The Property/Casualty insurance industry loves its data! Loss data, big data, data analytics, etc. Trends in artificial intelligence, automation, autonomous vehicles, wearable technology and smart homes and buildings all hold the promise of ever more data. But is waiting for the data before taking some action always the best way to go? Is it wise for an entire industry to drive by primarily looking in the rear view mirror?
“Without data, you’re just another person with an opinion.”
Frankly, it seems hard to argue with, but that’s precisely what I intend to do. Let’s start with underwriting. At its core, I think underwriting is: understanding the exposures, ensuring you’re getting an adequate rate for those exposures, making sure the risk falls within your company’s risk appetite and also making sure that policy/contract language includes all appropriate defenses given the coverage and exposures. When examining the last three, five or ten years of loss history, the premise is that what happened during those years will be reflective of what’s likely to happen in the next year – and for good reason. However, with the rapid pace of scientific, technological, and medical advances, changing social mores, court decisions abrogating various policy defenses, as well as new and/or amended laws and regulations, the past may be somewhat less indicative of the future. In addition, as faster computer chips are developed and the development of machine learning progresses, the pace of innovation promises to get even faster.
Most liability coverage (e.g. General Liability Commercial Umbrella, Personal Umbrella) is written on an occurrence basis – resulting in open ended liability up to the policy limits reaching out into the future. Keep in mind that some 17 jurisdictions currently allow limits stacking over policy years. This leaves underwriters in the difficult position of trying to assess risks with what’s known today, when the aforementioned advances – impacting so many classes of business – made five, ten or twenty years down the road could present a very different liability scenario than the one originally expected.
Emerging issues are the rub. The P/C industry has never, at least in my 39 years, faced so many emerging issues.
How can insurers monitor emerging issues when, by their very nature, they are not yet showing up in the data? The first place to start is with those scientific and medical advancements. Ongoing peer-reviewed studies, while not infallible, provide the best evidence of the potential for injury/illness and litigation. Information regarding technological advancements and even changing social mores are regularly reported in the popular press. With a little research one can often find information to help estimate the size of the potential plaintiffs’ pool and those potential classes of business that may be in the liability chain and named as defendants in any ensuing litigation. Previous litigation and claims activity may provide a comparison, at least to some extent, to some current emerging issues. For example, several studies have linked the inhalation of carbon nanotubes with asbestosis or mesothelioma like symptoms. Perhaps the size of asbestos verdicts could be instructive here?
Knowledge of coverage and policy/contract language is a must. After all, just because something may be significant to society, doesn’t mean it will translate into insurance claims, particularly if policy defenses have already been upheld by state appellate and/or state high courts. Therefore, one must also monitor coverage actions as even state high courts can reverse previous decisions which had upheld policy defenses. Coverage actions and legal activity are often reported in law and insurance journals, but having an online source (e.g. Lexis/Nexis, Westlaw) can be a big help.
Appellate court decisions ruling on the efficacy of the absolute pollution exclusion provide a great example of the importance of monitoring coverage actions. Appellate level courts in some 16 jurisdictions have not upheld the intent of the language under various fact patterns. That said, how many insurers provide GL and/or CU coverage in those jurisdictions without giving pollution exposures a second thought because, after all, the policy contains an absolute pollution exclusion?
When considering potential latent exposures, generally, the more peer-reviewed studies concluding that exposure causes harm, the more likely the litigation – although, admittedly, determining the tipping point can be quite daunting. When should an insurer take action (which could be anything from creating a new product, to amending underwriting guidelines and risk appetite, to adding a new exclusion)? This should vary by size of the insurer, risk appetite, estimated exposure to any given emerging issue, etc.
It’s not likely that any insurer would take action after one study (unless that action is to begin monitoring the issue). But, what about after a dozen studies? Fifty? How about a couple of hundred? The question is not as easy to answer as it may first seem.
