Insurance: The Liability Messenger
DENNIS R. CONNOLLY
At one point in the 1980s, there was a generalized "hard" insurance market. Insurance for a whole range of liability exposures, including product liability, was either very expensive or hard to come by, or both. Today, however, the insurance market is soft overall. In fact, the industry has the ability to write more insurance than it is currently writing. Even so, insurers are choosing not to divert this surplus capacity to cover certain classes of product liability risk, especially in industries such as pharmaceuticals, chemicals, automotives, and aviation.
Others have written eloquently about how costly U.S. product liability exposures are to consumers and manufacturers, principally by stifling innovation and competitiveness. The intent of this paper is to explain the process by which these risks become expensive to insure, or simply uninsurable. This is an important aspect of the product liability problem because, all too often, insurers are blamed for denying industry the affordable protection it needs as it goes about the business of innovating. In fact, both the blame and the solution lie elsewhere.
INSURANCE AS MESSENGER
To appreciate the role of the insurance system in managing corporate liabilities, it is helpful to think of the system as a message-bearer. Insurers send a message to insurance buyers whenever they set premium rates and establish the terms and conditions of coverage. When, for example, a company manufactures a product that causes injury to a consumer, compensates the consumer for his or her loss, and is itself indemnified under an
applicable insurance policy, the insurer may then turn to the buyer and, in effect, say, "Your product has caused this problem, so we're going to raise your insurance premium."
Insurance buyers that fail to get the message and take the necessary precautions to reduce future losses will get harsher and harsher messages along these lines. When, for whatever reason, loss experience is not turned around, one of two things happens: either buyers reach a point where they cannot afford to pay for the coverage they need, or sellers reach a point where they cannot afford to extend the needed coverage at any price.
The fact is that today there is a strong sense among insurers that they cannot afford to insure parties for injuries and property damage caused by certain products and substances. For insurers, the known costs are too high and the potential costs are too uncertain.
UNDERWRITING: RESPONSIBLE RISK TAKING
To understand why the insurance system breaks down in certain cases, it is necessary to understand exactly how it works when it works well.
To provide insurance coverage, insurers must first engage in the art of underwriting. This is not a process of avoiding all loss. If it were possible to avoid all loss, insurance would not be necessary. Rather, when an insurance company places its assets at risk, it does so on the basis of reasonable predictions about how often losses can be expected to occur and what size claim payments they will be required to make.
Life insurance is probably as near to a science as the art of insurance underwriting comes. It is relatively easy to determine actuarially sound rates for particular individuals based on a great deal of experience with people of similar ages and lifestyles, for example. Underwriting in the property/casualty insurance industry is generally less precise.
As with life insurance, property/casualty insurers start with the available data. In determining rates for automobile liability coverage, insurers will first ask: What is this car's loss history? There are, for example, data supporting the contention that red automobiles are involved in more crashes than cars painted any other color. In fact, when it comes to automobile crashes, a database has been built up over time that can provide virtually any kind of statistical correlation desired.
When it comes to many product liability exposures, however, this is simply not the case. For new products, the data do not yet exist. And for complicated products, the available data do not provide a sufficiently strong basis for making credible predictions.
For such products, insurers use manuals produced by the Insurance Services Office, an industry organization for data collection and analysis.
Complicated products are designated with the letter "A," which means that the underwriter's own subjective judgment of the reliability of the product and the company is a key factor in the rate-making process.
This can be a difficult task, made more difficult by the fact that an underwriter must set a premium that encompasses a company's full product line. For instance, 3M has roughly 29,000 products. There is a good deal of art to figuring the appropriate premium for all 29,000 products.
In addition, an underwriter may make a reasonable assumption that a product poses little danger when used for its obvious purpose—but what happens when it is used in some other, more dangerous, context? One such case involves a bicycle bell technology that was being used as an altimeter in blimps to warn of rapidly decreasing altitude. When a blimp crashed, a lawsuit followed, naming the manufacturer of the bicycle bell as one of the defendants. But the bicycle bell manufacturer's insurer, in assessing the risk involved, had charged a premium based on the assumption that the bell would be used for bicycles, not blimps.
