as a result of a subcontractor’s faulty workmanship expected, but renders damage to property of a third party caused by the same workmanship unexpected.” These jurisdictions, which include Florida, Kansas, Minnesota, Tennessee, Texas, and Wisconsin, find property damage to the work itself, as well as damage to the work of third parties, to be an occurrence that triggers CGL coverage.

What does the future hold?

In 2004, the Supreme Court of Nebraska stated, “the majority of courts have determined that faulty workmanship is not an accident and, therefore, not an occurrence.” [Auto-Owners Insurance Co. v. Home Pride Companies, Inc., 268 Neb. 528 (2004).] Since that case, however, courts in Florida, Kansas, Iowa, New Jersey, Georgia, West Virginia, Montana, Missouri, South Dakota, Tennessee, and South Carolina have reversed that majority, and it appears that a strong consensus has emerged that construction defects are occurrences that come within the initial grant of CGL coverage. The Ohio Supreme Court stemmed the unanimity of these authorities favoring coverage with its 2018 opinion in Ohio Northern University v. Charles Construction Services, Inc., from which the above example was taken.

Despite Ohio Northern, the clear trend favors initial coverage for defective workmanship. And, at least four state legislatures (Colorado, Hawaii, South Carolina, and Arkansas) have enacted statutes that define occurrences in the CGL context to include construction defects. Although such decisions are state-specific, they often rely on policy interpretations from other jurisdictions, a fact which suggests the majority view may continue to grow with support from new jurisdictions that have yet to consider the matter.

What does all this mean for the insurance industry and its insureds? This greater trend toward uniformity in construction defect coverage should provide optimism in the goal toward greater understanding and therefore fewer disputes among all parties (and fewer lawsuits against carriers) regarding the question of CGL coverage of damages arising from

Construction Defect Dispute Governed by Contract Disputes Act not yet Suited to being a “Suit”

William S. Bennett | SDV Insights | May 14, 2019

The Southern District of California recently held that a series of demands for a general contractor to investigate and repair several construction defects at a U.S. Army facility did not constitute a “suit” within the meaning of the general contractor’s commercial general liability (“CGL”) policy.

In Harper Construction Co., Inc. v. Nat’l Union Fire Ins. Co. of Pittsburgh, Pa., the U.S. Government hired Harper Construction Company (“Harper”) to construct a U.S. Army training facility for the Patriot Missile System in Fort Sill, Oklahoma. No. 18-cv-00471-BAS-NLS (S.D. Cal. Mar. 28, 2019). During the project, Harper hired Harper Mechanical Contractors (“Harper Mechanical”), an independent company, as a subcontractor “to perform demolition, grading, and other work at the Project.”

After Harper completed the project, the government informed Harper of property damage at the project, “including, but not limited to, gypsum wallboard cracks and binding doors.” Harper attempted to repair the issues, but the problems continued. The issues were apparently the result of Harper Mechanical’s grading work. Subsequently, the government sent two letters requesting an investigation and asking Harper to “propose a plan to correct the issues.” As Harper undertook an investigation spanning multiple years, the government became increasingly frustrated with the delays. The government threatened to initiate “formal administrative recourse” and to demolish the project, forcing Harper to re-build from the ground up. It also sent Harper another letter requesting Harper submit a formal proposal to correct the issues.

Harper’s general liability carrier was National Union Fire Insurance Company of Pittsburgh, PA (“National Union”). Harper Mechanical was listed as an additional insured on Harper’s policy. Four years after the government’s first notification to Harper of the issues with the project, Harper’s broker submitted a claim to National Union. The broker noted that Harper was seeking additional insured coverage for Harper Mechanical under Harper’s own policy for investigation and repair costs resulting from Harper Mechanical’s work.

National Union issued a reservation of rights letter and sought more information from Harper. The parties corresponded for the next year and half, until National Union issued a denial letter indicating that there was not a “suit” against Harper seeking damages because of “property damage,” based on the policy’s definition of “suit.”

