Construction defect (CD) claims are unusually complex. The staggering cost of litigation related to CD has hammered insurers underwriting residential construction in high-growth states including California, Nevada, Arizona, Colorado, Florida, the Carolinas, Texas, and Washington. Residential construction includes condominiums, townhouses, and planned communities consisting of multiple single-family dwellings. Whenever a housing area experiences a lot of population growth and expansion, there is a huge demand for new housing. Despite the recent credit crunch and the mortgage lending crisis, the U.S. housing boom is still going strong in many areas.
Unfortunately, there are never enough quality tradesmen to do the job properly. Builders sometimes turn to unskilled labor or cut corners in other ways. They may not inspect each home as it is built, an expensive proposition in a tract development. The result is predictable: Issues crop up due to shoddy workmanship, if not sooner, then later.
Even so, without a hostile litigation environment, there would be less of a problem. In California, where the issue arose in the 1980s, the plaintiffs bar has been very knowledgeable and well-versed in ways of extracting money from builders and insurers—something that has not yet happened in states like Georgia. In fact, CD has become a cottage industry. Even though they are on the side of the carriers, legal defense firms make even more money than the plaintiffs. Making the CD problem go away is not in the interest of either group.
As a result, CD has become the second-biggest insurance expense for builders, behind workers' compensation. But a solution—or at least a coping mechanism—may be at hand. A combination of actuarial scrutiny, savvy underwriting, and effective claims analysis can lead to better management of CD risk.
What is construction defect?
CD is one of two classes of claims that fall under general liability insurance for residential homebuilders. Sudden or accidental claims that result from premises or operations exposures are the other type.
A CD claim is typically related to a defect in design or construction material that does not cause a problem until years after the home was built—sometimes seven years or longer. Claims are typically based on problems that are perceived as dating back to the time the dwelling was built.
Take, for example, a window installer who uses an improper sealant. Over time, this may allow water intrusion that ultimately damages the internal studs as well as the interior of the home, resulting in the need to replace or respackle the sheet rock.
But here's the hitch: Not all damages are covered by insurance. In this example, the insurer pays for the result of the window installer's faulty work but not for his own work product. That means that if the windows need to be replaced, the insurer is not liable. But if the water seepage damages the wall or creeps into the floor, the insurer foots the bill. Of course, regardless of whether or not the builder has insurance, it still has a responsibility to the homeowner to tend to the problem.
How much of the damage is attributable to work product and how much is resultant or consequential damage? That's the million-dollar question. It's part and parcel of a CD claim, but it's not as straightforward as it may seem. This is why a damage investigation on these claims is more important than a liability investigation. It's not uncommon for plaintiffs to write up 200 pages of damages—all of which must be rebutted by an insurer's own experts. Legal expenses snowball and usually exceed the indemnity or losses.
But it's not always the insurer or general contractor (GC) that absorbs those expenses. Historically, general contractors' policies have not covered subcontractors. That said, subcontractors have often agreed to name the GCs and developers as additional insureds on their policies. In addition, their subs' contracts require them to indemnify and hold developers or GCs harmless for damages arising out of their work. The GCs and developers, in turn, have tendered their defenses for all their subcontractors. The result is that the so-called "mow and blow" subcontractors have paid a huge amount of the expense but incurred very little of the loss.
How does this happen? Let's assume that damages occur because of leaks in the roof or windows. The subcontractor that did the landscaping is brought into the lawsuit under the terms of his subcontract. It is clear that the landscaper is not at fault, but he has to pay defense costs or a portion of the defense costs just like all the other subcontractors—some of whom may have had a real role in causing the damages.
This situation reflects a significant mismatch between expenses and losses. But expenses, of course, become losses, because the carrier is often willing to pay to get out from under the expense burden.
Compounding the issue is the popularity of wrap policies, under which a single insurer covers construction work performed by all parties. Wrap policies have helped to streamline claims processing and reduce the frequency of claims, but because what would have been several different claims are now bundled into one monster claim, the policies have also increased the severity of the CD claims that do appear. According to a developer in Northern California, insurance for CD using a wrap policy cost him "$20,000 per door."
Estimating liabilities: A tricky business
A builder's obligation to a homeowner starts when he builds the house. The clock starts ticking when he actually sells the home and continues for a period of time equal to the statute of limitations. In some states, the statute of limitations or repose is 12 years—or more.
There can be a reporting lag of up to six years between the time a policy is issued and when a CD claim has been reported and all the facts of the case are known. If a lawsuit ensues, a discovery period can take another two years before an insurer knows what kind of exposure it faces, and it may be another three years before the case settles. In the end, it may be 15 years from the time a policy is issued to the time a claim against it is settled.
But that's just the beginning of how difficult it may be for an actuary to assess the CD risk. It’s not uncommon for a builder to switch insurers during the period in which a statute of limitations or repose lasts. So another key question that potentially muddies the waters is how to determine which insurer is responsible for what.
One reason it's so difficult for actuaries to estimate an insurer's past liability or forecast it into the future is that most clients don't have 15 years of data—nor is there very good benchmark data available. Also, policy terms change frequently, so losses that were covered in the past may not be covered on newer policies.
