The Financial Conduct Authority (FCA) v Arch Insurance (UK) Ltd & Ors

Introduction

This case has been brought by the Financial Conduct Authority (“the FCA”), the regulator of the defendant insurers, as a test case to determine issues of principle in relation to policy coverage under various specimen wordings underwritten by the defendants in respect of claims by policyholders to be indemnified for business interruption losses arising in the context of the COVID-19 pandemic and the advice of and restrictions imposed by the UK Government in consequence. We have been asked to determine those issues as to the correct construction of the policy terms and as to whether cover is available in principle by reference to a set of agreed facts (which we summarise in the next section of the judgment) and assumed facts (which were essentially illustrative factual scenarios as to how certain businesses have been affected). Since, in their oral submissions, the parties did not rely upon the assumed facts, we have not felt it necessary to set them out in this judgment, although we have taken them into consideration in reaching our conclusions.

These proceedings were commenced by the FCA on 9 June 2020, seeking declarations in respect of the relevant business interruption policy wordings, pursuant to a

“Framework Agreement” between the parties dated 31 May 2020. The Framework Agreement provided that the parties have a mutual objective of achieving the maximum clarity possible for the maximum number of policyholders and their insurers, consistent with the need for expedition and proportionality. The parties agreed to put forward for consideration a representative sample of the standard form business interruption policies issued by the defendant insurers in the test case. In these proceedings, the FCA represents the interests of the large number of policyholders who purchased the relevant policies, many of whom are small to medium sized enterprises.

On the same day as the Claim Form was issued, the FCA made an application, supported by the defendant insurers, for (i) the expedition of the trial and (ii) the Financial Market Test Case Scheme under Practice Direction 51M to apply to the claim. That Practice Direction provides for the Scheme to apply to any Financial List claim which raises issues of general importance in relation to which immediately relevant authoritative English law guidance is needed. It also provides that in a case of particular importance or urgency the trial may, at the court’s discretion, be heard by a court consisting of two Financial List judges, or a Financial List judge and a Lord or Lady Justice of Appeal. The first Case Management Conference (“CMC”) on 16 June 2020 was heard by Butcher J alone. Orders were made for the expedition of the trial and for the case to be heard under the Scheme with the trial being conducted by what was in effect a Divisional Court consisting of Flaux LJ and Butcher J. The trial was fixed for 8 days from 20 July 2020.

At that first CMC, there was an issue about the FCA’s proposed reliance on certain expert evidence about the prevalence of the disease at various dates. The Court ruled that there would be no expert evidence going to the issue of actual prevalence, from either the FCA or the insurers. However, it was left open to the parties to raise arguments about what type of proof would be sufficient to satisfy the onus of proof on the insured about the prevalence of the disease, and the issue was to be revisited at the next CMC.

At the second CMC heard by both of us, we acceded to the applications of the Hiscox Action Group (“the Hiscox Interveners”) and the Hospitality Insurance Group Action

The CMCs and the trial took place wholly remotely, using Skype for Business, and were broadcast by live-streaming pursuant to Schedule 25 of the Coronavirus Act 2020. We received detailed written opening submissions from all the parties and the two interveners and oral submissions from counsel for the FCA, each of the defendants and the interveners. It was and is evident that a tremendous amount of hard work went into the preparation and presentation of the case at trial, on the part of all those involved and their legal advisers. We are grateful to all those involved for the clarity and intellectual rigour of the written and oral submissions and for the professional way in which the trial was conducted.

In all we had some 21 “lead” policies to consider: one issued by Arch Insurance (UK) Limited (“Arch 1”), one issued by Argenta Syndicate Management Limited (“Argenta 1”), two issued by Ecclesiastical Insurance Office plc (“Ecclesiastical 1.1” and “Ecclesiastical 1.2”), four issued by Hiscox Insurance Company Limited (“Hiscox 1”, “Hiscox 2”, “Hiscox 3” and “Hiscox 4”), three issued by MS Amlin Underwriting Limited (“MSA 1”, “MSA 2” and “MSA 3”), three issued by QBE UK Limited (“QBE 1”, “QBE 2” and “QBE 3”), five issued by Royal & Sun Alliance Insurance plc (“RSA 1”, “RSA 2.1”, “RSA 2.2”, “RSA 3” and “RSA 4”), and two issued by Zurich Insurance plc (“Zurich 1” and “Zurich 2”). Certain of the lead policies are related to other policies issued by insurers who are not participants in the proceedings, such as RSA 4 which is in similar terms to certain policies issued by Aviva. Indeed, the FCA estimated that, in addition to the particular policies chosen for the test case, some 700 types of policies across over 60 different insurers and 370,000 policyholders could potentially be affected by the test case.

The relevant provisions of the lead policies essentially fell into three categories: (i) what the FCA termed “Disease Clauses”; (ii) what have been referred to as “Hybrid Clauses”; and (iii) clauses covering prevention of access and similar perils. We will deal with the three categories in turn. The structure of the remainder of this judgment is as follows: Section B, Factual Background, Section C, Applicable Principles of Construction, Section D, disease clauses, Section E, hybrid clauses, Section F, Prevention of Access and similar wordings, Section G, Issues of Causation (although as will become apparent we have reached the conclusion that most, if not all, of the issues of causation raised at some length by the parties are answered by the correct construction of the wordings) and Section H, Prevalence.

Factual Background

As we have already said, the factual background is essentially agreed between the parties and it is only necessary to set out the history of the disease to date and the government response to it to the extent necessary to put in context the issues of construction which we have to determine.

On 31 December 2019, the World Health Organization (“WHO”) was informed of pneumonia cases of unknown cause in the city of Wuhan, in the Hubei province in China. On 12 January 2020, the WHO announced that a novel coronavirus had been identified in samples obtained from cases in China. This announcement was subsequently recorded by Public Health England (“PHE”). The virus was named severe acute respiratory syndrome coronavirus 2, or SARS-CoV-2, and the associated disease was named COVID-19. We will refer to it throughout the remainder of this judgment as “COVID-19”.

On 22 January 2020, the UK Department of Health and Social Care (“Department of Health”) and PHE issued a statement to the effect that the coronavirus situation was being constantly monitored. The risk level was raised from “very low” to “low”. At this stage, there were no confirmed cases in the UK. On 28 January 2020, the UK Foreign Office advised against travel to Wuhan. On 30 January 2020, the risk level was raised again, this time from “low” to “moderate”.

On 30 January 2020, the WHO declared the outbreak of COVID-19 a “Public Health Emergency of International Concern”.

On 31 January 2020, the Chief Medical Officer for England confirmed that two patients had tested positive for COVID-19 in England.

On 10 February 2020, the Health Protection (Coronavirus) Regulations 2020 were introduced by the Secretary of State for Health and Social Care, pursuant to powers under the Public Health (Control of Disease) Act 1984 (“the 1984 Act”). In broad terms, the regulations provided for the detention and screening of persons reasonably suspected to have been infected or contaminated with COVID-19. The regulations were subsequently repealed on 25 March 2020 by the Coronavirus Act 2020.

On 22 February 2020, COVID-19 was made a “notifiable disease”, and SARS-CoV-2 made a “notifiable organism”, in Scotland by amendment to the Public Health etc. (Scotland) Act 2008. The same followed for Northern Ireland, England and Wales on the dates set out below. The notification regime is discussed further below.

On 25 February 2020, the UK Government announced that travellers from certain locations including Hubei province should self-isolate at home regardless of whether they had any symptoms. On the same day, the UK Government published guidance for employers and businesses about sanitisation and other matters. On 27 February 2020, the first case of COVID-19 was confirmed in Northern Ireland. The following day, the first case was confirmed in Wales. On 29 February 2020, COVID-19 was made a notifiable disease in Northern Ireland by amendment to the Public Health Act (Northern Ireland) 1967.

On 1 March 2020, the first case of COVID-19 was confirmed in Scotland. On 2 March 2020, the UK recorded the first death of an individual who had tested positive for COVID-19. The parties agreed that this is apparent from certain death data published by the National Health Service (“NHS”), although the first death from COVID-19 was publicly announced by the Chief Medical Officer for England on 5 March 2020.

On 3 March 2020, the UK Government announced an “action plan” setting out its response to the COVID-19 outbreak. That plan set out information that was known at the time about the virus and its spread, as well as the UK’s preparations for infectious disease outbreaks. It included discussion of four phases of the response: “contain”, “delay”, “research” and “mitigate”.

On 4 March 2020, the UK Government published guidance titled “Coronavirus (COVID-19): What is social distancing?” on the PHE Public Health Matters blog. It referred to the Government’s action plan from the previous day and the possibility of introducing social distancing measures. It asked people to think about how they could minimise contact with others.

On 5 March 2020, the Scientific Advisory Group for Emergencies (“SAGE”) held a meeting, the minutes of which record that there was epidemiological and modelling data to support the implementation of isolation measures for people with symptoms (and their families) within one to two weeks, and then for vulnerable people roughly two weeks after that. On the same day, 5 March 2020, COVID-19 was made a notifiable disease, and SARS-CoV-2 made a “causative agent”, in England by amendment to the Health Protection (Notification) Regulations 2010 (“the 2010 Regulations”). On 6 March 2020, the same followed for Wales by amendment to the Health Protection (Notification) (Wales) Regulations 2010.

Accordingly, by 6 March 2020, COVID-19 was a notifiable disease across the UK. Notifiable diseases are diseases in respect of which medical professionals and others have statutory responsibilities to notify public authorities. In England, for example, the notification regime is established by the 1984 Act together with the 2010 Regulations.

On 11 March 2020, the WHO declared COVID-19 to be a pandemic.

On 12 March 2020, the UK Government announced that it was moving from the “contain” phase to the “delay” phase of its action plan and raised the risk level from “moderate” to “high”. Those with symptoms were told to self-isolate for 7 days.

On 13 March 2020, SAGE held a meeting, the minutes of which record that “SAGE now believes there are more cases in the UK than SAGE previously expected at this point … The science suggests that household isolation and social distancing of the elderly and vulnerable should be implemented soon”.

