Introduction

The COVID-19 pandemic made business interruption (BI) insurance and business closure (BC) insurance widely known to the general public, mainly in the context of government-introduced lockdown orders, which were designed to prevent the spread of the disease but caused havoc to the daily operations of businesses of all kinds and size, although mostly to small and medium-sized enterprises (SMEs). These included restaurants, bistros, pubs, hotels, childcare institutions, retail shops, hairdressers and many others. Not all of the businesses affected carried insurance against the risk of this form of government interference prompted by the pandemic, but many had. Did those policies respond? That was the question that was widely asked.


Whilst BI insurance called up for COVID-19-induced losses was mainly issued in the international and U.K. markets, BC insurance, by contrast, was mostly sold in Germany (as ‘Betriebschließungsversicherung’).

Judicial response in the U.K. and in Germany

Not surprisingly, BI and BC insurance came under widespread judicial scrutiny. Thousands of claims were made against insurers, many consumer protection agencies and, indeed, lawyers were being consulted and many claims ended up being filed in courts of law in both countries. This huge tide of judicial review activity is only now about to ebb away, after the U.K. Supreme Court (U.K. SC) pronounced a landmark judgement in the well-known FCA test case (Financial Conduct Authority v Arch Insurance [UK] Ltd. V Hiscox Action Group [2021] UKSC 1) in January 2021, which concerned a number of crucial insurance market clauses in English BI policies. In Germany, the German Federal Supreme Court (Bundesgerichtshof, or BGH) handed down a first final and authoritative ruling (Judgement dd 26 January 2022, IV ZR 144/21) on 26 January 2022, which, although not formally labelled a test case, will have a very similar legal effect. It will guide many other sets of legal proceedings commenced in the German courts. The BGH judgement will now considerably lower water levels in the judicial and scholarly legal debate about whether German BC insurance cover responds to the COVID-19-induced closures of thousands of businesses in the country.

The astonishing outcome of these two test cases could not have been more different: in the U.K., the plaintiff policyholders won the legal battle; in Germany they lost it. In both countries, the surge of legal proceedings may well be perceived as a direct reflection of the considerable political and social pressures placed upon law and legal systems by the COVID-19 pandemic (del Carmen Boado-Penas et al. 2022).

Perhaps less surprisingly, the intensive judicial reviews exerted in both countries heralded groundbreaking developments in their insurance law. This may be illustrated, with respect to the FCA test case in the U.K., by reference to the fact that an older case of the English High Court (Orient-Express Hotels Ltd. V Assicurazioni General SpA (UK) [2010] EWHC 1186 [Comm]) was expressly overruled (even though, ironically, one of the judges who sat in the U.K. SC on the FCA test case previously sat in the High Court when Orient-Express was decided there, some 12 years earlier). Some of the basic principles of the law on causation were looked at and readjusted by the U.K. SC, in part. We will return to that aspect of the U.K. SC judgement in a moment. Suffice it here to say that the U.K. SC, in dealing with the pandemic, coined a few new principles of insurance law.

Similarly, in the BGH decision, the highest German court took the somewhat unusual step of letting the legal community know—albeit obiter—that four distinguished German Courts of Appeal and many courts of first instance from all over the country had previously got it wrong and had in fact misunderstood the pertinent insurance law concept underlying German BC insurance contracts. In other words, the BGH wanted to make absolutely clear to everybody that BC insurance policies responded not only to ‘localised’ infection scenarios, which happen to occur at insured businesses’ premises, but also to anticipatory COVID-19 lockdown orders that were to take effect across the country (BGH, Judgement dd 26 January 2022, IV ZR 144/21, notes 10/11).

These were remarkable developments. Whilst unusual things do happen from time to time within any kind of hierarchical system, they in fact rarely happen in judicial systems. In general, the law tends to be conservative and stable. It is not easily shaken, and principles of law and precedents are not often overruled, not even by dramatic events such as a global pandemic. Yet, as these unusual things did in fact happen in response to COVID-19-induced insurance law litigation, both in Germany and the U.K., the phenomenon might warrant closer scrutiny.

The research question might be formulated thus: are these seemingly opposite outcomes, suggestive of error, or are they, perhaps more likely, merely suggestive of differences in legal approach? The answer seems obvious. The extensive high-quality review undertaken in the courts of either country could be relied upon, certainly in this instance, to rule out the risk of error of judgement.

