Construction projects are always the subject of contracts among owners and contractors. Does an exclusion in the standard Comprehensive General Liability insurance policy that precludes coverage for “liabilities assumed under contract” apply to construction contracts? According to a recent, and long-awaited, decision of the Texas Supreme Court: Usually not.

To the uninitiated, it would appear counterintuitive. To the initiated: Not so much.

The laws of most states require that a construction project be governed by a written contract between the owner and the contractor. The provisions of most standard-form CGL policies preclude coverage for “liabilities assumed under contract.” If a construction project goes awry because of alleged faulty workmanship, or the failure of a contractor to perform in a “workman-like manner,” why wouldn’t this failure — presumably addressed in the provisions of the construction contract — be precluded from coverage by the contractual liability exclusion in the policy? The answer to this question lies in an exception to the exclusion that brings damage arising from certain kinds of contract-related claims back into coverage. It’s quite common, incidentally, for exclusions to contain exceptions that give back coverage that the exclusion would otherwise take away. It is one of the many aspects of complex commercial insurance policies that make them maddeningly hard to follow and understand.

The standard contractual liability exclusion provides as follows:

This insurance does not apply to:

“Bodily 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 for 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.”

The policy generally — though not universally — defines an “insured contract” narrowly as, for example, a “lease agreement” or a “sidetrack agreement” with a railroad.

It’s quite common for exclusions to contain exceptions that give back coverage that the exclusion would otherwise take away

Court dockets across the country are full of cases involving faulty construction claims by owners against contractors. These cases typically allege that the contractor both failed to comply with contract provisions that required the project to be done in a “workmanlike manner” and that the contractor or its sub-contractors were negligent in their construction of the project. Thus, there is usually a straight “contract” claim and a straight “tort” or negligence claim at issue in each of these suits. Two questions, then, arise when a carrier tries to get out of coverage on the basis of the Contractual Liability exclusion. First, what does it mean to “assume liability in a contract?” Second, when would the insured have liability “in the absence of the contract?” The Texas Supreme Court answered both of these questions.

Ewing v. Amerisure Ins. Co., No. 12-0061 was decided on January 17, 2014 (get a copy here). For those who reside in the insurance and construction subcultures (and in the even smaller, but still significant, “coverage-for-construction-defects subculture”) , the Ewing decision has been much-anticipated and a long time in coming. The wait has left contractors and insurers in Texas — and to a lesser extent, across the country — in limbo about whether insurance companies might be able to invoke the contractual liability exclusion in a very broad class of construction-defect coverage claims.

The drama (if the incongruity of using the words “insurance” and “drama” in the same breath can be excused) began nearly six years ago. Ewing Construction Company built a set of tennis courts in 2008 at a school in Corpus Christi. Shortly after construction was complete, according to the school district, “the courts started flaking, crumbling, and cracking, rendering them unusable for their intended purpose of hosting competitive tennis events.” 

The school district sued Ewing in Texas state court. It claimed that Ewing had breached its contract and that it had built the tennis courts in an unworkmanlike manner. Ewing turned the suit over to Amerisure, its CGL insurer, seeking a defense and indemnity. Amerisure denied all coverage, citing the contractual liability exclusion in its policy. This inspired Ewing to sue the carrier in federal district court for the Southern District of Texas. 

The parties filed motions for summary judgment against each other. The district court denied Ewing’s motion and granted Amerisure’s, dismissing Ewing’s claims on the basis of the contractual liability. The court relied on a prior Texas Supreme Court case that had interpreted a similar exclusion, Gilbert Texas Construction, L.P. v. Underwriters at Lloyd’s London, 327 S.W.3d 118 (Tex. 2010). The district court concluded that Gilbert “stands for the proposition that the contractual liability exclusion applies when an insured has entered into a contract and, by doing so, has assumed liability for its own performance under that contract.”