The policy contained the standard ISO CGL definition of “suit,” which is defined, in pertinent part, as “a civil proceeding in which damages because of … ‘property damage’ to which this insurance applies are alleged. ‘Suit’ includes: … b. Any other alternative dispute resolution proceeding in which such damages are claimed and to which the insured submits with our consent.”

Harper sued National Union. National Union moved for summary judgment. In opposition, Harper argued that the government’s demand constituted a “suit” because the demand falls within the Contract Disputes Act (“CDA”), which includes administrative and court proceedings and qualifies as “any other alternative dispute resolution proceeding” under the policy definition. The CDA applies to “contracts made by an executive agency for, among other things, the procurement of construction … of real property.”

The court acknowledged that the CDA applied to the contract, given the Army’s status as an executive agency. However, the CDA does not automatically consider all disputes to constitute a “claim.” A dispute does not become a “claim” unless one of the contracting parties issues a “[w]ritten demand or written assertion … seeking … the payment of money in a sum certain,” at which point “each claim by the Federal Government against a contractor relating to a contract shall be the subject of a written decision by the contracting officer.” Without the claim being “submitted for a written decision by the contracting officer, which is the first step in the dispute resolution process under the CDA,” the court determined that there was “no evidence that Harper was faced with a “civil proceeding in which damages … are alleged” or “any other alternative dispute resolution proceeding,” as required by the policy’s definition of “suit.” The court also noted that there was no evidence that National Union had consented to any of the processes involved in the dispute, which is a further requirement of the definition of “suit.”

The court granted summary judgment for National Union based on the conclusion that the CDA demands did not constitute a “suit.” This case is an unfortunate example of what can happen when a contractor does not consider coverage when making strategic decisions throughout the process of investigating and repairing construction defects. The result could potentially have been favorable for Harper had it notified National Union early (and often) of the issues, involved coverage counsel to work with its defense and/or general counsel to strategize about how to cast the proceedings as a “suit” under the CDA, and followed the proper channels under the CDA to solidify its position that the parties were involved in ADR proceedings under existing California law.

The Limited Scope of the “Care, Custody or Control” Exclusion

Kirk Pasich | Pasich LLP | June 10, 2019

General liability policies have long contained an exclusion for property “in the care, custody or control of the insured . . . .” Commercial General Liability Coverage Form, § I., Coverage A, ¶ 2.J.(4) (Insurance Services Office, Inc. 2012). The scope of this exclusion has been the subject of various court decisions over the last 60 years. However, on June 5, 2019, a California court of appeal rendered a decision confirming the narrow scope of this exclusion.  

In McMillin Homes Construction, Inc. v. National Fire & Marine Insurance Co., 2019 WL 2366468 (Cal. Ct. App. June 5, 2019), a general contractor was an additional insured on a general liability policy issued to a subcontractor. The insurer refused to defend the general contractor in a suit brought by homeowners for construction defects relating to work performed by the subcontractor. It argued, among other things, that its “care, custody or control: exclusion applied because the worksite was within the contractor’s control. The court disagreed.  

The court considered various earlier decisions. In doing so, it cited Home Indemnity Co. v. Leo L. Davis, Inc., 79 Cal. App. 3d 863, 870-71 (1978), noting, “‘Almost invariably where coverage is denied, physical control by the insured has been exclusive, even if such exclusivity was only momentary, so long as the damage occurred in that moment.’” Id. The court followed this rule, noting that the subcontractor “was responsible for controlling its jobsite and supervising the roofing work,” but the general contractor “was responsible for the whole project and coordinating schedules to ensure the project finished on time.” McMillin, at *8. 