BI and BC insurance—distant relations or no relation at all?

A few conceptual basics should provide the starting point for our brief expedition into the realms of Anglo-German comparative insurance law, as follows:

  1. 1.

    Traditional BI insurance is there to cover loss of earnings or extra expense flowing from the standstill of a business consequent upon the occurrence of property damage within the business or within its vicinity. Historically, BI developed as an add-on to fire insurance. Extended coverage policies gradually became available for other perils such as theft, burglary, storm and hail, by way of supplements to property insurance. Even all-risk policies may today be combined with BI cover. However, in all of these there would always be a—factual as well as legal—requirement to the effect that property damage must occur first, before BI cover would be triggered. This is known as the ‘material damage proviso’ (Riley 2021, at 2.37–2.38, on pages 53–56). Of course, in the case of the COVID-19 pandemic, hardly any of the businesses tied down by government lockdown orders could demonstrate to their BI insurers that property damage had in fact occurred at their own premises or in the vicinity. The English High Court decision in TKC London Ltd v Allianz Insurance plc ([2020] EWHC 2710 (Comm)) provides an illustration: the owner of a café that was required to close by U.K. government regulation in response to the pandemic tried to subsequently make a claim under a combined property and BI policy but could not show that his business had in fact suffered any property damage. The inability to use the café for its designed purpose did not amount to ‘damage’ sufficient to support a claim under the BI section of the policy. Consequently, the claim was dismissed in the High Court.

  2. 2.

    By contrast, typical BC insurance, as is sold in large numbers on the German insurance market, did not require the presence of property damage to trigger coverage. Instead, BC insurance would be directly triggered by the presence of infectious disease (or threat thereof), which would in turn cause public authorities to order the closure of the business concerned. BC insurance policies were mostly sold in the food processing and health sectors. It should be clear, therefore that BC insurance does not fall within the category of property insurance at all. It stands independently of any requirement of property damage. Instead, BC insurance is typically premised on the occurrence of infectious disease, with the relevant diseases being listed and named in the policy wording. All diseases appearing on that list need to be reported to the health authorities whenever and wherever they materialise. They would then prompt the relevant public authority in charge to order closure of the business concerned. The trouble with the COVID-19 lockdown orders for policyholders was that the list of infectious diseases in the policy wording did not, of course, mention the SARS CoV-2 virus nor COVID-19 disease, as they were unknown when the BC insurance cover was written and the policy issued.

  3. 3.

    Another typical feature of BI insurance, when compared to BC insurance, relates to the mechanism by which the extent of the loss normally comes to be measured. Whereas in the German market for BC insurance one would normally encounter a ‘taxed’ policy, i.e. a policy which identifies the measure of indemnity by reference to a pre-agreed, fixed sum of money (meant to reflect the presumed daily earnings of the business), multiplied by a certain number of days for which the business is actually interrupted (with a typical maximum of 30 days, though longer periods may of course be agreed upon individually). Once cover is engaged, little more than a simple mathematical calculation is needed to arrive at the amount payable under the policy by way of indemnity for any particular loss scenario. By contrast, in a run-of-the-mill commercial English BI policy, both parties to the bargain would need to embark in a second step upon a factual enquiry in any given loss scenario triggering coverage, whereby the causative effect between the lockdown order, on the one hand, and the actual drop in business income, on the other, would need to be established. Other, concurring causes would need to be taken into account, too, which might include, in the case of restaurants and hotels for example, any natural inclination of people to stay at home during times of crisis and increased risk of infection (or because the government has urged its people to stay at home). Such ‘counterfactual’ elements and their causative effects would need to be recognised, potentially diminishing any individual claim. This would normally happen during the loss adjustment process. With the exceptional situation brought about by the COVID-19 pandemic in mind, it is easy to see that this kind of case-by-case factual enquiry, which would need to be conducted in effect simultaneously for a huge number of individual claims, might bring the entire insurance and loss adjustment system to the brink of collapse. Long delays in the settlement process are inevitable.

All in all, whereas BI and BC insurance cover tends to cater for very similar, if not identical, needs of the pertinent business community concerned, particularly within the SME sector, the respective basic structure of the insurance product, in terms of cover as well as indemnity, turn out to be mostly—and fundamentally—different.

What, if any, enlightenment can be derived from the highest judicial authorities as to what might properly be called ‘justice’ with respect to insurance coverage against COVID-19 BI and BC losses?