On 16 March 2020, there was a further SAGE meeting, the minutes of which record that “SAGE advises that there is clear evidence to support additional social distancing measures be introduced as soon as possible”. The same day, the UK Government published guidance on social distancing. The guidance advised vulnerable people to

The same day, on 16 March 2020, the Prime Minister, the Rt Hon Boris Johnson MP, made a statement which included the following:

“Last week we asked everyone to stay at home if you had one of two key symptoms: a high temperature or a new and continuous cough. Today, we need to go further, because according to SAGE it looks as though we’re now approaching the fast growth part of the upward curve. And without drastic action, cases could double every 5 or 6 days.

So, first, we need to ask you to ensure that if you or anyone in your household has one of those two symptoms, then you should stay at home for fourteen days. That means that if possible you should not go out even to buy food or essentials, other than for exercise, and in that case at a safe distance from others. If necessary, you should ask for help from others for your daily necessities. And if that is not possible, then you should do what you can to limit your social contact when you leave the house to get supplies. And even if you don’t have symptoms and if no one in your household has symptoms, there is more that we need you to do now.

So, second, now is the time for everyone to stop non-essential contact with others and to stop all unnecessary travel. We need people to start working from home where they possibly can. And you should avoid pubs, clubs, theatres and other such social venues. It goes without saying, we should all only use the NHS when we really need to. And please go online rather than ringing NHS 111. Now, this advice about avoiding all unnecessary social contact, is particularly important for people over 70, for pregnant women and for those with some health conditions …

So third, in a few days’ time – by this coming weekend – it will be necessary to go further and to ensure that those with the most serious health conditions are largely shielded from social contact for around 12 weeks …

And it’s now clear that the peak of the epidemic is coming faster in some parts of the country than in others. And it looks as though London is now a few weeks ahead. So, to relieve the pressure on the London health system and to slow the spread in London, it’s important that Londoners now pay special attention to what we are saying about avoiding non-essential contact, and to take particularly seriously the advice about working from home, and avoiding confined spaces such as pubs and restaurants.

Lastly, it remains true as we have said in the last few weeks that risks of transmission of the disease at mass gatherings such as sporting events are relatively low. But obviously, logically as we advise against unnecessary social contact of all kinds, it is right that we should extend this advice to mass gatherings as well. And so we’ve also got to ensure that we have the critical workers we need, that might otherwise be deployed at those gatherings, to deal with this emergency. So from tomorrow, we will no longer be supporting mass gatherings with emergency workers in the way that we normally do. So mass gatherings, we are now moving emphatically away from.”

On 17 March 2020, the Chancellor of the Exchequer, the Rt Hon Rishi Sunak MP, announced financial measures for businesses in response to COVID-19. That announcement included the following:

“Second, as well as access to finance, businesses need support with their cashflow and fixed costs. Following the changed medical advice yesterday, there are concerns about the impact on pubs, clubs, theatres and other hospitality, leisure and retail venues. Let me confirm that, for those businesses which do have a policy that covers pandemics, the government’s action is sufficient and will allow businesses to make an insurance claim against their policy.”

On 17 March 2020, a meeting took place between the Economic Secretary to the Treasury and various representatives of the insurance industry. Of the current defendants only Hiscox, RSA and Zurich attended the meeting. After the meeting, the Chancellor of the Exchequer made various statements in Parliament about this meeting, to the effect that insurers had agreed to “do the right thing”. The FCA pleaded this meeting and what the Chancellor of the Exchequer had said against the insurers. Although at the second CMC we declined to accede to an application by Mr Gavin Kealey QC for Ecclesiastical Insurance Office plc and MS Amlin Underwriting Limited (neither of which had been represented at the meeting) to strike out this plea, we indicated our firm view that, even in relation to the insurers represented at the meeting (and a fortiori in relation to those who were not), the meeting and what was said at it or after it was of no relevance to the issues of construction which we had to decide. In the event, the FCA made no mention of the meeting in their oral submissions and it will not be necessary to refer to it again.

On 18 March 2020, HM Treasury published a Fact Sheet entitled “How to access government financial support if you or your business has been affected by COVID-19”. That fact sheet included the following (emphasis in original):

“If the only barrier to your business making an insurance claim was a lack of clarity on whether the government advising people to stay away from businesses, rather than ordering businesses to shut down, was sufficient to make a claim on business interruption insurance:

•

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 - provided all other terms and conditions in their policy are met. Businesses should check the terms and conditions of their specific policy and contact their providers if in doubt.

•

However, most businesses have not purchased insurance that covers pandemic related losses. As such, any affected businesses should note the government’s full package of support, including the Coronavirus Business Interruption Loan Scheme and business rates holiday.”

On 18 March 2020, SAGE held a further meeting, the minutes of which record that “SAGE considers that the UK is 2 to 4 weeks behind Italy in terms of the epidemic curve” and that school closures should occur as soon as practicable. According to the minutes, SAGE advised that the measures that had already been announced should have a “significant effect, provided compliance rates are good and in line with the assumptions”. In respect of social distancing, SAGE advised that it should be based on places of leisure, such as restaurants and bars, and indoor workplaces.

On 18 March 2020, the Prime Minister made an announcement reiterating his statement from two days earlier to avoid non-essential contact and travel, and to work from home if possible. In that announcement, the Prime Minister said the following in respect of schools:

“And we come today to the key issue of schools where we have been consistently advised that there is an important trade off.

…

So looking at the curve of the disease and looking at where we are now – we think now that we must apply downward pressure, further downward pressure on that upward curve by closing the schools. So I can announce today and Gavin Williamson making statement now in House of Commons that after schools shut their gates from Friday afternoon, they will remain closed for most pupils – for the vast majority of pupils – until further notice. I will explain what I mean by the vast majority of pupils. …

But of course, as I’ve always said, we also need to keep the NHS going and to treat the number of rising cases. So we need health workers who are also parents to continue to go to work. And we need other critical workers with children to keep doing their jobs too – from police officers who are keeping us safe to the supermarket delivery drivers, social care workers who look after the elderly and who are so vital. We will be setting out more details shortly about who we mean in these groups.

So we therefore need schools to make provision for the children of these key workers who would otherwise be forced to stay home. And they will also need to look after the most vulnerable children.”

On 20 March 2020, the Prime Minister made an announcement including the following:

“And that means we have to take the next steps, on scientific advice and following our plan, we are strengthening the measures announced on Monday which you will remember. And of course people have already made a huge effort to comply with those measures for avoiding unnecessary social contact. But we need now to push down further on that curve of transmission between us.

And so following agreement between all the formations of the United Kingdom, all the devolved administrations, we are collectively telling, telling cafes, pubs, bars, restaurants to close tonight as soon as they reasonably can, and not to open tomorrow. Though to be clear, they can continue to provide take-out services.

We’re also telling nightclubs, theatres, cinemas, gyms and leisure centres to close on the same timescale. Now, these are places where people come together, and indeed the whole purpose of these businesses is to bring people together. But the sad things [sic] is that today for now, at least physically, we need to keep people apart. And I want to stress that we will review the situation each month, to see if we can relax any of these measures.

And listening to what I have just said, some people may of course be tempted to go out tonight. But please don’t. You may think you are invincible, but there is no guarantee you will get mild symptoms, and you can still be a carrier of the disease and pass it on to others. So that’s why, as far as possible, we want you to stay at home, that’s how we can protect our NHS and save lives.”

On 20 March 2020, the Chancellor of the Exchequer announced a Coronavirus Job Retention Scheme which included the furloughing of workers and other financial assistance.

On 21 March 2020, the Health Protection (Coronavirus, Business Closure) (England) Regulations 2020 (“the 21 March Regulations”) were made by the Secretary of State for Health and Social Care pursuant to powers under the 1984 Act. Equivalent provisions were introduced in Wales under the Health Protection (Coronavirus, Business Closure) (Wales) Regulations 2020.

The 21 March Regulations provided for the closure of certain businesses. Regulation 2 provided as follows:

“Requirement to close premises and businesses during the emergency

2.—(1) A person who is responsible for carrying on a business which is listed in Part 1 of the Schedule must—

(a)

during the relevant period—

(i)

close any premises, or part of the premises, in which food or drink are sold for consumption on those premises, and

(ii)

cease selling food or drink for consumption on its premises; or

(b)

if the business sells food or drink for consumption off the premises, cease selling food or drink for consumption on its premises during the relevant period.

(2)

For the purposes of paragraph (1)(a), food or drink sold by a hotel or other accommodation as part of room service is not to be treated as being sold for consumption on its premises.

(3)

For the purposes of paragraph (1)(a)(ii) and (b), an area adjacent to the premises of the business where seating is made available for customers of the business (whether or not by the business) to be treated as part of the premises of that business.

(4)

A person responsible for carrying on a business which is listed in Part 2 of the Schedule must cease to carry on that business during the relevant period.

(5)

If a business listed in the Schedule (“business A”) forms part of a larger business (“business B”), the person responsible for carrying on business B complies with the requirement in paragraph (1) if it closes down business A.

(6)

The Secretary of State must review the need for restrictions imposed by this regulation every 28 days, with the first review being carried out before the expiry of the period of 28 days starting with the day after the day on which these Regulations are made.

(7)

As soon as the Secretary of State considers that the restrictions set out in this regulation are no longer necessary to prevent, protect against, control or provide a public health response to the incidence or spread of infection in England with the coronavirus, the Secretary of State must publish a direction terminating the relevant period.

(8)

A direction published under paragraph (7) may terminate the relevant period in relation to some of the businesses listed in the Schedule, or all businesses listed in the Schedule.

(9)

For the purposes of this regulation—

(a)

“coronavirus” means severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2);

(b)

a “person responsible for carrying on a business” includes the owner, proprietor, and manager of that business;

(c)

the “relevant period” starts when these Regulations come into force and ends on the day specified in a direction published by the Secretary of State under paragraph (7).”

Regulation 3 of the 21 March Regulations provided that contravention of Regulation 2 without reasonable excuse was an offence, punishable on summary conviction by a fine. Regulation 4(1) provided that a person designated by the Secretary of State may take action as necessary to enforce a closure or restriction imposed by Regulation 2.