Although the court found the exclusion to be unambiguous, it then addressed the interpretation of the exclusion assuming that it was ambiguous. Once again, it rejected the insurer’s argument. It noted that because construction defect litigation “‘is typically complex and expensive, a key motivation in procuring an additional insured endorsement is to offset the cost of defending lawsuits where the general contractor’s liability is claimed to be derivative.’” Id. Therefore, the court concluded that reading the “care, custody or control” exclusion “in a manner that nullifies the broad coverage provision for a general contractor sued for construction defects is not consistent with an insured’s objectively reasonable expectations.” Id. at *9. It emphasized that the insurer’s construction “bears little connection to the risk involved or the reason for a general contractor to seek coverage as an additional insured. Its stance might be ‘reasonable in the abstract,’ but it is inconsistent with the basic rule that limitations on a promised defense duty must be conspicuous, plain, and clear.” Id

McMillin stands as a clear statement that an “additional insured’s mere status as a general contractor—with overall responsibility for and nominal control of the entire project—does not meet [the] standard” of “exclusive or complete control” sufficient to trigger application of the “care, custody or control” exclusion. Id. at *5. 

“Rip-and-Tear Damages” In Construction: A Roadmap For Coverage Where None Existed?

Ashley Veitenheimer | Kane Russell Coleman Logan | May 22, 2019

The insuring agreement in most commercial general liability policies states that the carrier “will pay those sums that the insured becomes legally obligated to pay as damages because of…’property damage’ to which this insurance applies.” In addition, most policies exclude coverage for the defective work of the named insured. Questions have arisen, however, as to whether and when there is coverage for damages commonly known as “rip-and-tear,” which are those damages caused to other property by the necessity of removing, replacing, and correcting defective work.

Prior to 2015, Texas law held that rip-and-tear damages were covered if there was underlying covered property damage in the first instance. See Lennar Corp. v. Markel Amer. Ins. Co., 413 S.W.3d 750 (Tex. 2013). That all changed with U.S. Metals v. Liberty Mutual Ins. Group, Inc., 490 S.W.3d 20, 22 (Tex. 2015). In U.S. Metals, the Court appears to hold that damages are covered even when they are not “because of” property damage, leading to vexing issues for the insurance carrier regarding when the duty to defend is triggered and whether rip-and-tear costs are covered when they are not “because of” property damage.

In U.S. Metals, U.S. Metals sold ExxonMobil 350 flanges for use in constructing diesel units. When ExxonMobil conducted post-installation testing, it discovered that several flanges leaked and did not meet industry standards such that it was necessary to replace them to avoid the risk of explosion. For each flange, this process involved stripping the temperature coating and insulation (which were destroyed in the process); cutting the flange out of the pipe; removing the gaskets (which were also destroyed in the process); grinding the pipe surfaces smooth for re-welding; replacing the flange and gaskets; welding the new flange to the pipes; and replacing the temperature coating and insulation. 

After ExxonMobil sued U.S. Metals and the parties settled, U.S. Metals sought indemnification from its insurer. On appeal, the parties disputed whether the installation of the faulty flanges physically injured the diesel units.  The Court noted that “the installation of the leaky flanges…can certainly be said to have injured – harmed or damaged – the diesel units by increasing the risk of danger from their operation and thus reducing their value.” However, no physical injury resulted because ExxonMobil replaced the flanges in order to avoid the risk of such injury.

The Court concluded that the diesel units were physically injured in the process of replacing the flanges because the flanges were welded to the pipes, and the removal process “necessitated injury to tangible property, and the injury was unquestionably physical.”  That tangible property was the original welds, coating, insulation, and gaskets. Because the diesel units were restored by replacing the flanges, they were impaired property to which Exclusion M applied.[1] Id. But it also concluded that the insulation and gaskets were destroyed in the process and replaced such that Exclusion M did not apply. Therefore, the Court held that these rip-and-tear costs, were covered because the items were physically injured and constituted “property damage.”

After U.S. Metals was decided, the Western District of Texas issued an opinion illustrating the problems the holding created. In Travelers Lloyds Ins. Co. v. Cruz Contracting of Texas, LLC, the Western District relied on the U.S. Metals holding to conclude that rip-and-tear damages were covered. 2017 WL 5202891 (W.D. Tex. Sept. 7, 2017). There, Cruz, the subcontractor, was hired by D & D to install utility systems, which were later discovered to be faulty. D & D alleged that, in order to replace the sewer system installed by Cruz, it had to tear out and redo roadways, curbs, and parkways.