Of course, no common thread immediately springs to mind for scholarly analysis in the two aforementioned judgements of the U.K. SC and the BGH. Clearly each judgement was premised on its particular set of facts, as this came to be placed before the court. And, of course, the rulings were based on the specific insurance terms and wordings used in the underlying insurance contracts. However, there might still be some merit in trying to look behind the practical results achieved at the end of those proceedings, from a broader perspective. In particular, one might ask whether ‘justice’ (or lack thereof) was actually dispensed to those assured policyholders seeking judicial relief in the face of widespread financial loss and dramatic business disruption. Should ‘justice’ not demand that similar results are being achieved by the highest legal authorities anywhere, with regard to seemingly similar insurance products? Policyholders naturally looked to their insurers for help when being faced with stringent lockdown orders, i.e. for help to weather the financial consequences of the COVID-19 pandemic. Yet they were mostly turned down. They then turned to the courts for legal redress.

With that line of enquiry in mind for both judgments, the U.K. SC’s and BGH’s answers to the COVID-19 pandemic at first strike one as being correct on legal principle. They also follow a sound interpretation of, and give effect to, underlying policy wordings. Both courts correctly identified both parties’ intentions as those had been expressed in the policy wordings. Yet, in the German rally about BC insurance litigation, policyholders suffered a resounding defeat in front of the BGH, whereas their counterparts in the U.K. SC came away with a triumphant victory. How could this be? In light of the sketchy analysis above, one might be tempted to ask whether these diverging results had anything to do with the structural differences between BI and BC insurance, i.e. with the lack of common ancestry between them.

A few interesting features that distinguish BI insurance from BC insurance emerge upon closer analysis:

  1. 1.

    One of the salient features of the coronavirus is its exponentially high degree of contagion—its ability to spread fast and wide, by almost any form of contact, however, light, amongst humans. Under normal circumstances the virus will spread extremely quickly and easily.

    1. (a)

      It was this specific propensity of the SARS CoV-2 virus that caused some governments to impose, to varying degrees, lockdown orders and similar restrictions. Some governments directly opted for very stringent measures, whereas others first acted in a more restrained form. In almost all instances, however, whatever government measures were eventually imposed, they were usually made to apply throughout the entire territorial expanse of the respective geographical territory over which the respective government body exercised jurisdiction and control. It is, therefore, not surprising that it was considered irrelevant whether individual business premises within those territories could be specifically identified as being ‘infected’ by COVID-19 or not.

    2. (b)

      Even where an incidence of specific COVID-19 infection could be demonstrated by reference to any particular local area, the U.K. SC ultimately considered it largely irrelevant, as far as the policy’s response to the lockdown orders was concerned. In BI policies issued in the U.K., such an area was often identified by reference to a geographical radius of one mile or 25 miles around the insured premises. In other words, the U.K. SC came to assist the plaintiff policyholders by significantly relaxing the rules of evidence. As a result and in short, geographical proximity to actual incidents of COVID-19 mattered little, if at all, in deciding whether a particular BI insurance policy was to respond to any loss of income flowing from the COVID-19 pandemic.

  2. 2.

    It is interesting to see that this aspect of local proximity to any actual COVID-19 infestation featured prominently in both judgements under review, albeit in slightly different juridical disguises. In the U.K. SC, it surfaced in connection with the aforementioned one or 25 miles radius, whereas in the Federal German Supreme Court the debate centred around whether an incident of actual COVID-19 infestation at the insured premises had to be shown in order for BC cover to be triggered. In both cases, the courts got around these geographical restrictions and held, in effect, that proximity between the insured premises and incidences of COVID-19 was (almost) irrelevant.

    1. (a)

      The U.K. SC disguised this conclusion in an ingenious judicial finding on causation and whilst applying rules of evidence. It held that, in order for the policyholder to show that the loss that had struck the insured business was proximately caused by one or more occurrences of illness resulting from COVID-19, it was sufficient to prove that the interruption was a result of government action taken in response to all cases of disease, which included at least one case of COVID-19 within the geographical area covered by the insurance wording. This finding was based on the legal reasoning that each individual case of illness resulting from COVID-19 within the U.K. was a separate and equally effective cause of the action of the U.K. government (FCA v Arch, UK SC [2021 UKSC 1, para 212).