The businesses listed in Part 1 of the Schedule included restaurants, cafes, bars and public houses. Those businesses were required to close or cease carrying on the business of selling food and drink other than for consumption off the premises under Regulation 2(1). The businesses listed in Part 2 of the Schedule included cinemas, theatres, nightclubs, bingo halls, concert halls, museums, galleries, casinos, spas and gyms. Those businesses were required to cease carrying on business under Regulation

On 22 March 2020, the Prime Minister announced the next stage of the UK Government’s plan, which included “shielding” measures for vulnerable people and advising the public to stay 2 metres apart even when outdoors. The guidance for people to stay 2 metres apart was reiterated by PHE the following day.

On 23 March 2020, there was a further SAGE meeting, the minutes of which provide that “high rates of compliance for social distancing will be needed to bring the reproduction number below one and to bring cases within NHS capacity”. The reproduction number or “R” is a reference to the average number of secondary infections produced by 1 infected person. The minutes also record that “Public polling over the weekend on behaviour indicated significant changes but room for improvement in compliance rates”.

On 23 March 2020, the Prime Minister made an announcement which included the following:

“From this evening I must give the British people a very simple instruction - you must stay at home. Because the critical thing we must do is stop the disease spreading between households.

That is why people will only be allowed to leave their home for the following very limited purposes:

•

shopping for basic necessities, as infrequently as possible

•

one form of exercise a day - for example a run, walk, or cycle - alone or with members of your household;

•

any medical need, to provide care or to help a vulnerable person; and

•

travelling to and from work, but only where this is absolutely necessary and cannot be done from home.

That’s all - these are the only reasons you should leave your home. You should not be meeting friends. If your friends ask you to meet, you should say No. You should not be meeting family members who do not live in your home. You should not be going shopping except for essentials like food and medicine - and you should do this as little as you can. And use food delivery services where you can. If you don’t follow the rules the police will have the powers to enforce them, including through fines and dispersing gatherings.

To ensure compliance with the Government’s instruction to stay at home, we will immediately:

•

close all shops selling non-essential goods, including clothing and electronic stores and other premises including libraries, playgrounds and outdoor gyms, and places of worship;

•

we will stop all gatherings of more than two people in public – excluding people you live with;

•

and we’ll stop all social events, including weddings, baptisms and other ceremonies, but excluding funerals.”

On 23 March 2020, the UK Government issued guidance to businesses about closures.

On 24 March 2020, the UK Government issued guidance to holiday accommodation providers to the effect that they should have taken steps to close for commercial use and to remain open only for limited prescribed purposes, for example supporting key workers or homeless people.

On 25 March 2020, the Coronavirus Act 2020 was passed. The Act applies across the UK, although different provisions have come into force in different nations at different times. In broad terms, the Act provides for emergency arrangements in relation to health workers, food supply, inquests and other matters. Sections 37 and 38, and Schedules 16 and 17, of the Act provide the Secretary of State with power to give temporary directions to close “educational institutions”, though to date that power has not been exercised.

On 26 March 2020, the Health Protection (Coronavirus, Restrictions) (England)

The 26 March Regulations revoked most of the 21 March Regulations and introduced a more expansive regime for business closures, as explained below. The 21 March Regulations remain in force to the limited extent that they provide for offences committed between 21 March 2020 and 25 March 2020 (Regulation 2(2) of the 26 March Regulations).

Regulation 3 of the 26 March Regulations defined the emergency period as follows:

“The emergency period and review of need for restrictions

3.—(1) For the purposes of these Regulations, the “emergency period”—

(a)

starts when these Regulations come into force, and

(b)

ends in relation to a restriction or requirement imposed by these Regulations on the day and at the time specified in a direction published by the Secretary of State terminating the requirement or restriction.

(2)

The Secretary of State must review the need for restrictions and requirements imposed by these Regulations at least once every 21 days, with the first review being carried out by 16th April 2020.

(3)

As soon as the Secretary of State considers that any restrictions or requirements set out in these Regulations are no longer necessary to prevent, protect against, control or provide a public health response to the incidence or spread of infection in England with the coronavirus, the Secretary of State must publish a direction terminating that restriction or requirement. …”

Regulation 4 of the 26 March Regulations was in a similar form to Regulation 2 of the 21 March Regulations and provided as follows:

“Requirement to close premises and businesses during the emergency

4.—(1) A person responsible for carrying on a business which is listed in Part 1 of Schedule 2 must—

(a)

during the emergency period—

(i)

close any premises, or part of the premises, in which food or drink are sold for consumption on those premises, and

(ii)

cease selling food or drink for consumption on its premises; or

(b)

if the business sells food or drink for consumption off the premises, cease selling food or drink for consumption on its premises during the emergency period.

(2)

For the purposes of paragraph (1)(a), food or drink sold by a hotel or other accommodation as part of room service is not to be treated as being sold for consumption on its premises.

(3)

For the purposes of paragraph (1)(a)(ii) and (b), an area adjacent to the premises of the business where seating is made available for customers of the business (whether or not by the business) is to be treated as part of the premises of that business.

(4)

A person responsible for carrying on a business or providing a service which is listed in Part 2 of Schedule 2 must cease to carry on that business or to provide that service during the emergency period.

(5)

Paragraph (4) does not prevent the use of—

(a)

premises used for the businesses or services listed in paragraphs 5, 6, 8, 9 or 10 of that Part to broadcast a performance to people outside the premises, whether over the internet or as part of a radio or television broadcast;

(b)

any suitable premises used for the businesses or services listed in that Schedule to host blood donation sessions.

(6)

If a business listed in Part 1 or 2 of Schedule 2 (“business A”) forms part of a larger business (“business B”), the person responsible for carrying on business B complies with the requirement in paragraph (1) if it closes down business A.”

Regulation 5 introduced new closures for retail shops, holiday accommodation providers and places of worship:

“Further restrictions and closures during the emergency period

5.—(1) A person responsible for carrying on a business, not listed in Part 3 of Schedule 2, of offering goods for sale or for hire in a shop, or providing library services must, during the emergency period—

(a)

cease to carry on that business or provide that service except by making deliveries or otherwise providing services in response to orders received—

(i)

through a website, or otherwise by on-line communication,

(ii)

by telephone, including orders by text message, or (iii) by post;

(b)

close any premises which are not required to carry out its business or provide its services as permitted by sub-paragraph

(a);

(c)

cease to admit any person to its premises who is not required to carry on its business or provide its service as permitted by subparagraph (a).

(2)

Paragraph (1) does not apply to any business which provides hot or cold food for consumption off the premises.

(3)

Subject to paragraph (4), a person responsible for carrying on a business consisting of the provision of holiday accommodation, whether in a hotel, hostel, bed and breakfast accommodation, holiday apartment, home, cottage or bungalow, campsite, caravan park or boarding house, must cease to carry on that business during the emergency period.

(4)

A person referred to in paragraph (3) may continue to carry on their business and keep any premises used in that business open—

(a)

to provide accommodation for any person, who—

(i)

is unable to return to their main residence;

(ii)

uses that accommodation as their main residence;

(iii)

needs accommodation while moving house;

(iv)

needs accommodation to attend a funeral;

(b)

to provide accommodation or support services for the homeless,

(c)

to host blood donation sessions, or

(d)

for any purpose requested by the Secretary of State, or a local authority.

(5)

A person who is responsible for a place of worship must ensure that, during the emergency period, the place of worship is closed, except for uses permitted in paragraph (6).

(6)

A place of worship may be used— (a) for funerals,

(b)

to broadcast an act of worship, whether over the internet or as part of a radio or television broadcast, or

(c)

to provide essential voluntary services or urgent public support services (including the provision of food banks or other support for the homeless or vulnerable people, blood donation sessions or support in an emergency).

(7)

A person who is responsible for a community centre must ensure that, during the emergency period, the community centre is closed except where it is used to provide essential voluntary activities or urgent public support services (including the provision of food banks or other support for the homeless or vulnerable people, blood donation sessions or support in an emergency).

(8)

A person who is responsible for a crematorium or burial ground must ensure that, during the emergency period, the crematorium is closed to members of the public, except for funerals or burials.

(9)

If a business referred to in paragraph (1) or (3) (“business A”) forms part of a larger business (“business B”), the person responsible for carrying on business B complies with the requirement in paragraph (1) or (3) to cease to carry on its business if it ceases to carry on business A.”

Regulation 6(1) set out the general prohibition against people leaving the place where they were living “without reasonable excuse”. Regulation 6(2) provided a nonexhaustive list of reasonable excuses, including the need to obtain basic necessities, take exercise, seek medical assistance, or fulfil a legal obligation. Regulation 7 prohibited gatherings in public places of more than two people other than in limited circumstances.

Regulation 8(1) provided “relevant persons” with the power to take such action as necessary to enforce any requirements imposed by Regulations 4, 5 or 7. Under Regulation 8(3) and (4), where a relevant person considered that a person was outside the place they were living contrary to Regulation 6(1), a relevant person could direct them to return to the place they were living or remove them to the place where they were living, including by using reasonable force if necessary. “Relevant person” was defined in Regulation 8(12)(a) to include a constable, a police community support officer or a person designated by the Secretary of State.

Regulation 9(1) provided that a contravention of Regulations 4, 5, 7 or 8 without reasonable excuse was an offence and any contravention of Regulation 6 was an offence. Such offences were punishable on summary conviction by a fine (Regulation 9(4)). The parties agreed that there were several reports of enforcement action being taken under these sections in the months after 26 March 2020. Regulation 10(1) provided “authorised persons” with powers to issue fixed penalty notices. “Authorised person” was defined to include a constable, a police community support officer or a person designated by the Secretary of State. Regulation 11 provided that the Crown Prosecution Service, and any person designated by the Secretary of State, could bring proceedings for an offence under the regulations. Regulation 12(1) provided that the regulations would expire at the end of the 6 month period beginning with 26 March 2020.