Based on U.S. Metals, the court found that D & D suffered property damage in the form of rip-and-tear damages “to access faulty equipment installed by an insured…”. The problem with this conclusion is that no damages “because of” property damage existed prior to the rip-and-tear process being undertaken. Rather, as the court concedes, the adjoining utility work was “not physically disturbed by Cruz’s defective work” but was “rendered useless by the defective work.” Consequently, the court apparently relied upon the loss of use as the trigger for the insurer’s duty to defend.

This, in turn, raises the pivotal issue of when the alleged property damage actually occurs. In other words, since there was no “property damage” prior to the tear-out and replacement of Cruz’s work – there was merely faulty work (which is typically excluded from coverage) – when did the “covered” property damage occur? The court’s opinion states that the property damage “occurred when the utility systems installed by Cruz failed testing, rendering them inoperable and unusable.”. Although the court relies upon Don’s Buildingfor this proposition, this is a rather questionable conclusion because there was no property damage prior to the damage caused in accessing the faulty work.[2]

Take as an example pipe work that is performed before pouring a concrete floor. No damage exists at the time the pipes are installed; however, there is later discovered a leak in one of the connections that requires replacement. If suit is filed merely alleging that the pipe was faulty and that the concrete needed to be torn out, is this sufficient to trigger a duty to defend in Texas because the rip-and-tear is in itself property damage? And, if so, does the insurer for the connection supplier owe a duty to defend the entire lawsuit when the concrete flooring, pipes, and other building components are damaged in an effort to repair and replace the connection? If that is the case, almost every suit for construction defects may plead a covered claim because it will involve rip-and-tear costs.

Equally confounding it the issue of “when is the occurrence.” If the rip-and-tear is itself the “property damage,” then can an insured create its own trigger for defense by alleging that the installation was improperly performed and required the rip-and-tear damages to replace the faulty connection? These are the questions created by the holding in U.S. Metals that have yet to be answered, but the Cruz holding certainly got this issue wrong. That is because U.S. Metals clearly identifies when the occurrence is:

We have further held that, for purposes of a duty to defend under an occurrence-based policy period, damage due to faulty workmanship “occurs” not at the time the damage manifests (when it is discovered or discoverable) nor when the plaintiff is exposed to the agent that will eventually cause the damage (when it is installed, presumably). Rather, under a straightforward reading of the policy, we concluded that “[o]ccurred means when damage occurred, not when discovery occurred.” Since a defective product that causes damage is not an occurrence until the damage actually happens, it would be inconsistent to now find that a defective product that does notcause damage is nevertheless an occurrence at the time of incorporation.

Cruz, however, held that the “occurrence” happened when the utility systems failed testing without any related property damage. This is one example of the myriad of questions created by the U.S. Metalsholding, relied upon by the Cruz court, and the lower courts’ application of the ruling, which may create the potential for a huge shift in coverage law as to when the duty to defend is triggered.


[1] Exclusion M denies coverage for damages to impaired property, which is defined as property that can be “restored to use by the…replacement” of the faulty flanges. Id.

[2] Interestingly, the Southern District relied upon U.S. Metals to conclude that rip-and-tear damages are covered when the utility of component parts is destroyed “[a]s a consequence of their having been encased in bad concrete.” See Lauger Cos., Inc. v. Mid-Continent Cas. Co., 2017 WL 8677353 (S.D. Tex. Aug. 2, 2017). This creates the same problems as Cruz and gives rise to the same concerns of when the duty to defend is actually triggered.

The 10th Circuit Correctly Construes “That Particular Part” Narrowly

David Smith | Farella Braun & Martel | April 12, 2019

We do not often write about coverage opinions from jurisdictions as far away as Oklahoma; however, a recent case from the Federal Tenth Circuit looked at one of our favorite topics and came out with a much better reasoned opinion than recent decisions from the Ninth Circuit.

I’ve written before on the topic of the meaning of “that particular part” as the phrase is used in exclusions j (5) and j(6) of the Commercial General Liability (“CGL”) policy.  The “j” exclusions exclude coverage for damage to certain property.  Specifically, the j (5) and (6) exclusions state that the insurance does not apply to:

(5)    That particular part of real property on which you or any contractors or subcontractors working directly 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.