The industry’s action (or inaction) regarding exposure to nanomaterials may help explain. There have been several hundred studies that have linked exposure (inhalation, ingestion and/or dermal) to various nanomaterials with serious adverse health effects. Exposure to nanomaterials is a potentially onerous emerging issue. Like with asbestos, nanomaterial manufacturers and manufacturers incorporating these materials into other products could result in litigation based upon occupational exposures. In addition, however, there is already far more consumer exposure to nanomaterials than there ever was for asbestos. Yet many insurers have taken no action. It could be that they believe they have little or no exposure. After all, many mutual companies will not provide coverage for nanomaterial, or even products, manufacturers. However, many nanomaterials are manufactured overseas and many of the products containing those nanomaterials are also made overseas and sold in many small U.S. retail stores. Should there be allegations of illness arising from exposure, the U.S. distributors and retailers would likely be in the liability chain. These defendants could certainly be covered by mutual insurers. Many small mom and pop shops are often written for minimum premium, some even on BOP policies, but perhaps with more exposure than may be realized. Sure, distributors may be the first defendant named if the manufacturers are overseas, but as we’ve seen in previous emerging issues litigation, distributors can quickly declare bankruptcy leaving only U.S. retailers left in the liability chain.
Insurers now have over 20 years of exposure to nanomaterials. Maybe carriers are waiting for that first multimillion dollar verdict to show up in the rear-view mirror.
Another example of an emerging issue with an increasing number of studies linking exposure to adverse health effects that could impact both large and small commercial insurers is E-cigarettes/vaping fluids. They entered the U.S. marketplace in about 2007 and quickly became available at a variety of retailers, even car washes. Many insurers don’t even have an application question asking whether the insured sells e-cigarettes or vaping fluid – let alone underwrite or charge for the exposure. Insurers already have some 12 years of potential limits stacking for, in many cases, risks that were charged minimum premium. Of course, not all emerging issues will cause illness or involve limits stacking. Take the advent of autonomous vehicles. It is predicted they could reduce the frequency of accidents by 80% – 90% and dramatically reduce fatalities. While great for society, autonomous vehicles present a huge challenge for the P/C industry – which gets roughly half of its premium from personal and commercial auto coverage. Many insurers have not taken steps to address the coming changes brought about by autonomous vehicles, believing they have many years before they need to act. But figuring out how to replace half the industry’s premium will not be easy. Development of, and growth in, other insurance products could take many years to replace lost auto premium. Meanwhile, the technology continues to advance. Trucking companies are already running revenue-generating autonomous trucks to haul cargo on highways in several states. Grocers are already using autonomous vehicles to deliver groceries. McKinsey & Co. recently predicted that “in 7 to 10 years autonomous trucks will operate throughout the interstate-highway system and other ‘geofenced’ areas.” Grocery chain, Stop and Shop, will be experimenting with driverless delivery vehicles in Boston this spring, joining Walmart and Door Dash in moving towards driverless grocery/food delivery. Waymo is already operating driverless taxis in the Scottsdale, Arizona area, and driverless shuttles are operating in several U.S. cities. In addition, two states have already passed laws deeming accidents caused by autonomous driving systems to be products liability, not auto liability. Will other states follow suit? Yes, keeping up with emerging issues is time consuming. After over 30 years of monitoring emerging issues, believe me I understand. There’s also the current push for “ease of use” resulting in fewer application questions, but considering that just a few prior emerging issues are slated to cost insurers over $200 billion in unexpected losses (e.g. asbestos, pollution, EIFS [exterior insulation and finish systems], construction defects, mold, lead), the cost of not monitoring some 200 new emerging issues and not adding application questions could be very steep. Still, most insurers still don’t have a single full time Emerging Issues Officer or Analyst position.
So, back to the question I raised in the first paragraph, how comfortable should you be driving while primarily looking at the review mirror?
Retired Vice President and Emerging Issues Officer at Gen Re