THE ROOTS OF UNINSURABILITY
Underwriting product liability generally entails accepting more uncertainty than, say, underwriting life insurance. Theoretically, the insurance purchaser, through appropriately priced premiums, protects the insurer from some, if not all, of this additional uncertainty. Unfortunately, for certain product liabilities, uncertainty rises to the point of wild unpredictability. As a result, no price may be high enough to compensate insurers for the potential losses they face.
In the pharmaceutical, chemical, automotive, and aviation industries, it is impossible to develop very useful risk assessment procedures because the number of variables is too great and the opportunity to isolate and analyze individual factors, and individual victims, does not exist.
But the usual difficulties of underwriting product risks are compounded manyfold by factors external to the nature and harmfulness of the products themselves. Probably the most pernicious source of added uncertainty is the U.S. tort liability system, which, it may be fairly said, has helped kill the messenger that is our insurance system.
LEGAL LIABILITY: SHORT-CIRCUITING THE INSURANCE MESSAGE
Product liability is, historically, a state matter. Thus, manufacturers must deal with 51 separate statutes and 51 separate bodies of case law interpreting those statutes. At the very least, it can be said that U.S. legislatures and courts send a profusion of mixed messages to business. But it is
the liability principles built into these laws that most seriously interfere with the insurance message.
Liability Principles and Uninsurability
The business of commercial insurance involves spreading one party's risk among others. In determining the appropriate price for assuming the risk of a client with toxic or other product liability exposures, the insurer must assess risk on a case-by-case basis. However, joint-and-several liability, a staple of state product liability laws, is a major impediment to individual risk assessment. Under most state laws, individual policyholders may be forced to pay enormous court awards wholly out of proportion to their conduct. An underwriter who has reviewed the conduct of a particular insured and found it to be exemplary cannot simply develop a premium reflecting that fact. This is because the insured may become entangled in litigation involving actors whose conduct may be less commendable, but whose financial status is insufficient to bear their fair share of the liability burden. Thus, the ultimate losses incurred by individual insured's are highly unpredictable.
State laws can be insurer-unfriendly in other important ways:
The use of strict liability magnifies risk, because companies that have done nothing wrong can still be held liable for court awards that are potentially life-threatening to them.
Companies may be held liable for conduct which, at the time, complied with government standards. This is equivalent to changing the rules of the game after the bets have been made.
The absence of caps on noneconomic, especially punitive, damages can result in awards hundreds of times as large as the economic loss caused.
Companies may also be held liable even if victims cannot identify the actual manufacturer responsible. For instance, in some states, a company's liability for DES injuries is dependent on its market share at the time the victim's mother took the drug.
Finally, in understandable sympathy for innocent victims, courts have been known to overreach. In one case, a dice manufacturer was sued when its dice allegedly emitted toxic fumes during a casino fire. While it was clear that the tiny volume of toxic fumes emitted by the dice did not contribute to any injury, that did not prevent the court, in a compassionate mode, from trying to arrange a nice compensation package for the injured party. When insurers are obliged to pay compensation for losses their policyholders had no hand in causing—not even accidentally—it is time for them to reconsider their underwriting practices.
Science, Pseudo-Science, and Uninsurability
For a number of reasons, underwriters must also concern themselves with the advance of science. First, new dangers may be found in products that were previously perceived as safe. Asbestos is a prime example of this. Policies of companies producing asbestos were underwritten in the 1940s. To have avoided the $1.5 billion settlement insurance companies will be paying to settle asbestos claims, the underwriters would have had to anticipate that later a firm causal link between asbestos and certain health problems would be established, and that the plaintiff's bar would be successful in the resulting lawsuits.
Second, the more scientifically advanced the product, the more uncertainty it is likely to engender in insurers. Precisely because it is such a departure from other products, it has no track record and thus provides no solid basis for predicting and pricing the risks involved.
Third, a product that can do a great deal of good for large numbers of people can sometimes do serious harm to a few. Vaccines are a prime example, especially vaccines that use live viruses that, in a relatively small number of cases, cause the disease they have been designed to protect against. If there is a vaccine for AIDS, it will prove simultaneously a boon for mankind and, perversely, a concern for product liability insurers.