In point of fact, the Gilbert decision was significantly narrower than the district court’s interpretation of it, but that’s getting ahead of the story. The district court in Ewing also considered the exception to the exclusion and concluded that it did not apply because the school district’s claims against Ewing sounded only in contract, not in tort (negligence), and that, since Ewing had contracted to be liable for failing to perform under the contract, the case did not entail any liabilities that would have existed in the absence of the contract, itself.

Thus far, it was not looking good for Ewing or for any other construction contractor in Texas.

Ewing appealed to the United States Court of Appeals for the Fifth Circuit, which is the court in the federal system just below the United States Supreme Court. The Fifth Circuit initially affirmed the district court’s decision against Ewing and in favor of Amerisure. There was, however, a dissenting opinion on the Fifth Circuit’s three-judge panel. Ewing made a motion for the court to rehear the case. The Fifth Circuit then withdrew its opinion and certified the question at issue in the case to the Texas Supreme Court. (In cases such as this, federal courts are required to apply the law of the state in which the federal district court sits and, when questions of state law are unsettled, the federal courts will sometimes “certify” or ask the state supreme court to resolve the issue.) That is where things stood about 19 months ago.

Insurance companies know how costly it can be to resolve coverage disputes while litigating an underlying claim at the same time

Here it is worth pausing to consider the time, effort, and expense that went into getting the case to this point of standstill. Because of the position Amerisure took with its insured, the policyholder had to fight a protracted and, presumably, very expensive, two-front war — one with the school district, in which it had to defend itself instead of relying on the defense coverage in its policy, and the other with its CGL carrier. Litigating one case against a determined adversary is terribly expensive. The expense of litigating two cases at the same time can take one’s breath away. Insurance companies know how costly it can be to resolve coverage disputes under such trying circumstances. The expense, the diversion of resources, the uncertainty, and the often discouraging outcome of intermediate court rulings are all powerful incentives for insureds to avoid coverage litigation altogether, to give up half-way through, or to settle for much less than they would otherwise be entitled to recover. It often takes a Ewing Construction Company, with the apparent resources and the assiduity to keep fighting in the face of daunting odds and dispiriting set-backs, to see such litigation through to the very end for the benefit of itself and for other policyholders.

The Texas Supreme Court first looked at the nature of the claims the school district had brought against Ewing, actually quoting from the complaint. The first was purely a contract claim, alleging that “Defendant Ewing Construction has breached its contractual commitments, proximately causing damages to Plaintiff.” The second was a negligence claim: “Defendant Ewing Construction and/or its subcontractors was/were guilty of negligence proximately causing damages to Plaintiff.” Let’s pause and consider, once again, the frustration and discouragement that litigation of this kind can involve.

Recall that the district court’s decision dismissing Ewing’s coverage case was based on a finding that the school district’s claims against Ewing sounded purely in contract and not in tort. Yet, there it is, in plain English in the underlying complaint: a claim that Ewing was allegedly “guilty of negligence proximately causing damages to Plaintiff.” Imagine how disheartening it must have been for Ewing and its attorneys to lose in the district court on the ground that there was supposedly no negligence claim in the underlying complaint. Then try to imagine the frustration of having that decision affirmed by the Fifth Circuit. It takes a stout and unyielding constitution — not to mention a very deep pocket — not to succumb to the urge of running up the white flag in the face of two such decisions.

The Texas Supreme Court stated the certified question as follows: “Does a general contractor that enters into a contract in which it agrees to perform its construction work in a good and workmanlike manner, without more specific provisions enlarging this obligation, ‘assume liability’ for damages arising out of the contractor’s defective work so as to trigger the Contractual Liability Exclusion?” The phrase in the question, “without more specific provisions enlarging this obligation” is a key to the Ewing decision. The phrase provides insight into the boundaries of the contractual liability exclusion.