Schedule 2 to the 26 March Regulations was entitled “Businesses subject to restrictions or closure” and provided a list of businesses in three parts:

“PART 1

For the purposes of these proceedings, the parties adopted a categorisation of businesses in light of the 21 March Regulations and the 26 March Regulations as follows:

On 4 April 2020, the Secretary of State for Health and Social Care designated local councils, including district councils, county councils and London borough councils, as “relevant persons” and “authorised persons” under the enforcement provisions in the 26 March Regulations. Specifically, local councils were empowered to take action and issue fixed penalty notices under Regulations 8 and 10 for the enforcement of Regulations 4 and 5. They were also empowered under Regulation 11 to bring proceedings for an offence under Regulations 4 and 5.

On 16 April 2020, the restrictions in the 26 March Regulations were continued for a further three week period. On 28 April 2020, the Secretary of State for Health and Social Care, the Rt Hon Matt Hancock MP, responded to a question in a daily press conference in the following terms:

“There was a big benefit, I think, as we brought in the lockdown measures, of the whole country moving together. We did think about moving with London and the Midlands first, because they were more advanced in terms of the number of cases, but we decided that we are really in this together, and the shape of the curve, if not the height of the curve, has been very similar across the whole country. It went up more in

London but it’s also come down more, but the broad shape has been similar, which is what you’d expect, given that we’ve all been living through the same lockdown measures. The other thing to say is that it isn’t just about the level, it’s also about the slope of the curve, and if the R goes above one anywhere, then that would eventually lead to an exponential rise and a second peak and an overwhelming of the NHS in that area unless it’s addressed, so although the level of the number of cases is different in different parts, the slope of the curve has actually been remarkably similar across the country, so that argues for doing things as a whole country together.”

On 10 May 2020, the Prime Minister made an announcement in respect of lifting restrictions in stages over the coming months:

“And so no, this is not the time simply to end the lockdown this week. Instead we are taking the first careful steps to modify our measures. And the first step is a change of emphasis that we hope that people will act on this week. We said that you should work from home if you can, and only go to work if you must. We now need to stress that anyone who can’t work from home, for instance those in construction or manufacturing, should be actively encouraged to go to work. And we want it to be safe for you to get to work. So you should avoid public transport if at all possible – because we must and will maintain social distancing, and capacity will therefore be limited. So work from home if you can, but you should go to work if you can’t work from home.

And to ensure you are safe at work we have been working to establish new guidance for employers to make workplaces COVID-secure. And when you do go to work, if possible do so by car or even better by walking or bicycle. But just as with workplaces, public transport operators will also be following COVID-secure standards.

And from this Wednesday, we want to encourage people to take more and even unlimited amounts of outdoor exercise. You can sit in the sun in your local park, you can drive to other destinations, you can even play sports but only with members of your own household. You must obey the rules on social distancing and to enforce those rules we will increase the fines for the small minority who break them …

In step two – at the earliest by June 1 – after half term – we believe we may be in a position to begin the phased reopening of shops and to get primary pupils back into schools, in stages, beginning with reception, Year 1 and Year 6. …

And step three - at the earliest by July - and subject to all these conditions and further scientific advice; if and only if the numbers support it, we will hope to re-open at least some of the hospitality industry and other public places, provided they are safe and enforce social distancing…”

On 11 May 2020, the UK Government published a “recovery strategy”.

The 26 March Regulations, and their equivalents in Wales, Northern Ireland and Scotland, were amended on several occasions. For example, on 13 May 2020, Category 3 businesses, that is those businesses listed in Part 3 of Schedule 2, were expanded to include garden centres and outdoor sports courts. Similarly, on 1 June 2020, that list was expanded further to include outdoor markets and certain showrooms.

On 3 July 2020, the Secretary of State made specific regulations in relation to a “protected area” defined to be an area surrounding the city of Leicester in the Health Protection (Coronavirus, Restrictions) (Leicester) Regulations 2020 (“the Leicester Regulations”). The Leicester Regulations adopt a similar approach to the 26 March Regulations, in that they specify an emergency period and impose restrictions on certain businesses and individuals.

On 4 July 2020, the 26 March Regulations were revoked and replaced with more limited restrictions in the Health Protection (Coronavirus, Restrictions) (No. 2) (England) Regulations 2020 in England (“the 4 July Regulations”). The 26 March Regulations remain in force, however, in relation to offences committed in contravention of those regulations between 26 March 2020 and 4 July 2020 (Regulation 2(2) of the 4 July Regulations).

Principles of construction

Each business interruption policy is a contract between an individual policyholder, whose interests are represented in this case by the FCA, and an individual insurer. The ordinary principles of contractual construction apply.

The general principles of construction were not in dispute. The court must ascertain what a reasonable person, that is, a person who has all the background knowledge which would reasonably have been available to the parties in the situation in which they were at the time of the contract, would have understood the contracting parties to have meant by the language used: Rainy Sky SA v Kookmin Bank [2011] UKSC 50, [2011] 1 WLR 2900 at [14]. This means disregarding evidence about the subjective intentions of the parties: Rainy Sky at [19]; Arnold v Britton [2015] UKSC 36, [2015] AC 1619 at [15].

In Wood v Capita Insurance Services Ltd [2017] UKSC 24, [2017] AC 1173, Lord Hodge set out the applicable principles following Rainy Sky and Arnold v Britton as follows:

“10.

The court’s task is to ascertain the objective meaning of the language which the parties have chosen to express their agreement. It has long been accepted that this is not a literalist exercise focused solely on a parsing of the wording of the particular clause but that the court must consider the contract as a whole and, depending on the nature, formality and quality of drafting of the contract, give more or less weight to elements of the wider context in reaching its view as to that objective meaning. In Prenn v Simmonds [1971] 1 WLR 1381, 1383H1385D and in Reardon Smith Line Ltd v Yngvar HansenTangen (trading as HE Hansen-Tangen) [1976] 1 WLR 989, 997, Lord Wilberforce affirmed the potential relevance to the task of interpreting the parties’ contract of the factual background known to the parties at or before the date of the contract, excluding evidence of the prior negotiations. When in his celebrated judgment in Investors Compensation Scheme Ltd v West Bromwich Building Society [1998] 1 WLR 896, 912-913 Lord Hoffmann reformulated the principles of contractual interpretation, some saw his second principle, which allowed consideration of the whole relevant factual background available to the parties at the time of the contract, as signalling a break with the past. But Lord Bingham of Cornhill in an extrajudicial writing, “A New Thing Under the Sun? The Interpretation of Contracts and the ICS decision” (2008) 12 Edin LR 374, persuasively demonstrated that the idea of the court putting itself in the shoes of the contracting parties had a long pedigree.

11.

Lord Clarke of Stone-cum-Ebony JSC elegantly summarised the approach to construction in the Rainy Sky case [2011] 1 WLR 2900, para 21f. In the Arnold case [2015] AC 1619 all of the judgments confirmed the approach in the Rainy Sky case: Lord Neuberger of Abbotsbury PSC, paras 13-14; Lord Hodge JSC, para 76 and Lord Carnwath JSC, para 108. Interpretation is, as Lord Clarke JSC stated in the Rainy Sky case (para 21), a unitary exercise; where there are rival meanings, the court can give weight to the implications of rival constructions by reaching a view as to which construction is more consistent with business common sense. But, in striking a balance between the indications given by the language and the implications of the competing constructions the court must consider the quality of drafting of the clause (the Rainy Sky case, para 26, citing Mance LJ in Gan Insurance Co Ltd v Tai Ping Insurance Co Ltd (No 2) [2001] 2 All ER (Comm) 299, paras 13, 16); and it must also be alive to the possibility that one side may have agreed to something which with hindsight did not serve his interest: the Arnold case, paras 20, 77. Similarly, the court must not lose sight of the possibility that a provision may be a negotiated compromise or that the negotiators were not able to agree more precise terms.

12.

This unitary exercise involves an iterative process by which each suggested interpretation is checked against the provisions of the contract and its commercial consequences are investigated: the Arnold case, para 77 citing In re Sigma Finance Corpn [2010] 1 All ER 571, para 12, per Lord Mance JSC. To my mind once one has read the language in dispute and the relevant parts of the contract that provide its context, it does not matter whether the more detailed analysis commences with the factual background and the implications of rival constructions or a close examination of the relevant language in the contract, so long as the court balances the indications given by each.

13.

Textualism and contextualism are not conflicting paradigms in a battle for exclusive occupation of the field of contractual interpretation. Rather, the lawyer and the judge, when interpreting any contract, can use them as tools to ascertain the objective meaning of the language which the parties have chosen to express their agreement. The extent to which each tool will assist the court in its task will vary according to the circumstances of the particular agreement or agreements. Some agreements may be successfully interpreted principally by textual analysis, for example because of their sophistication and complexity and because they have been negotiated and prepared with the assistance of skilled professionals. The correct interpretation of other contracts may be achieved by a greater emphasis on the factual matrix, for example because of their informality, brevity or the absence of skilled professional assistance. But negotiators of complex formal contracts may often not achieve a logical and coherent text because of, for example, the conflicting aims of the parties, failures of communication, differing drafting practices, or deadlines which require the parties to compromise in order to reach agreement. There may often therefore be provisions in a detailed professionally drawn contract which lack clarity and the lawyer or judge in interpreting such provisions may be particularly helped by considering the factual matrix and the purpose of similar provisions in contracts of the same type. The iterative process, of which Lord Mance JSC spoke in Sigma Finance Corpn [2010] 1 All ER 571, para 12, assists the lawyer or judge to ascertain the objective meaning of disputed provisions.”

The unitary exercise of interpreting the contract requires the court to consider the commercial consequences of competing constructions, but as Lord Neuberger said in Arnold v Britton at [19]-[20], commercial common sense should not be invoked retrospectively, or to rewrite a contract in an attempt to assist an unwise party, or to penalise an astute party. Where the parties have used unambiguous language, the court should apply it: Rainy Sky at [23].

