Business Interruption Insurance (COVID-19) – Quantum Wars Part 3 - The Insurers’ Case(s)

01 July 2020

This article is an update to my previous article (which can be found here) which set out the quantum position put forward by the FCA on behalf of policyholders in its Particulars of Claim.

On Tuesday 23 June, seven separate Defences were filed from the eight insurers named in the Test Case (Arch, Argenta, Ecclesiastical, Hiscox, MS Amlin, QBE, RSA and Zurich).

Whereas the Particulars of Claim was an all-encompassing document setting out the policyholders’ claim against all parties, the insurers’ Defences respond directly to the issues relevant to their own policy wordings which are included in the Test Case.

I don’t think it would be unfair to say that the insurers’ Defences currently provide more nuanced arguments than the Particulars of Claim as they are responding to specific wordings – however, the FCA will have an opportunity to provide a Reply to these Defences (due on 3 July) and I would expect those to be similarly nuanced.

From reading the insurers’ Defences, there is a blurring of the coverage (liability) and Causation (quantum) arguments i.e. insurers are - in part - denying coverage because they believe the main cause of the loss is not covered.

The variation and nuance in both the policy wordings, and the arguments put forward by insurers makes it challenging to provide an overarching summary, but I have attempted to capture the key themes and arguments relating to quantum below.

Proximate Cause - Government measures

The FCA position is that:

The Government response, in the form of advice, instructions and legislation, was a single body of public authority intervention which did prevent and hinder access to and use of business premises, cause closure of and restrictions on activities at business premises, and interrupt and interfere with activities at business premises, within the meaning of the Wordings.”

Insurers have adopted a different position whereby they consider that the Government advice can be segregated, and that only the Government mandated closures would trigger coverage. The Arch Defence sets this out in detail:

For these purposes, the appropriate counterfactual scenario is where there was no insured peril, i.e. no government or local authority action or advice preventing access to the Premises, but where all other factors remain unchanged. Those factors include but are not limited to the following:

  1. COVID-19 existed and was prevalent in all or most parts of the United Kingdom;
  2. The various other official control measures remained in force, including the UK Government regulations and advice on social distancing, the “lockdown” and the requirement of self-isolation;
  3. COVID-19 and the control measures affected levels of employment, consumer behaviour, economic activity and confidence, both generally and in the particular sector in which the policyholder operates.”

Interestingly, Arch also pleads an alternative case of only points (1) and (3) which – depending on how the wide area damage principles are applied, may not be dissimilar to the FCA’s alternative case.

An alternative approach is set out in the Hiscox Defence which seeks to draw comparisons to countries where a full lockdown was not mandated:

In broad terms, which are all that is relevant, the Hiscox counter-factual occurred in Sweden in relation to COVID-19, as to which Hiscox relies on the facts set out in its proposed additional Agreed Facts Document 6 sent to the FCA on 17 June. Even in the absence of relevant mandatory measures, many businesses in Sweden may have experienced business or trading losses. As to this and for the avoidance of doubt, Hiscox does not seek to prove in the Test Case that any particular number of businesses in Sweden suffered such losses, or the amount of any losses, or there was in fact a reduction in economic activity of any actual amount in Sweden, but rather that there may have been a reduction of some amount, and that businesses may have suffered such losses.”

As discussed in further detail in Part 2 of this series, the principle of the Orient Express case on wide area damage is that the “but-for” position considers only the policy trigger. In the context of the Hiscox policies (Hiscox 1-4), the Public Authority clause is as follows:

“Your inability to use the insured premises due to restrictions imposed by a public authority during the period of insurance following: … an occurrence of any human infectious or human contagious disease, an outbreak of which must be notified to the local authority.”

The key question would appear to be is the trigger “an authority shutdown” or “an authority shutdown of the policyholder’s premises”. The former would mean that in the “but-for” position, the policyholder is open, but so are all of its competitors; in the latter, just the policyholder is open – clearly there would be a significant difference in the levels of expected revenue where the policyholder is the only game in town versus full competition.

Without wanting to stray too far into legal opinion, a possible analogy of the Hiscox position compared to the findings in Orient Express, is that Hiscox have considered the impact of the “hurricane” rather than “the damage caused by the hurricane to the policyholder’s premises” – in that case the latter was held to be the case.

The Hiscox position would appear to be a best-case position for insurers, in the same way that “no COVID-19, no Government restrictions” is the best-case position for policyholders. In my view, neither of these positions seems to really follow the principles of the Orient Express case, which seem to better fit the FCA’s alternative case of an “island in a sea of disease” - albeit there is a question of whether that island has travel restrictions and requires social distancing!

I should note that the FCA’s “island” alternative case was set out in Paragraph 79 of the Particulars of Claim. It is perhaps telling that insurers have chosen either to not plead against this paragraph, or – in the case of Hiscox – provide only a cursory response:

Paragraph 79 is denied and is in any event likely to be of no or minimal significance as regards Hiscox 1-4 because of the very small area of the putative “island”.”

