"The single biggest problem with communication is the illusion that it has taken place"
(George Bernard Shaw)
Tens of thousands of small businesses will receive insurance payments covering losses from the first national lockdown, following a ruling by the Supreme Court of England in favor of small businesses receiving payments under business interruption insurance policies. The case is Financial Conduct Authority & Others v. Arch Insurance (UK) Ltd. & Others [2021] UKSC 1 (January 15, 2021). The verdict marked a historic victory, affecting approximately 60 insurers that sold similar products, and all valid claims will be settled as quickly as possible. For some companies, this could be a lifeline, allowing them to survive the coronavirus crisis. The decision could cost the insurance industry hundreds of millions of pounds.
What is the ruling about? The case was essentially a “consultative opinion” provided by the Supreme Court at the request of a government regulator (the Financial Conduct Authority) and a number of insurers regarding whether “business interruption” insurance policies apply to costs incurred by the suspension or closure of business operations due to the lockdown requirements imposed by the UK government. During last spring’s lockdown, a huge number of small businesses filed claims under business interruption insurance policies for loss of income when they were forced to close. Despite proof of payment of the annual premium, many insurers refused to pay, arguing that only a few more detailed policies provided coverage for unprecedented events, such as COVID-19. Thus, it was agreed that a selection of terms included in the policies should be tested in court to establish the parameters for what would be considered a valid claim.
The complex ruling addressed issues such as “disease clauses,” whether businesses were denied access to their properties, and the point in time at which the loss of economic earnings occurred. The Supreme Court’s decision provides reliable guidance for the policies under discussion and, potentially, for other similar policies not involved in the litigation, covering a broader set of 700 policies and potentially affecting 370,000 small businesses, covering their main costs—particularly rent—and the losses suffered by small businesses forced to close while awaiting the decision.
Beyond the economic outcome of the decision, we cannot ignore the significant discussion regarding the causal link. The majority of the judges held that, for insurance purposes, the accepted approach is to ask whether there is an “immediate cause” between the event and the policy. The answer for an insurance contract will not be the same as that obtained in matters of civil liability. The concept of “a cause” is distinct from another issue, namely, which cause-and-effect links are relevant in the specific legal context. In the tort of negligence (the most common common-law tort), this issue is referred to as “remoteness of damage” (UK) or “scope of responsibility/liability” (USA). In the interpretation of insurance policies, however, this separate issue has been labeled “proximate cause.” Like us, the British also grapple with the issue of causal connection, and there is an endless debate over the precise definitions of the terms “cause,” “contribution,” and “responsibility.”
The policy in question states the following:
“We shall indemnify you for business interruption or interference during the following indemnity period:
III. any occurrence of a Notifiable Disease within a 25-mile radius of the Facilities;
“Indemnity Period” means the period during which business results will be affected as a result of the occurrence.”
The crucial interpretive question before the UKSC was what “as a result” means in this context. There must be some kind of connection between the disease and the business interruption. Interpretation is necessary to reach a conclusion; it is not enough merely to identify the common meaning of “cause” or “consequence,” or anything else.
Just as in Brazilian law, causation is a basic prerequisite for civil liability under common law, whether the defendant personally caused the damage—requiring a causal connection between the defendant’s conduct and the damage—or where there are “joint tortfeasors,” or in cases of “vicarious liability,” in which certain individuals are held indirectly liable for the conduct of others who are under their legal or contractual authority. In any case, in England, factual causation requires the “but-for test,” a hypothetical inquiry into what would have happened had the defendant not engaged in a given conduct. According to the “but-for test,” conduct is the cause of harm when, but for that conduct, the harm would not have occurred. If the answer to the test is negative, then the conduct caused the harm.
However, SCUK demonstrated that the “but-for test” is not always decisive. If the test were followed to the letter, the business interruption clause regarding “disease” would be interpreted to mean that the insurer would only compensate for the effects of COVID-19 cases occurring within the specified radius of the insured premises—approximately 25 miles. However, the policyholders’ claim concerned the government’s business closure rules, which obviously affected all of England. That is to say, when considering any single case of COVID-19, can it be said that that case was the cause of the lockdown rules? In other words, no individual instance of the disease can be considered a “but-for” cause of the government’s lockdown rules. If, as the insurers argue, the relevant test when considering the government’s measures taken in March 2020 is to ask whether the government would have acted in the same way on the assumption that there were no cases of COVID-19 within a 25-mile radius of the insured’s premises, but all other cases elsewhere in the country had occurred as they actually did, the answer—with respect to any particular policy—must be that it would likely have acted in the same manner.