There may be certain cases, however, where the background and context drive a court to the conclusion that “something must have gone wrong with the language”: Chartbrook Ltd v Persimmon Homes Ltd [2009] 1 AC 1101 at [14] (Lord Hoffmann); Investors Compensation Scheme Ltd v West Bromwich Building Society [1998] 1 WLR

Arguments which rely on what is absent from the drafting of the contract are to be treated with caution and in many cases provide little assistance: Netherlands v Deutsche Bank AG [2019] EWCA Civ 771 at [59]. In the context of an insurance policy, if one cover is subject to an exclusion whereas another is not, the absence of that exclusion in respect of the latter cover is not decisive as to its scope: Burger v Indemnity Mutual Marine Assurance Co [1900] 2 QB 348 at 351.

In respect of certain policies, reference was made by the parties to the maxims or canons of construction ejusdem generis (of the same kind) and noscitur a sociis (known by its associates). These are specific applications of the primary principle, which is to read the words of a particular provision in context: see Colinvaux’s Law of Insurance (12^th ed) at §3-055.

For instance, if a clause in an insurance policy covers, or excludes, the risk of damage to a number of items, it is likely that the words used denote things of the same genus (ejusdem generis), and each word can take its meaning from the words with which it is linked or surrounded (noscitur a sociis). In Watchorn v Langford (1813) 170 ER 1432, the insurance policy covered “stock in trade, household furniture, linen, wearing apparel and plate”. When the insured’s linen drapery goods were destroyed in a fire, the House of Lords held that the policy did not respond because the reference to “linen” must have been to household linen or linen in clothing, rather than drapery.

A more recent illustration can be seen in Tektrol Ltd (formerly Atto Power Controls Ltd) v International Insurance Co of Hanover Ltd [2005] EWCA Civ 845, [2006] 1 All ER (Comm) 780 where an insurance policy excluded liability for “erasure loss distortion or corruption of information on computer systems”. Sir Martin Nourse (agreeing with Buxton LJ) noted that “loss” in this context was a reference to loss by electronic means, rather than the burglary of a computer, citing the maxim noscitur a sociis (at [29]). That case also involved consideration of the meaning of “malicious person” within another exclusion containing the phrase “rioters strikers locked-out workers persons taking part in labour disturbances or civil commotion or malicious persons”. In that context, given the other categories of persons in the list, malicious person was held not to be a reference to a person who hacked in remotely to the computer systems in question (at [11]-[12]).

The principle of noscitur a sociis is, however, one which only operates if there can be said to be a common characteristic of the surrounding words, and it is a principle which must in any event give way if the particular words, or other features of the contract so dictate.

Reference was also made to the so-called contra proferentem rule, which provides that where there is a doubt about the meaning of a contract, the words will be construed against the person who put them forward (the proferens). Given the difficulties in identifying the proferens in different contexts, several different lines of authority have developed: Lewison on the Interpretation of Contracts (6^th ed) at §7.08. What is said to be potentially relevant for present purposes is the rule that, in the event of ambiguity, an exclusion clause in an insurance policy should be read narrowly (Cornish v Accident Insurance Co Ltd (1889) 23 QBD 453 at 456). It is clear that that principle, if it still has any validity, can only apply if there is genuine ambiguity, which cannot otherwise be resolved by applying the ordinary principles of construction. In other words, it should not be relied on to create ambiguity where there is none: McGeown v Direct Travel Insurance [2003] EWCA Civ 1606, [2004] 1 Lloyd’s Rep IR 599 at 603 [13] (Auld LJ).

The Supreme Court’s decisions in Rainy Sky, Arnold v Britton and Wood v Capita make no reference to the contra proferentem rule. In Impact Funding Solutions Limited (Respondent) v AIG Europe Insurance Ltd (formerly known as Chartis Insurance (UK) Ltd) [2016] UKSC 57, [2017] AC 73, the Supreme Court had to consider the interpretation of a solicitors’ professional indemnity insurance policy which contained an exclusion for loss associated with “breach by any insured of terms of any contract or arrangement for the supply to, or use by, any insured of goods or services in the course of providing legal services”. The issue was whether this exclusion applied to the lending arrangement between Impact Funding and the solicitors. The court held that the exclusion clause applied. After stating the general principles of contractual construction, Lord Hodge (with whom Lord Mance, Lord Sumption and Lord Toulson agreed) made the following observations about interpretation contra proferentem:

“6.

… As I see no ambiguity in the way that the Policy defined its cover and as the exclusion clause reflected what The Law Society of England and Wales as the regulator of the solicitors’ profession had authorised as a limitation of professional indemnity cover, I see no role in this case for the doctrine of interpretation contra proferentem. …

7.

The extent of AIG’s liability is a matter of contract and is ascertained by reading together the statement of cover and the exclusions in the Policy. An exclusion clause must be read in the context of the contract of insurance as a whole. It must be construed in a manner which is consistent with and not repugnant to the purpose of the insurance contract. There may be circumstances in which in order to achieve that end, the court may construe the exclusions in an insurance contract narrowly. The judgment of Carnwath LJ in Tektrol Ltd (formerly Atto Power Controls Ltd) v International Insurance Co of Hanover Ltd [2006] 1 All ER (Comm) 780, to which counsel for Impact referred, is an example of that approach. But the general doctrine, to which counsel also referred, that exemption clauses should be construed narrowly, has no application to the relevant exclusion in this Policy. An exemption clause, to which that doctrine applies, excludes or limits a legal liability which arises by operation of law, such as liability for negligence or liability in contract arising by implication of law: Photo Production Ltd v Securicor Transport Ltd [1980] AC 827, 850 per Lord Diplock. The relevant exclusion clause in this Policy is not of that nature. The extent of the cover in the Policy is therefore ascertained by construction of all its relevant terms without recourse to a doctrine relating to exemption clauses.”

Lord Toulson (with whom Lord Mance, Lord Sumption and Lord Hodge agreed) said at [35]:

“The fact that a provision in a contract is expressed as an exception does not necessarily mean that it should be approached with a pre-disposition to construe it narrowly. Like any other provision in a contract, words of exception or exemption must be read in the context of the contract as a whole and with due regard for its purpose. As a matter of general principle, it is well established that that if one party, otherwise liable, wishes to exclude or limit his liability to the other party, he must do so in clear words; and that the contract should be given the meaning it would convey to a reasonable person having all the background knowledge which is reasonably available to the person or class of persons to whom the document is addressed. (See, among many authorities, Dairy Containers Ltd v Tasman Orient Line CV [2005] 1 WLR 215, para 12, per Lord Bingham.) This applies not only where the words of exception remove a remedy for breach, but where they seek to prevent a liability from arising by removing, through a subsidiary provision, part of the benefit which it appears to have been the purpose of the contract to provide. The vice of a clause of that kind is that it can have a propensity to mislead, unless its language is sufficiently plain. All that said, words of exception may be simply a way of delineating the scope of the primary obligation.”

The position was correctly summarised by Mr Peter MacDonald Eggers QC sitting as Deputy High Court Judge in Crowden v QBE Insurance (Europe) Ltd [2017] EWHC 2597 (Comm); [2018] Lloyd’s Rep IR 83, at [65], where, after considering Impact Funding Solutions, he said:

“65.

In my judgment, applying this approach, the Court must adopt an approach to the interpretation of insurance exclusions which is sensitive to their purpose and place in the insurance contract. The Court should not adopt principles of construction which are appropriate to exemption clauses - i.e. provisions which are designed to relieve a party otherwise liable for breach of contract or in tort of that liability - to the interpretation of insurance exclusions, because insurance exclusions are designed to define the scope of cover which the insurance policy is intended to afford. To this end, the Court should not automatically apply a contra proferentem approach to construction. That said, there may be occasions, where there is a genuine ambiguity in the meaning of the provision, and the effect of one of those constructions is to exclude all or most of the insurance cover which was intended to be provided. In that event, the Court would be entitled to opt for the narrower construction. This result may be achieved not only by the applicable of the contra proferentem approach, but also the approach adopted by Lord Clarke, JSC in Rainy Sky SA v Kookmin Bank [2011] UKSC 50; [2011] 1 WLR 2900, that in the case of ambiguity, the Court may opt for the more commercially sensible construction, at paragraph 21: “If there are two possible constructions, the court is entitled to prefer the construction which is consistent with business common sense and to reject the other”. That said, as Lord Clarke, JSC also said, at paragraph 23 of his judgment: “Where the parties have used unambiguous language, the court must apply it”. This would, however, be subject to considerations of absurdity or where something plainly has gone wrong with the language of the contract.”

When interpreting the provisions of a contract, the court can only take into account facts or circumstances which existed at the time that the contract was made, and which were known or “reasonably available” to both parties: Arnold v Britton at [21]; Investors Compensation Scheme at 912-913.

The material which is reasonably available to the parties includes the relevant legal background (see Lewison on the Interpretation of Contracts (6^th ed) at §4.06), and the parties are “taken to have contracted against a background which includes the previous decisions upon the construction of similar contracts” (Toomey v Eagle Star Insurance Co Ltd [1994] 1 Lloyd’s Rep 516 at 520 (Hobhouse LJ)). It is also part of the factual matrix known or taken to be known to both parties that both statute law and the common law develop over time, so they can be taken as having agreed that their agreement be interpreted in the light of the general law from time to time: Lymington Marina Ltd v MacNamara [2007] EWCA Civ 151 at [33] (Arden LJ).

An example in the context of insurance is Sunport Shipping Ltd v Tryg-Baltica International (UK) Ltd (The Kleovoulos of Rhodes) [2003] EWCA Civ 12, [2003] 1 Lloyd’s Rep 138. In that case, the claimants’ vessel was insured under a policy of marine insurance which contained a standard Institute clause excluding cover for arrest or detainment “by reason of infringement of any customs or trading regulations”. The parties were at issue in particular about the breadth of the expression “customs regulations” and therefore whether the clause in question excluded cover in relation to an infringement of laws prohibiting the import of drugs (at [15]). In his judgment, Clarke LJ (with whom Scott Baker LJ and Peter Gibson LJ agreed) referred to the principle that “where the relevant expression has been given a settled meaning by the Courts the Court should so construe it in the same context in the future” (at [25]). He went on as follows at [28]:

“Where a contract has been professionally drawn, as in the case of the Institute Clauses, the draftsman is certain to have in mind decisions of the Courts on earlier editions of the clause. Such decisions are part of the context or background circumstances against which the particular contract falls to be construed. If the draftsman chooses to adopt the same words as previously construed by the Courts, it seems to me to be likely that, other things being equal, he intends that the words should continue to have the same meaning.”