Finally, when a product is improved, that should be a cause for celebration. But drug manufacturers have been known to celebrate too soon—unaware of the potential suits they face when they produce a new generation of drug that eliminates the side effects of the previous generation. Drug makers have been held liable for injuries caused by an old drug even though they lacked the scientific know-how to produce a drug free of side effects at the time.
While the above are examples of social goods posing certain difficulties for insurers, the onward march of science has a dark, insurer-unfriendly side, and that is the way our court system treats scientific evidence. In the United States, putatively harmful substances are well publicized and presumed to be lurking everywhere. As a result of this generalized paranoia, underwriters must also factor into their risk assessments the possibility that a court will impose a large bodily injury judgment on an insured based on what is either minority-supported evidence or, in some cases, just plain "junk science." Hence, uncertainty is piled on top of uncertainty.
The Litigation Crisis and Uninsurability
Finally, there is U.S. society's taste for litigation. Underwriters know that for every dollar spent in indemnifying a loss, 70 cents is spent on defense costs, and these costs are increasing 10 percent annually. Not surprisingly,
in the areas where the plaintiff's bar has been most successful, insurance has virtually disappeared. Most pharmaceuticals, including vaccines, are "insured" through risk-funding mechanisms that are tantamount to self-insurance.
The tale of the swine flu vaccine is especially interesting. In 1975 Americans were urged to get vaccinated after an outbreak of swine flu. The insurance industry, anticipating the potential liability exposure for a drug being taken by such a large number of people, canceled the coverage of all four vaccine manufacturers. Congress then passed special legislation so that drug manufacturers would be liable only for manufacturing defects ("bad batch"); the government would take on the liability for any remaining exposure (design testing, for example).
During the time these decisions were being made, there was conflicting opinion on the extent of liability exposure involved. Many consumer groups and many in the plaintiff's bar said that the insurance industry was overreacting, that there would not be a great deal of litigation. Their arguments persuaded the government to assume some of the liability. The four companies paid an $8 million premium for coverage of manufacturing defects, but there was ultimately no litigation against them. However, the federal government did have to pay out more than $100 million to settle suits for the liability it assumed—despite the widespread assumption that it would not lose a penny—and, 20 years later, litigation is still pending.
One lesson this experience teaches is that it is very difficult to predict the extent of the liability that may result from a particular product. Another is that a joint effort by government, insurers, and manufacturers made the vaccination program workable: it could not have gone forward without such cooperation. There is much to be said about the pros and cons of private-public sector risk-sharing, but that is a topic for another paper.
With a legal liability system like this, it is not hard to see how the insurance message gets lost. For insurance to work as a message-bearer, it must be possible for insurance buyers to take measures to improve their loss experience. But the existence of a legal liability system that punishes good behavior as often as bad discourages insurers from staying in the game and leaves corporate risk managers with little to do but lobby their congressional representatives for national tort reform.
Other Causes of Uninsurability
The inflation of loss severity has also played a major role in souring insurers on certain product liability risks. The world has changed since the 1960s, when a judgment of $25,000 was considered a large loss. Several years later, $250,000 was considered a big loss. Today we counsel insurance
companies to settle a class action lawsuit for silicone breast implants for $4.75 billion.
The increased frequency of lost lawsuits is also an issue. Product liability suits rank second behind medical malpractice in the number of nonvehicular cases resulting in million-dollar court awards. Toxic torts are especially onerous to insurers.
The continued growth in product liability awards and in statutory and judicial liabilities poses a major threat to a company's financial health. Insurers, who are in the business of responsible risk-taking, are understandably reluctant to sacrifice their own financial health to save their clients.
TORT REFORM: REVIVING THE MESSENGER
Not only does the out-of-control tort liability system undermine the insurance system, but the absence of insurance in turn undermines a principal aim of the tort liability system—to force wrongdoers to indemnify the innocent victim of their conduct. Companies without insurance are left totally exposed; when a loss occurs, they must pay it all themselves. The result is an increase in insolvent defendants, leading to an increase in uncompensated victims.
A world with dangerous products and toxic exposures is a perilous world indeed. The only thing worse is a world without insurance for the risks inherent in some classes of products or in innovation. Federal tort reform could, if it addresses the sources of insurer uncertainty, go a long way toward making uninsurable risks insurable again.