Therefore, the central issue was whether the specific cases covered by the clause “caused” the damage. Bearing in mind that this was a matter of insurance and not of civil liability, most judges approached the case through the lens of the theory of “over-determined causation”—the classic example of which is that of a hunter struck by two bullets at the same time, each of which would have killed him. A hypothetical case discussed during oral arguments involves 20 individuals who band together to push a bus off a cliff. Suppose it is shown that only 14 people would have been necessary to produce this result. One could not then say that the participation of a particular individual was necessary or sufficient to cause the destruction of the bus. However, it seems appropriate to describe each person’s involvement as the cause of the loss. If we consider the “but for” test as a minimum threshold that must always be exceeded if X is to be considered a cause of Y, we would reach the absurd conclusion that none of the 14 actions caused the bus’s destruction.
This line of reasoning is useful because it has been argued—and accepted—that the whole of Great Britain could be covered by about 20 circles with a radius of 25 miles. Suppose that Company X is located in Circle Y and that the government would not have imposed the lockdown if only one circle had COVID-19, but would have done so if, say, 13 circles had the virus. So, under this approach, to show that COVID in circle Y was the cause of the lockdown, all one needs to do is show that circle Y is a necessary element of a sufficient set (12 other circles and circle Y) that led to the lockdown.
No single case of COVID-19 was necessary for the government action that led to the suspension of business operations, but each case made a real contribution to the prevalence of the disease that triggered that action, even if that prevalence was excessive (i.e., there were more cases than were necessary to trigger that action, so that no single case was necessary on its own).
The Supreme Court was persuaded by the argument that, for a proper construction of insurance policies expressly triggered by notifiable diseases (a small set of the most dangerous or infectious human diseases, including SARS), the indemnity clause should have contemplated diseases extending beyond the radius in question and that the main route of business interruption would be government intervention in response to the disease. That being the case, the proximate cause that the parties intended to be satisfied was met by the link between local occurrences and the national response to cases across the country, even though the “but-for” test was not satisfied. In other words, the parties would not have intended for the test to be applied to the local occurrences themselves, but only to the larger context of which they are a part. Therefore, the policy must provide that the insured risk (disease within the radius or, in other clauses, the public authority’s response to the disease) does not compete with other consequences of the underlying fortuitous event (COVID-19 itself), which means they must be treated as a single entity. The 25-mile clauses provide coverage because they are designed to cover diseases that will occur both within and outside the area and will affect businesses through actions taken by public authorities in response to the widespread disease.
Another aspect that impacts the interpretation of these clauses involves insurers’ practice of reviewing contracts in a “standardized manner,” without taking into account the specific factual circumstances between the parties. By way of illustration, let’s consider a restaurant seeking insured coverage in light of the pandemic. What about other factors contributing to the restaurant’s closure or financial loss? Consider a restaurant that would have closed anyway (or seen its revenue decline) because its “star chef” was about to leave—and did so. Why would the event of the star chef’s departure (not covered by insurance) be any different from the event of all COVID-19 cases outside a 25-mile radius (also not covered by insurance)? Or, consider a restaurant that would have had no revenue anyway, even without the restrictions, because it’s a terrible restaurant. In this case, clearly, there would be no loss. Since this is a claim for compensation for business interruption, the counterfactual is how the restaurant would have fared had there been no interruption. Consequently, if the establishment were of poor quality or the chef had left anyway, the compensation would be lower than if it had generated high revenue and profit.
The question is whether the compensation will reflect the situation of the restaurant as it would have been operating in the real world—without the restriction on access but with the rest of the effects of COVID-19—or without the restriction on access and without the effects of COVID-19? The Supreme Court interpreted the policies based on the second alternative. Therefore, it is a subtle approach to how the parties must have intended causation to apply. The counterfactual of the original cause (the disease) and everything that stems from it is removed—not just the “insured risk” in its strict sense.
As for the concept of cause, the Supreme Court held that the insured risk (for example, a COVID-19 case within the relevant geographic radius) was a cause of the government action. In other words, it was recognized that the concept of being “a cause” in private law is broader than being a necessary factor (as in the “but-for test”), regardless of the context. A factor is a cause of an injury if, without it, the injury would not have occurred or a real contribution to an element of the positive requirements for the injury to occur would not have existed.
The SCUK held that the insured risk was a cause of the government action and that it was not the parties’ intention that a proximate causal link between this cause and the government action also be required. The parties must have contemplated illnesses outside the relevant area; and this indicated that the proximate cause test the parties intended to require was satisfied by the link between local occurrences and the national response, even if the “but-for test” was not satisfied because it appears contrary to the negotiating intent of the clause, to address uninsured cases of notifiable diseases occurring outside the territorial scope of coverage, thereby depriving the insured of compensation for business interruption also caused by cases of disease that the policy is intended to cover.
It is clear that, from both a theoretical and practical perspective, causation is a thorny issue in any context.
__________