Clarke LJ held that part of the relevant context against which the exclusion should be understood was an earlier decision of the Court of Appeal (Panamanian Oriental Steamship Corporation v Wright (The Anita) [1971] 1 Lloyd's Rep 487) which addressed the meaning of the expression “customs regulations”. Against that background, it was clear that the phrase had a broad meaning and therefore the exclusion did cover laws prohibiting the import of goods (The Kleovoulos of Rhodes at [47], [56]).

The approach exemplified in The Kleovoulos of Rhodes is one based on the clause having a settled meaning, by reason of its being used against a background of longstanding and clear authority. There was argument in the present case as to whether the decision in Orient-Express Hotels Ltd v Assicurazioni Generali SpA [2010] EWHC

Disease clauses

The Court is asked to construe a number of wordings which contain non-damage “extensions” to the “standard” Business Interruption (“BI”) cover provided by the relevant insurers. That “standard” cover is contingent on the occurrence of physical or material damage to the insured premises. There is no dispute before the Court about whether there is cover under such “standard” BI cover. In this section, we will consider the proper construction of a series of provisions to which the FCA refers as “disease clauses”. Provided that this way of referring to them is not permitted to influence questions of the ambit of the coverage provided, it is a helpful way of distinguishing a group of coverage provisions which cover prevention of access and similar perils, which we analyse later in this judgment. The clauses with which we will deal in this section are: (i) the “Defective Sanitation, Notifiable Human Disease, Murder or Suicide” cover in Argenta 1; (ii) the “Notifiable disease, vermin, defective sanitary arrangements, murder and suicide” covers in MSA 1 and MSA 2; (iii) the “Murder, suicide or disease” cover in QBE 1 and the “Infectious disease, murder or suicide, food or drink poisoning” covers in QBE 2 and QBE 3; and (iv) the “Infectious Diseases” cover in RSA 3 and the “Notifiable Diseases and Other Incidents” cover in RSA 4.

It will be necessary to consider the terms of each of these policies separately as it is, of course, impossible to determine questions of policy coverage in the abstract. What these policies share, however, are provisions which, in broad terms, provide coverage in respect of business interruption in consequence of or following or arising from the occurrence of a notifiable disease within a specified radius of the insured premises. In relation to each, there arises the question of whether there is cover in respect of a pandemic where it cannot be said that the key matters which led to business interruption, and in particular the governmental measures, would not have happened even without the occurrence of COVID-19 within the specified radius, as a result of its occurrence or feared occurrence elsewhere.

We propose to commence by considering what has been referred to as RSA 3, because it raises a number of different issues which are echoed in the other covers which will be considered in this section. This is a Commercial Combined Policy underwritten by Eaton Gate Commercial, a Managing General Underwriter on behalf of RSA. It was taken out by the owners of a variety of businesses, including building contractors, landscape gardeners and manufacturers and wholesalers of electronics, fabrics and metal goods. The policy has nine sections: Property Damage, Business Interruption, Money and Assault, Goods in Transit, Employers’ Liability, Public and Products Liability, Contract Works, All Risks and Commercial Legal Expenses. Which sections

In Section 2, which covers BI, the basic cover is in respect of loss resulting from interruption of or interference with the relevant business “in consequence of loss or destruction of or Damage insured under Section 1 [the Property Damage section]”. The insuring clause is in the following terms (here, as elsewhere, the text of the policy is reproduced though the formatting may not be identical to that which appears in the policy; except where it matters, no significance is attached to the formatting of the wordings):

“Cover

In the event of Business Interruption We will pay to You in respect of each item in the Schedule the amount of loss resulting from such interruption or interference provided that at the time of the happening of the loss destruction or Damage there is an insurance in force covering Your interest in the Property at the Premises against such loss destruction or Damage and that:

a)

payment shall have been made or liability admitted therefore

[sic]; or

b)

payment would have been made or liability admitted therefore [sic] but for the operation of a proviso in such insurance excluding liability for losses below a specified amount.”

The policy provides, however, for a series of “extensions” to Section 2. These include the following:

Of most significance for present purposes amongst the extensions to Section 2 is Extension vii “Infectious Diseases”. It provides as follows:

“We shall indemnify You in respect of interruption or interference with the Business during the Indemnity Period following:

a)

any

i.

occurrence of a Notifiable Disease (as defined below) at the Premises or attributable to food or drink supplied from the Premises;

ii.

discovery of an organism at the Premises likely to result in the occurrence of a Notifiable Disease;

iii.

occurrence of a Notifiable Disease within a radius of 25 miles of the Premises;

b)

the discovery of vermin or pests at the Premises which causes restrictions on the use of the Premises on the order or advice of the competent local authority;

c)

any accident causing defects in the drains or other sanitary arrangements at the Premises which causes restrictions on the use of the Premises on the order or advice of the competent local authority; or

d)

any occurrence of murder or suicide at the Premises.

Additional Definition in respect of Notifiable Diseases

1.

Notifiable Disease shall mean illness sustained by any person resulting from:

i.

food or drink poisoning; or

ii.

any human infectious or human contagious disease excluding Acquired Immune Deficiency Syndrome (AIDS) or an AIDS related condition an outbreak of which the competent local authority has stipulated shall be notified to them

2.

For the purposes of this clause:

Indemnity Period shall mean the period during which the results of the Business shall be affected in consequence of the occurrence discovery or accident beginning:

i.

in the case of a) and d) above with the date of the occurrence or discovery; or

ii.

in the case of b) and c) above the date from which the restrictions on the Premises applied; and ending not later than the Maximum Indemnity Period thereafter shown below.

Premises shall mean only those locations stated in the Premises definition. In the event that the section includes an extension which deems loss destruction or Damage at other locations to be an Incident such extension shall not apply to this clause.

3.

We shall not be liable under this clause for any costs incurred in the cleaning repair replacement recall or checking of Property

4.

We shall only be liable for the loss arising at those Premises which are directly affected by the occurrence discovery or accident Maximum Indemnity Period shall mean 3 months”

A series of exclusions applicable to Section 2 are set out on pages 41-43 of 93 of the policy wording. None is said to be applicable. RSA does, however, rely on one of the General Exclusions, namely General Exclusion L, which appears on page 93 of 93 and is in the following terms:

“Applicable to all sections other than Section 5 – Employers’

Liability and Section 6 – Public Liability

Contamination or Pollution Clause

a)

The insurance by this Policy does not cover any loss or Damage due to contamination pollution soot deposition impairment with dust chemical precipitation adulteration poisoning impurity epidemic and disease or due to any limitation or prevention of the use of objects because of hazards to health.

b)

This exclusion does not apply if such loss or Damage arises out of one or more of the following Perils:

•

Fire, Lightning, Explosion, Impact of Aircraft

•

Vehicle Impact Sonic Boom

•

Accidental Escape of Water from any tank apparatus or pipe Riot, Civil Commotion, Malicious Damage

•

Storm, Hail Flood Inundation Earthquake

•

Landslide Subsidence Pressure of Snow, Avalanche Volcanic Eruption

a)

(bis) If a Peril not excluded from this Policy arises directly from Pollution and/or Contamination any loss or Damage arising directly from that Peril shall be covered.

b)

(bis) All other terms and conditions of this Policy shall be unaltered and especially the exclusions shall not be superseded by this clause.”

The Policy Definitions include the following:

“Business

Activities directly connected with the Business shown in the

Schedule and no other for the purposes of this Policy …

…

Damage

Material loss destruction or Damage.

…

Premises

The part of the Premises at the address or addresses specified in the Schedule which You occupy for the purposes of the Business and otherwise occupied as offices or private dwellings unless otherwise agreed with Us.”

The sectional definitions for Section 2 include:

“Business Interruption

Business Interruption shall mean loss resulting from interruption of or interference with the Business carried on by You at the Premises in consequence of loss or destruction of or Damage insured under Section 1 to Property used by You at the Premises for the purpose of the Business

…

Incident

a)

Loss or destruction of or Damage to Property used by You at the Premises for the purpose of the Business; or

b)

Loss destruction of or Damage to Your books of account or other Business books or records at the Premises in respect of Book Debts.”

The Basis of Claims Settlement provisions for the BI section provide in part:

“Section 2 – Gross Profit/Estimated Gross Profit (if shown as operative in the Schedule)

The insurance is limited to loss of Gross Profit due to:

a)

reduction in Turnover; and

b)

increase in cost of working; and the amount payable as indemnity shall be:

a)

in respect of a reduction in Turnover:

b)

the sum produced by applying the Rate of Gross Profit to the amount by which the Turnover during the Indemnity Period shall fall short of the Standard Turnover in consequence of the Incident …

Section 2 – Gross Revenue/Estimated Gross Revenue

(if shown as operative in the Schedule) The insurance is limited to

a)

loss of Gross Revenue;

…

and the amount payable as indemnity shall be:

a)

in respect of loss of Gross Revenue: the amount by which the Gross Revenue during the Indemnity Period shall fall short of the Standard Gross Revenue in consequence of the Incident …”

After the sectional definitions for Section 2 there appears the following “special provision applicable to this section” (referred to as the “trends clause” in the discussion below):

“Under Rate of Gross Profit, Annual Turnover, Standard Turnover, Annual Rent receivable, Standard Rent, Receivable Annual Gross Revenue and Standard Gross Revenue adjustments shall be made as may be necessary to provide for the trend of the Business and for variations in or other circumstances affecting the Business either before or after the Incident or which would have affected the Business had the Incident not occurred so that the figures thus adjusted shall represent as nearly as may be reasonably practicable the results which but for the Incident would have been obtained during the relative period after the Incident.”