Recall that the district court’s decision favoring Amerisure had cited the 2010 decision in Gilbert Texas Construction, L.P. v. Underwriters at Lloyd’s London. In that case, the contractor had agreed in writing to do something more than simply perform in a good and workmanlike manner. It had also agreed to protect against, and to pay for, any damage to property adjacent to the job site. Predictably, it happened during the course of the construction that some harm had come to property adjacent to the job site and the owner sued Gilbert for the damage. Although Gilbert had a common-law obligation to perform its work competently and without negligence, it did not have a common-law duty to the owner to avoid damage to the adjacent property. That was an obligation it had assumed solely under the contract and the obligation was beyond the liability it would have had in the absence of the contract. It was under these circumstances that the Texas Supreme Court held that the claim for coverage as to the damage to the adjacent property was subject to the contractual liability exclusion in Gilbert’s policy.

The contract Ewing had signed required the contractor to build the tennis courts in a good and workmanlike manner, and nothing more. According to Ewing, its agreement with the school district therefore “did not add anything to the obligation it has under general law to comply with the contract’s terms and to exercise ordinary care in doing so.” The Texas Supreme Court agreed.

The word “assumption” in the exclusionary phrase “assumption of liability under contract” means something more than simply signing a contract that requires what is already required by law. “Assumption” must add something to the phrase “assumption of liability under contract.” Otherwise, the phrase “assumption of liability” would be nothing but surplusage. According to the Ewing court, the contract claims that Ewing failed to perform in a good and workmanlike manner “are substantively the same as its claims that Ewing negligently performed under the contract because they contain the same factual allegations and alleged misconduct.” Failure to perform in a “good and workmanlike manner” is functionally and substantively the same as performing negligently. “Accordingly,” the Ewing court said, “we conclude that a general contractor who agrees to perform its construction work in a good and workmanlike manner, without more, does not enlarge its duty to exercise ordinary care in fulfilling its contract, thus it does not ‘assume liability’ for damages arising out of its defective work so as to trigger the Contractual Liability Exclusion.”

This result is not especially striking. It would have been far more newsworthy had the court ruled in a way that was consistent with the district court or the Fifth Circuit. If the Contractual Liability exclusion was meant to exclude all claims arising from contract, the insurance industry would never have included the exception for liability the insured would have in the absence of the contract. The Ewing decision is fully consistent with the way the majority of courts around the country have resolved the same question. That doesn’t make it any less valuable to Ewing and other Texas contractors. Nor does it mute the sigh of relief among policyholder counsel, who know only too well that this insurance stuff isn’t easy and that courts often get it wrong.

There is one other aspect of the Ewing decision worth mentioning. Although the issue was not relevant to its holding, the Ewing court wrote an epilogue of sorts addressing the carrier’s claim that failing to exclude coverage for faulty workmanship would turn the CGL policy into a “performance bond.” This is a fatuous (or at least perhaps a disingenuous) argument that carriers frequently make in construction-defect coverage cases. Anyone who works in the insurance arena should know better.

The head-on response to that argument is that, regardless of the coverage provided by a CGL policy, it could never be confused or conflated with a performance bond because the two kinds of coverage are totally different. They don’t even cover the same insureds. A performance bond is issued to, and covers, the owner of the project and secures only the owner’s right to a timely and competently built project. A CGL policy covers the contractor against claims by third parties. The two kinds of insurance cover different insureds for different risks, and they exist for entirely different purposes altogether. When a CGL policy pays out in defense or indemnity, it pays the insured contractor. When a performance bond pays out, it pays the owner. To say that finding coverage for faulty workmanship under a CGL policy transforms it into a performance bond is like saying that finding coverage for a hospital in a case of negligence by one of its doctors transforms the hospital’s third-party liability policy into the doctor’s malpractice policy. No insurance lawyer would ever make such a claim with a straight face.

Instead of meeting this argument head-on, the Ewing court merely pointed out that it was not transforming a CGL policy into a performance bond because other exclusions in the CGL policy might be applicable to faulty workmanship claims. That’s fair enough, as far as it goes. But it doesn’t get to the heart of the matter.

Nonetheless, it’s hard to find fault with a decision that explains a counterintuitive insurance concept in a way that makes sense even to the uninitiated.