The part of these exclusions that some courts consistently get wrong is the meaning of the phrase “that particular part.”  In particular, in June 2017 I wrote about the way the Ninth Circuit (supposedly applying California law) has on several occasions ignored the insurance industry’s own explanation of the meaning of the phrase “that particular part” and applied the exclusion to the entire project a contractor was working on.

For example, the insurance industry’s own text books give the example of a general contractor dropping a steel roofing beam during the construction of a steel-frame building. The dropped beam fell while being hoisted and “struck and damaged other steel components that had already been erected.” Because at the time the accident occurred the contractor was working on the beam that dropped, and not the other damaged building components, the damage to those building components was intended to be covered because it was not the “particular part” being worked on. Donald S. Malecki, et al., Commercial Liability Insurance and Risk Management, 95-96 (3d ed. 1995) (“No coverage applies to ‘that particular part’ of real property damaged while work is being performed on it. But, by inference, coverage does apply for damage to any other part of the property besides ‘that particular part’….”) (emphasis in original). A number of other industry writings support this analysis, including materials published by the drafters of the actual policy wording, the Insurance Services Office.

California law specifically allows the use of industry materials and resources to interpret insurance policies. See, e.g., Maryland Cas. Co. v. Reeder, 221 Cal. App. 3d 961, 973, n.2 (1990) (relying on industry publications to interpret the scope of exclusions in the 1973 CGL policy). Furthermore, California Courts of Appeal have already applied the exclusions narrowly. See, e.g., Pulte Home Corp. v. American Safety Indemn. Co., 14 Cal. App. 5th 1086 (2017); Global Modular, Inc. v. Kadena Pacific, Inc., 15 Cal. App. 5th 127 (2017).

Additionally, when a policy clause is ambiguous and capable of at least two reasonable interpretations, the court is supposed to construe the clause against the drafter – which in insurance contexts, is invariably in favor of coverage. This is true nationwide. See, e.g., Foster-Gardner, Inc. v. Nat’l Union Fire Ins. Co., 18 Cal. 4th 857, 869 (1998); Haworth v. Jantzen, 172 P.3d 193, 197 (Okla. 2006).

A recent case out of the Tenth Federal Circuit looked at the “that particular part” issue in a published opinion, MTI, Inc. v. Employers Insurance Company of Wausau, __ F.3d __, 2019 WL 321423 (10th Cir. 2019), applying Oklahoma law. MTI, a contractor, was hired to replace corroded anchor bolts and castings in a wooden cooling tower. The contractor removed the old bolts but was delayed in installing the new ones. High winds blew the cooling tower down while it was left unsupported.

MTI’s insurer, Wausau, denied MTI’s claim for the damage caused to the cooling tower and equipment inside, citing exclusions j (5) and j (6). MTI sued Wausau. Noting that the Oklahoma Supreme Court had not explicitly addressed the issue, the court looked at the way other jurisdictions had interpreted the clauses, finding that courts had ruled both ways (i.e., some courts held that the language restricted the exclusion’s application to only the part of a project being worked on; others applied it to the entire project).

Holding that both interpretations were reasonable, the MTI court held that the exclusions were ambiguous and thus must be construed against the insurer. The court also opined that the insured’s reasonable expectations would be for coverage for the cost of replacing the entire tower. While from a purist’s point of view it’s unfortunate that the MTI court did not look at the insurance industry resources, it nevertheless reached the right conclusion, adding to the majority of jurisdictions that interpret these exclusions narrowly. We can only hope that eventually the weight of opinion (together with California law) will sway the Ninth Circuit on this issue.

As an aside, the court also dismissed Wausau’s argument that providing coverage of the tower converted the CGL policy into a performance bond. This is a red herring argument often used by insurers. The court noted that a performance bond guarantees completion of a contract upon the contractor’s default, while a CGL policy spreads the contractor’s risk. The monetary values at issue underlined this point: the value of the contract was $46,000; the value of the damage was $1.4 million. A performance bond for such a project would not result in liability of this amount.