As explored further below, I’m not convinced that is true. I also think it will be interesting to see if the FCA doubles down on its primary case in its Reply or whether it seeks to reinforce its alternative case.

Insurers’ case – a closer look

Putting aside the Hiscox position, I’m going to take the example of a restaurant in Central London -as I did for the policyholder case – and explore the impact of the insurers’ case of taking only the shutdown as the policy trigger, and not any of the other Government measures.

In this case, the restaurant is open, but – insurers say – it will be empty anyway due to other Government measures and general loss of confidence. Is that right? Throughout the pandemic there have been essential shops that are open and serving a number of customers – what is also apparent as we move out of lockdown is that different people have differing views on the risk of COVID-19.

If the policy trigger is an “authority closure” at the very least the but-for position must be that the restaurant is allowed to be open. Based on the wide area damage principles from Orient Express if there is no reference to a “Relevant Policy Area” then it could be argued that in the but-for position that restaurant is the only restaurant that is allowed to open… in Central London.

Even with travel restrictions, social distancing etc. in my view it is an unsustainable position to suggest that nobody would go to that restaurant. It might even require the input of a pandemic behavioural expert to give a steer as to what proportion of the population might visit a restaurant in such circumstances, but in the case of a London-based business it would only take a very small proportion of the local populace to be prepared to venture out in spite of Government advice to fill up a single restaurant many times over.

I note that there are arguments put forward by the FCA as to the applicability of the trend clauses - clauses that are used in policies to adjust the projected turnover of the business to account for other issues that would have impacted the business regardless of the insured event - to non-damage extensions. That aside, I would normally expect the balance of the impacts of the other measures (assuming they are not covered), the fact of the business being the only restaurant open and the behaviour of potential customers in a pandemic to be factored into a trend clause to arrive at an assumed but-for position… factoring in different variations in wording and the specific circumstances of each policyholder, that will be an incredibly complex undertaking.

To be clear, some insurers have acknowledged that some - in their view, very limited - level of loss may be covered. Hiscox has pleaded:

An insured may be able to prove that part of the loss which it has sustained was caused by an insured peril, but the burden of proof is on it to do so.”

The Court’s decision on which Government measures should be considered to have triggered which policy wordings, and the application of wide area damage principles would seem to be central to determining whether policyholders are likely to see any substantial recovery of financial losses under their Business Interruption insurance policies.

Prevention or hindrance to access – other issues impacting quantum

In addition to challenging the Government measures being one single intervention, insurers are challenging a number of aspects on the “Denial of Access” claims. This includes:

  • Vicinity – In regard to vicinity, insurers are putting policyholders to proof on whether there has been a case of COVID-19, and that any losses are directly attributable to those localised cases.
  • Policy language - Where the policy language speaks of “an emergency that is likely to endanger life” there seems to be an acceptance by insurers that this would be triggered by COVID-19. However, where the language speaks of an “occurrence” or “incident” insurers are arguing this requires a “specific” and “local” event. Some insurers are also arguing that that the authority shutdown did not “follow” the local occurrence of COVID-19 and so should not be covered.
  • Mandated closure or advice – insurers are looking to distinguish between businesses that were mandated to close (for which they accept that is a triggering event) versus business that were just advised to close which they argue is insufficient to trigger both prevention or hindrance to access covers as it was not “imposed” on the policyholder.
  • Interruption or interference – some Denial of Access covers refer to “interruption or interference” to a business. Where only “interruption” is referred to, insurers are arguing that this requires “a complete cessation of the business at the premises” or else no cover will be provided.

Insurers have put forward a lot of technical arguments for the Courts to consider, the impact of which will be very much dependant on the individual circumstance of the policyholder. The vicinity arguments might have little impact on a restaurant relying on local custom, but a much larger impact on a holiday cottage business relying on a wider geographical customer base. The cover may also be wider if the policy language speaks of “prevention or hindrance” rather than just “prevention” of access.

The most controversial aspect may be the arguments over whether the Government measures were mandated or merely advice. The FCA’s case is that a Government statement on 18 March – purportedly having been made after discussions with insurance companies - provided clarity on the order v advice issue, stating that:

The government’s medical advice of 16 March is sufficient to enable those businesses which have an insurance policy that covers both pandemics and government ordered closure to make a claim

However, the insurers involved in the test case have universally distanced themselves from having been involved in this decision, and do not accept that any such agreement was made with them.


Ostensibly, there are two types of non-damage extensions being considered in the test case – Public Authority / Disease cover and Denial of Access cover. However, the huge variation in wordings is resulting in a complex web of issues – only once clarity has been provided on the meaning of such wordings can we get a real picture of how quantum might realistically be calculated.

Next up is the FCA’s Reply on behalf of policyholders – due on 3 July - which will likely provide further nuance on the issues above.

Richard Cameron-Williams 
Partner - Forensic Accounting and Valuation Services