    2. (b)

      The German Supreme Court, by contrast, could not, in dealing with BC insurance, resort to the concept of causation, which is so fundamental to property insurance. But it also dealt with the aspect of geographical proximity. It did so by unequivocally dismissing the insurers’ argument that in order for BC insurance cover to engage, an actual incidence of COVID-19 disease had to be shown to have occurred at the insured premises. In the course of that argument, which had been advanced (and had actually succeeded) in many of the German first instance as well as appellate courts, the point was summarised by postulating that BC insurance only engaged with respect to ‘intrinsic’ risks, i.e. risks of infection specifically affecting the insured business. It was argued that cover did not engage where the relevant generic government orders imposed business closures and were being motivated by trying to prevent (future) infection rather than by fighting an actual incidence of disease that had already manifested at the insured premises. In dismissing that argument, the BGH admonished one should look carefully at the policy wording instead. Since no distinction was drawn there between ‘intrinsic’ perils (i.e. local perils relating specifically to the insured premises) and other perils, the conclusion of subsequently writing any such distinction into the insurance policy was not open to the court; the whole argument was misconceived (BGH, Judgement dd 26 January 2022—IV ZR 144/21, notes 10–11).

  3. 3.

    Next, it would be worthwhile to look at the concept of causation again, in further pursuing the comparative relationship enquiry between BI and BC insurance. After all, insurance is invariably premised on the fundamental principle of causation (one of the few exceptions here is so-called parametric insurance [Riley 2021, para 14.6. at p. 593–595] which, however, is not relevant for the COVID-19 pandemic).

    1. (a)

      In the case before the German Federal Supreme Court, the court did not have to examine the question of causation. It simply admonished that the list of infectious diseases enumerated in the BC policy would need to be examined for SARS CoV-2 or COVID-19. Since this search, not surprisingly, turned out zero results, there could be no cover (BGH, Judgement dd 26 January 2022, IV ZR ZR 144/21, para 13–20). The BGH also observed that the purpose of BC insurance lies in protecting the assured against the risk of suffering economic loss (BGH, Judgement dd 26 January 2022, IV ZR ZR 144/21, para 42). This latter finding shows that BC insurance, in the opinion of the highest German civil court, must not be counted amongst property insurance (in German: Schadenversicherung). Property insurance ‘protects’ the insured object by offering the assured an indemnity aimed at restoring or replacing the insured property, whether lost or damaged. BC insurance, by contrast, does not hinge on loss of or damage to property, it is triggered solely by government orders disrupting the business protected by insurance.

    2. (b)

      Moreover, the point about causation between operation of the peril insured against and any financial loss suffered by materialisation of that peril does not really arise in BC policies in practice. Even if COVID-19-induced business closure had in fact been triggered in a scenario where the government order had been provoked by one of the diseases enumerated (and, therefore, covered) in the BC policy, that additional enquiry would still have been largely moot. As mentioned above, this is because in BC insurance contracts the enquiry as to whether the assured actually does or does not suffer any financial loss as a result of being closed down and ‘locked’ in by a specific government order has typically already been made obsolete at the time of entering into the contract. Usually. the parties already agree on the level of indemnity any future closure event would be deemed (and agreed) to produce, by reference to the number of days the order would be in effect and apply to the insured business. This means that other and (actually or potentially) concurrent reasons for causing turnover or earnings of the insured business to suddenly drop off, are normally irrelevant to the measure of indemnity payable under typical German market placed BC cover.

    3. (c)

      By contrast, in the FCA test case, the U.K. SC surprised everybody in the insurance community with some novel ideas regarding the law on causation and new principles that the court developed and directly proceeded to apply to the concept of causation in relation to the effects of the COVID-19 pandemic on SMEs. Specifically. in relation to COVID-19-induced BI insurance, the court emphasised that insurers could not rely on clauses relating to trends or other circumstances in these policies, no matter their craftily designed individual wordings, to reverse the finding that cover was engaged (FCA v Arch, [2021] UKSC 1, para 237, 240). In other words, the enquiry directed at the extent of the indemnity could not be used to undo the decision previously arrived at on cover, namely that cover was in fact engaged and that BI insurance, on the basis of that finding, was meant to respond to the policyholder’s benefit.