The FCA’s case is that there was cover under RSA 3, Extension vii (a)(iii) (“interruption of or interference with the Business during the Indemnity Period following … any occurrence of a Notifiable Disease within a radius of 25 miles of the Premises”), on the following basis. There was a Notifiable Disease in all parts of the UK by 6 March 2020. There was an occurrence of that Notifiable Disease within 25 miles of an insured’s premises when a person or persons with COVID-19 was within 25 miles of those premises. There was interruption of or interference with the business from 16 March 2020, or from a later date to be determined by the Court, as a result of the government’s instructions and/or announcements as to social distancing, selfisolation, lockdown and restricted travel and activities, or alternatively, in cases where businesses were ordered to close, from 23 March 2020. Any losses as insured were sufficiently causally connected with the interruption or interference and the interruption or interference “followed” the occurrence of COVID-19 if they would not have occurred had there been no COVID-19 outbreak or intervention by the government. In relation to this last aspect, the FCA contended that the word “following” deliberately connotes an event “which is part of the factual background and represents a looser causal connection than ‘resulting from’ and similar”.

For its part, RSA contended that there was no cover under RSA 3 on the basis advanced by the FCA. As elaborated by Mr David Turner QC, RSA made three groups of points, as follows. In the first place, he submitted that the cover was only against business interruption or interference proximately caused by a local outbreak of a Notifiable Disease (i.e. one within 25 miles of the premises). That, he submitted, imported a requirement of “but for” causation. The use of the word “following” did not alter this. Had there been no occurrence of COVID-19 within a 25 mile radius of insured premises, the insureds’ businesses would still have suffered from a general reduction of demand after 1 March 2020, and would also have suffered from the impact of the government’s social distancing measures from 16 March 2020, and from any closure measures, because those would have been introduced anyway by reason of the occurrence or feared occurrence of the disease in areas other than the 25 mile radius. Secondly, RSA submitted that cover for an epidemic such as COVID-19 was, in any event, excluded by the terms of General Exclusion L. Thirdly, RSA relied on the socalled “trends clause” in the BI section of the policy as limiting its liability under the relevant extension to any loss which would have been sustained had the insured peril not occurred, which RSA contended meant had the local occurrence not occurred. This was said to provide an alternative route to the same result as reached by reason of the first two arguments. We will consider each of these points in turn.

In relation to the first argument it is helpful to deal at the outset with two preliminary points. The first is that it is common ground that COVID-19 was a Notifiable Disease for the purposes of Extension vii (a)(iii) in all parts of the United Kingdom by 6 March 2020. The second is as to what constituted an “occurrence” of COVID-19, and in particular what constituted such an “occurrence” within the 25 mile radius provided for in Extension vii. The FCA’s case is that there will have been an occurrence of the disease whenever or wherever a person had contracted COVID-19 such that it was diagnosable, whether or not it had been verified by diagnosis, and whether it was symptomatic. RSA’s pleaded case is that nothing less than an actual diagnosis of COVID-19 would be sufficient to establish any relevant “occurrence”. We consider that there will have been an “occurrence” of COVID-19 within an area when at least one person who was infected with COVID-19 was in the relevant area. We do not consider that it is necessary for there to have been an “occurrence” of the disease that the case should have been diagnosed. The definition of Notifiable Disease is in relevant part “illness sustained by any person resulting from … any human infectious or human contagious disease…” Such a Disease thus “occurs” when the illness is “sustained” by a person, which we consider means, in simple terms, that they are suffering from it, not that they have been diagnosed with it. This fits in with the other parts of the Extension. For example, in sub-clause a(i) of Extension vii, if there were cases of food poisoning at the premises, which led to business interruption, but it took some time for it to be diagnosed that this was due to a Notifiable Disease, we would consider that the Notifiable Disease had “occurred” when there were the first cases of food poisoning, and that the “occurrence” was not postponed until there was diagnosis.

The next issue which needs to be addressed, and on which there was extensive argument, is as to what is required by the causal connectors within Extension vii, and in particular as to whether interference or interruption with the business was part of the loss or part of the insured peril, and as to the meaning and effect of the word “following”. Taking the first of those particular points, for our part we consider that the “peril insured against” is the composite peril of interruption or interference with the Business during the Indemnity Period following one of (a) to (d). Accordingly the requirement of proximate causation, codified in marine insurance by section 55(1) of the Marine Insurance Act 1906, but which is a reflection of the common law, including as to non-marine insurance, is between the loss claimed by the insured and such “interruption or interference with the Business”. The requirement that the business interruption or interference should “follow” one of (a) to (d) is, as we see it, a requirement within the composite peril insured.

As to what is intended by the requirement that the business interruption or interference should “follow” one of (a) to (d), we consider that this imports more than merely a temporal relationship. We consider that the FCA is correct to say that it involves a requirement that one of (a) to (d) should have a causal connection to the business interruption, but not necessarily one of proximate causation. While it is the case that courts tend to avoid drawing nice distinctions between causative phrases in an insurance context, and that it is often appropriate to infer that the intention of the parties was to refer to the established test of proximate causation, it does not appear that it is appropriate to construe “following” here as having that meaning. As we have said, we do not consider that it is a link between the loss and the peril, which is the central case where it is generally to be inferred that the parties intended a test of proximate causation. In any event, it appears to us that, in context, it was used to denote a looser form of link than that of proximate causation. Its natural meaning does not itself import such a link, and it makes good sense that a word implying a looser link should be used in recognition of the fact that, of the matters referred to within Extension vii, (a) and (d) would not of themselves directly cause interruption to or interference with the business, but would in almost every case have such an effect only via the reaction of the authorities and/or of the public. In (b) and (c) that part of the chain is specified in the clause, but not in (a) and (d), and thus the use of the word “following” is entirely understandable.

We were not persuaded by RSA’s arguments to the effect that the word “following” must denote proximate causation by reason of other provisions of Extension vii. There were two such arguments. One focused on the terms of the definition of “Indemnity Period” given in the extension, and in particular on the words “shall be affected in consequence of the occurrence discovery or accident”. The phrase “in consequence of”, RSA submitted, must be read as requiring proximate causation. In our view, however, the phrase is used simply to refer to the requirement that the interruption should “follow” one of (a) to (d) which is specified in the main body of the extension. “Following” on any view, and as the FCA accepts, imports some causal connection and as we see it, the phrase “in consequence of” is intending to refer to the same connection. It is significant that, within the definition of “Indemnity Period”, what is envisaged is that the “occurrence discovery or accident” shall have business interruption or interference as a “consequence”. But, as we have already said, the “occurrences” in (a) and (d) would not have a direct effect on the business. “In consequence of” is accordingly intended to embrace, at least, indirect causation.

The second argument is based on the terms of Clause 4 under the heading “Additional Definition in respect of Notifiable Disease” in Extension vii, and in particular the words “We shall only be liable for the loss arising at those Premises which are directly affected by the occurrence discovery or accident…” We consider that this provision is aimed at ensuring that there is no cover for the effects on the business which do not arise from those premises which are impacted by the contingency insured (whether it be (a), (b), (c) or (d) within Extension vii). Accordingly, for example, this provision makes clear that, if vermin or pests were discovered at an insured’s premises A and as a result the competent local authority shut down the insured’s premises B as well as a result of concerns about the insured’s safety standards, the business interruption caused by the shutting of premises B would not be covered. In our view, however, it is not intended to narrow the cover provided under (a) to cases where the relevant occurrence or discovery occurred at premises A or had an effect upon premises A without any intermediate cause. That would be inconsistent with the grant of cover in respect of the business interruption following any occurrence of a notifiable disease within 25 miles of the Premises, which clearly envisages that the occurrence of the disease will be at a distance of up to 25 miles from the Premises in question. A disease occurring at such a distance would only be likely to have an effect on the business at the Premises indirectly, namely via the reaction of the authorities and/or the public.

Putting on one side for the moment its arguments as to General Exclusion L and the “trends clause”, RSA did not develop its arguments as to why there was no cover if it is wrong as to its submissions as to the word “following” importing a requirement of proximate causation. Nonetheless, given that the FCA accepts that even the term “following” imports a causal link, we regard it as necessary to consider how that can be said to have been fulfilled. Furthermore, the FCA made it clear that, while its primary case was that Extension vii does not import a requirement of proximate causation, it contended that it would not make any difference if it did: that there would be cover under the clause even if it needed to be shown that a proximate cause of the business interruption was the occurrence of a Notifiable Disease within a radius of 25 miles of the Premises. We will consider that argument here as well, in case the matter goes further and because it is a theme relevant to other “disease clauses” which we have to consider.

The way in which Mr Colin Edelman QC, on behalf of the FCA, put this case was that in each case when and where there was the occurrence of a Notifiable Disease within the 25 mile radius, that occurrence was an effective cause of the subsequent government measures and interference with the insured’s business. The government’s decisions were based on the fact that there were such occurrences in a very large number of places. The decisions were accordingly based on all of those occurrences. He said that the way in which this should be analysed is either that there was one indivisible cause, namely the disease, of which all the outbreaks formed part, or that there were many different concurrent effective causes, none of which is excluded. That analysis, Mr Edelman QC submitted, meant that the looser “following” test was certainly satisfied, but also that there was proximate causation, if that needed to be shown.

While much of the argument was understandably put in terms of the nature of the causal requirements, we consider that what underlies the dispute in relation to causative requirements is a difference as to the nature of the peril insured, and that this depends on a proper construction of the relevant terms of Extension vii. Once that question of construction is answered, it seems to us that the issues of causation will also largely have been answered, and in particular it will have been established which matters can be said to be separate, non-insured causes which could be seen as distinct from the insured peril.

RSA’s contention is that the insured peril is the effect of a local occurrence of a Notifiable Disease. Its case is that if there is a local outbreak of a disease occurring more widely, then it is only the effects of the disease occurring locally, and only insofar as they can be distinguished, which are covered. RSA submits that this is the purpose and effect of the 25 mile radius provision, and that the FCA’s case reduces the requirement that there should have been an occurrence of the disease within 25 miles to a senseless, or at least arbitrary, “tick box” condition for cover.