So what's the upshot? Even with a benign line of business like personal automobile liability, which is normally predictable, actuaries still just estimate losses. There is always a range around that estimate due to uncertainty. But the uncertainties related to CD-based litigation exacerbate the uncertainty of any "best estimate." Reasonable reserve ranges sometimes vary by 50% or more. In addition, observed loss ratios in some states have been as high as 400%—and reported loss ratios reach higher still in the U.K.
Underwriting CD: Understanding risk
Mitigating the nightmare that actuaries face when they try to estimate a carrier's liability begins with the underwriter. Given the disparity from one state to another, any CD underwriting activity must begin with a thorough understanding of the situation in the state for which coverage is being written.
The CD underwriter needs to understand the risk environment, including the following:
- The extent of involvement of plaintiffs' law firms
- The specific state's case law and statute of limitations
- The state's "right to repair" laws, which require homeowners and their builders to communicate before a lawsuit is filed
- "Your work" exclusions, which preclude coverage for property damage caused by a breach of contract
- The situation in "continuous trigger" states, in which the insurer is hit by claims on houses built in prior years that the carrier did not intend to cover
All these state-by-state specifics make it difficult to even define an occurrence of CD. In fact, any attempt to formulate a uniform approach to CD has been thwarted by a lack of agreement over what constitutes construction defect and when it occurred.
A savvy underwriter then faces a tough question: How to manage CD risks from an underwriting perspective?
Underwriters can begin by pricing realistically—at levels commensurate with exposure. In the process, CD coverage should be engineered defensively. Examples of this include:
POLICY LANGUAGE. Changing the wording to better define or narrow exposure can reduce the carrier's risk. The price of coverage can also be altered, in some cases by as much as 25% to 40% below or above the state's manual rate.
POLICY EXCLUSIONS. Common CD exclusions for synthetic stucco and mold have contributed to the litigation explosion and massive losses that carriers have accrued.
RECOGNIZING THE COST OF DESIGN-BUILD. This approach merges the design and construction phases of a project, thereby decreasing the overall time required. This approach carries new CD risk, however, as design subcontractors are introduced into the equation.
WARRANTIES. The CD warranty solutions offered by some insurers are agreements between the builder and the customer that constitute a good-faith effort to deal with as much damage as possible via a warranty mechanism, thereby limiting the amount of litigation that takes place and reducing the overall cost impact of CD.
Once the policy is written, underwriters can insist on certain loss- and risk-control mechanisms—essentially ensuring appropriate attention to risk throughout the process. These may include an evaluation of subcontractors and the GC's relationship with them, a third-party peer review of new buildings before they are turned over to the owners, and an increased focus on the builder's own quality control and customer service functions.
Finally, it is essential that underwriters learn from past mistakes by analyzing claims data. Underwriters generally review between seven and 10 years of claims, which can reveal how a builder has changed the practices responsible for those claims.
Controlling risk through improved claims analysis
Improving the claims analysis process itself can also help. A major reason that insurers have suffered heavy losses on CD is that claims personnel do not always use the endorsements and exceptions that were put in place to protect the insurer. If carriers don’t employ people who understand these exclusions and endorsements and possess the skills to use them, they will end up in unnecessary litigation resulting in unnecessary loss and payments. In addition, the investigation conducted to determine a policyholder's involvement on a project is frequently inadequate and results in delays in estimating appropriate case reserves.
That said, certain endorsements are further complicating the CD claims environment. The big culprit here is "additional insured" endorsements. When developers or GCs hire tradesmen or subs, they require that they be named as additional insureds on the sub's policy. When the developer or GC is served with a suit, the additional insurer endorsement wraps those subs into the claim—even if they had nothing to do with the damage. Thus, every suit facing the developer or GC becomes the problem of every sub that worked on the project. Often, the subs do not end up paying any loss, but they will pay expenses because they are still involved in the litigation process. This situation has created a kind of CD cost shift. Developers' costs are down 20% to 30%; GCs, subs, and others have seen their costs go up.
To date, courts have ruled that additional insured endorsements are unambiguous. There is no way around them for GCs or subs. The prevalence of these endorsements across the CD universe complicates what already was a convoluted claims atmosphere.
In light of this complexity, what can be done to control risk at the claims end? A few questions need to be asked:
- Can lessons to be learned from claims data?
- When faced with a claim, does the builder perform its own damage assessment?
- If data is available, what is the typical lag between policy issuance and claim occurrence? How much time passes between the occurrence of a claim and when it is reported? How long until settlement?
- Is the insurer's or organization's customer service acceptable to claimants, or is it creating friction that might escalate a claim?
- How might warranties be used to help control claims?
CD presents a difficult-to-manage risk, one that continues to evolve in the face of tort reform, building booms, and increased attention to the complexity of the problem. Because claims can appear years after construction is completed, the CD trend shows no sign of a slowdown. As owners, general contractors, and subcontractors wrangle over who owns the risk, actuaries, underwriters, and claims specialists try to stay several steps ahead of them. They should all get used to the shuffle: The complication is here to stay.
WILLIAM R. AZZARA is a principal and senior consultant with the Princeton, N.J., office of Milliman. Bill has 40 years of claims management experience, including senior claim management positions at Liberty Mutual and American Re-Insurance Company. He has been involved in the handling of all primary and excess casualty and property lines of business, including workers' compensation, product liability, professional liability errors and omissions, and medical malpractice. His consulting experience includes evaluation of claims-handling practices and procedures, assessment of litigation management/expense controls, and management of due diligence reviews related to mergers and acquisitions.