    4. (d)

      Put simply, the SC held that an insurer could not, on the one hand, promise cover for government-ordered closure of business operations, brought about or motivated by the perceived threat of contagion with COVID-19, only to turn around and deny indemnity on the basis that that very same threat would have kept all potential customers of that business at home anyway, for fear of catching COVID-19. The normal ‘but for’ test in the law on causation was held not to apply. This means that the fact that a dramatic drop-off of business might have been inevitable in any event, independently of the government order, will not be taken into account. In the pursuit of justice in the same direction, the U.K. SC was forced to go so far as to expressly overrule an earlier insurance-related judgment on causation (Orient Express Hotels Ltd. v Assicurazione Generali SA (UK) [2010] Lloyd’s Rep. IR 531), which had gone the opposite way. This juridical turnaround vividly illustrates that the U.K. SC, seeing that insurers had expressly extended traditional property covers to respond (also) to scenarios that did not involve physical loss of or damage to property, now insisted that such extension would not—at the second stage of the enquiry—be brought back in again, thereby defeating cover by operation of the principle of causation.

  4. 4.

    Some of the salient insights which these rulings of the courts of highest authority in the U.K. and Germany might offer are, therefore, quite remarkable:

    In deciding on coverage as well as indemnity, either in BI or BC insurance policies, one of the fundamentals of insurance law, namely the enquiry into the causal link between peril and loss, may have been absent ab initio, because it had been dealt with conclusively by the parties at the time of entering into the insurance contract. This is the case with typical BC insurance as was sold in large numbers in the German market. Alternatively, the same enquiry has been considerably reduced in scope and depth, as is the case with BI insurance as interpreted by the U.K. SC in the FCA test case.

    With this finding in mind, one is tempted to ask why the supreme courts in the U.K. and Germany, knowing full well that their judgments would have far-reaching and widely persuasive effects for thousands of insurance claims in either country, produce diametrically opposed results? Why, in spite of the widely similar, if not identical, business objectives for buying BI and BC insurance, did the two judgements leave plaintiffs triumphant in the U.K. and crushed by defeat in Germany? In both cases, the government’s objectives were the same: by imposing lockdown orders, they tried to respond to the health crisis brought about the COVID-19 pandemic.

Conclusion

The above analysis allows only for one answer, and two brief observations on potential consequences for reinsurance cover.

  1. 1.

    The two forms of insurance protection, though they appear strikingly similar and are aimed at largely identical risk scenarios, are in fact quite different. Traditional BI cover is (and remains, even when extended to non-property-damage scenarios such as prevention of access clauses) essentially property insurance. This type of insurance should be looked at in light of the FCA test case findings on extent of cover and application of (partly new) principles of causation and rules of evidence. By contrast, the BC insurance offered on the German market is not at all based on property risks; it is based on contingency risk. BI and BC insurance are indeed (very) distant relations, if they can properly be considered related at all.

  2. 2.

    Having said that, the structural differences between traditional BI insurance and German BC insurance must not be overstressed, lest it lead to confusion. Insurance markets have shown many recent developments from classic BI to various forms of modern contingent BI, including many situations in which no physical damage is required. This clearly presents a challenge to those scholars advocating traditional, yet mostly academic, views on how BI and BC insurance must be distinguished on a conceptual level. From a commercial perspective, both types of insurance profess to offer a similar type of protection against the risk of government interference prompted by infectious disease. Whilst conceptually they seem to be coming from different directions, in practical terms, they aim at providing similar types of insurance protection.

  3. 3.

    This conclusion might prove to have far-reaching consequences once the insurance law issues engendered by the COVID-19 pandemic start their ascent upwards from primary insurance into various levels of reinsurance and retrocession. In particular, the question looms large whether the global COVID-19 pandemic might come to be regarded as one or more ‘events,’ in the context of Cat Loss XL covers (see Rhein 2022, pp. 385–394; Schwepcke 2022, pp. 467–485; Stahl 2022, pp. 501–516). Whilst that question currently still appears to be free of authoritative court rulings anywhere and certainly from the highest courts in the U.K. and Germany, some academic opinions are beginning to be formulated already (Rhein 2022; Schwepcke 2022; Stahl 2022; Woloniecki et al. 2021).

  4. 4.

    However, another, quite different question might have an even greater impact on the availability of reinsurance protection against COVID-19-induced losses incurred globally: contingency risks in Cat XL covers are often sublimited or form the subject matter of carefully worded exclusion clauses in Cat XL reinsurance and retrocession contracts. On the latter question, the above analysis might conceivably have some bearing and might warrant revisiting at some point in the future.