This is undoubtedly a significant argument, but it is one which we are unable to accept because we have concluded that it does not withstand detailed consideration of the nature of a cover in relation to Notifiable Disease in the terms of that provided for in Extension vii. Two matters are fundamental to this conclusion. The first is the language of the particular clause. Extension vii (a) is not expressly confined to cases where the interruption has resulted only from the instance(s) of a Notifiable Disease within the 25 mile radius, as opposed to other instances elsewhere. Nor in our view does the language used in this clause implicitly have that effect. Instead, the clause can and should properly be read as meaning that there is cover for the business interruption consequences of a Notifiable Disease which has occurred, i.e. of which there has been at least one instance, within the specified radius, from the time of that occurrence. The wording of the clause, in other words, indicates that the essence of the fortuity covered is the Notifiable Disease, which has come near, rather than specific local occurrences of the disease.

The second matter is the implications of the fact that the cover relates to occurrences of a Notifiable Disease. By reason of the terms of 1(ii) of the Additional Definition of Notifiable Disease, Notifiable Diseases include those which are required to be officially notified. The relevant scheme for notification, as mentioned above, is contained in the 1984 Act and the 2010 Regulations. Under the 2010 Regulations, registered medical practitioners have a duty to notify proper officers of the relevant local authority, amongst other things, of grounds for suspecting that a patient has a notifiable disease. Under the same Regulations, the relevant local authority must disclose the fact of a notification of a notifiable disease to, amongst others, PHE, and to the proper officer of the local authority in whose area the patient usually resides, if different. At the time of the conclusion of the contracts of insurance in question, there were some 31 diseases specified in Schedule 1 to the 2010 Regulations as notifiable, including cholera, plague, typhus, yellow fever and SARS. Under RSA 3, furthermore, the cover in Extension vii is not limited to Notifiable Diseases which have been included on Schedule 1 at the outset of the insurance, but extends to others which may be added thereto, as COVID19 has been.

While there is clearly a spectrum of diseases within the category of Notifiable Diseases, it includes diseases which are capable of widespread dissemination, such as SARS (Severe Acute Respiratory Syndrome), which is a viral respiratory illness caused by the SARS-associated coronavirus for which there is no vaccine. It is in the nature of human infectious and contagious diseases that they may spread in highly complicated, often difficult to predict, and what might be described as “fluid”, patterns. Furthermore, the list of diseases includes some which might attract a response from authorities which are not merely local authorities, and which is not a purely local response. The requirement under the Regulations of notifications to PHE, and to other local authorities facilitates such wider responses. Moreover, in terms of Extension vii, the fact that it is envisaged that the occurrence of a notifiable disease up to 25 miles away might be followed by interruption of business at the insured’s premises demonstrates, in our view, that the parties must have contemplated that there might be relevant actions of public authorities which affect a wide area. They must also have contemplated that the authorities might take action in relation to the outbreak of a notifiable disease as a whole, and not to particular parts of an outbreak, and would be most unlikely to take action which had any regard to whether cases fell within or outside a line 25 miles away from any particular insured premises.

In light of these matters, if RSA is correct as to the meaning to be ascribed to the clause, the parties would have been agreeing to the production of highly anomalous results. By way of example, if 20 people contracted a Notifiable Disease in a town 24 miles away from the premises and the authorities decided as a result to take “locking down” or other action which affected the insured’s premises, there would on RSA’s case (as we understood it) be cover, even though the aim of the public authorities was in large part to prevent the disease spreading elsewhere including outside the area of the 25 mile radius. On the other hand, if 20 people in a town 26 miles from the premises contracted the Notifiable Disease and the authorities decided to act by imposing a lockdown or other measures, there would be no effective cover for the resulting interruption or interference with the business, notwithstanding that some of those 20 people might have subsequently moved into, or infected people within, the 25 mile radius, and notwithstanding that a part of the motivation of the authorities in imposing the measures was to prevent or slow the spread of the disease within the 25 mile radius.

Equally, and on a more general level, it is not difficult to conceive of a disease which spread rather more slowly than COVID-19, which triggered a series of local lockdowns or other public health measures, which ultimately covered all or large parts of the country. On RSA’s case, if the local measures were caused by the occurrence of the disease within the 25 mile radius, then there would be cover for their effects. But if the disease developed and spread more quickly, so that the response was national, and simultaneous, then there would be no effective cover in any area, because the response was not taken specifically in relation to any particular area.

Those considerations support a construction of the cover in Extension vii (a)(iii) such that it is not confined to the effects only of the local occurrence of a Notifiable Disease. The construction we favour avoids the result that there would be no effective cover if the local occurrence were a part of a wider outbreak, and where, precisely because of the wider outbreak, it would be difficult or impossible to show that the local occurrence(s) made a difference to the response of the authorities and/or public. We do not consider that a reasonable person, equipped inter alia with knowledge of the 1984 Act and the 2010 Regulations, would have understood the parties in using the words they did to mean that or intend such results.

The point which RSA, and other insurers, particularly pressed us with was that the FCA’s interpretation of the cover provided by Extension vii (and other “disease clauses”) gave no sensible meaning or effect to the 25 mile radius stipulation at all. They argued that it became, on the FCA’s case, just a condition which had to be fulfilled, so that, if there were just one case of a disease within the 25 miles, then there would be cover for the effects of an epidemic which had no other link to the locality, but if no case came within 25 miles, then there would not be. This was described by RSA, and other insurers, as a matter of “happenstance” and productive of arbitrary results.

In our view, the FCA is correct to say that the 25 mile radius provision, interpreted as it submits it should be, makes sense. It has the effect that diseases which make no local appearance cannot lead to there being cover. While it is possible to think of anomalous cases, where it is a matter of chance whether an infected person came within the 25 mile radius or not, it appeared to us that any such anomalies were considerably less significant that those inherent in RSA’s interpretation, some of which we have indicated already.

If we are correct in our view as to the nature of the cover provided in Extension vii (a)(iii), then the issues as to causation largely answer themselves. If, properly construed, there is cover for the effects of a disease which may occur both within and outside the specified radius, and which may trigger a response of the authorities and the public to the outbreak as a whole, then it would be inconsistent with the nature of the cover to regard the occurrence of the disease outside the radius, or the response of the authorities or the public to that occurrence of the disease, as being alternative, uncovered, causes of the business interruption which could be relied on as supporting an argument that there would have been the same business interruption in the absence of the insured peril.

If, as we consider to be the case with RSA 3, what is required by the word “following” is a looser causal relation than proximate cause, we would regard that as being clearly satisfied by the occurrence of a case of the disease within the radius if that occurrence was part of a wider picture which dictated the response of the authorities and the public which itself led to the business interruption or interference. Even if the word “following” imports the requirement of proximate causation we would consider that, given the nature of the cover as we consider it to be, this is to be regarded as satisfied in a case in which there is a national response to the widespread outbreak of a disease. In such a case we consider that the right way to analyse the matter is that the proximate cause of the business interruption is the Notifiable Disease of which the individual outbreaks form indivisible parts.

Alternatively, although we regard this as being less satisfactory, each of the individual occurrences was a separate but effective cause. On this analysis they were all effective because the authorities acted on a national level, on the basis of the information about all the occurrences of COVID-19, and it is artificial to say that only some of those which had occurred by any given date were effective causes of the action taken at that date; and still more artificial to say that because the action was taken in response to all the cases, it could not be regarded as taken in response to any particular cases. As Mr Edelman QC submitted, there is material in the Agreed Facts which provides a sufficient basis for this analysis. He pointed to the information which the government was acting upon, and a number of SAGE minutes, which show that the government response was the reaction to information about all the cases in the country, and that the response was decided to be national because the outbreak was so widespread. As Mr Edelman QC pointed out, the Secretary of State for Health and Social Care, Mr Hancock, on 28 April 2020 stated that thought had been given to imposing measures first on London and the Midlands, but it had been decided that “we are really in this together”, and that “the shape of the curve … has been very similar across the whole country”. Given this, it appears to us that it is not unrealistic to say that all the cases were equal causes of the imposition of national measures.

The result is that, subject to a consideration of the second and third points raised by RSA, there is cover under RSA 3 for any business interruption which an insured can show resulted from COVID-19, including by reason of the actions, measures and advice of the government, and the reaction of the public in response to the disease, from the date when the disease occurred in the relevant 25 mile radius.

RSA’s second argument is that cover for the effects of an epidemic such as COVID-19 was excluded by a specific exclusion in the policy, namely General Exclusion L. We have already set out the terms of this exclusion, and where it appears in the policy.

This exclusion has the hallmarks of one which has been included without detailed consideration of the extent to which its terms might, if applied literally, cut down specific covers provided in the insurance. Thus, the exclusion seems to exclude “loss or Damage” due to “poisoning”. However, food poisoning is one of the notifiable diseases which is included in Schedule 1 to the 2010 Regulations, and is also the most obvious Notifiable Disease which might be “attributable to food or drink supplied from the Premises” within Extension vii (a)(i). Indeed, in the “Additional Definition in respect of Notifiable Diseases” within Extension vii, Notifiable Disease is defined to include an illness sustained by any person resulting from “food or drink poisoning”. Similarly, General Exclusion L seems to exclude “loss or Damage” due to “disease”. This is notwithstanding that Extension vii quite clearly provides express cover for business interruption following Notifiable Diseases. We have no doubt that a reasonable person would not understand the insurance to be expressly giving cover with one hand and taking it away by the other in the form of the list of matters referred to in General Exclusion L.

While RSA seeks to rely on the appearance of the word “epidemic” within Exclusion L, it appears to us that this reliance carries no conviction. “Epidemic” is not alone in the list but, as we have said, is in the company of “poisoning” and “disease”. As we have said, we find it impossible to consider that loss as a result of these matters insofar as expressly covered under Extension vii, was intended to be excluded by this General Exclusion. Furthermore, the list of diseases which, even at the time of the conclusion of the contracts of insurance, were required to be notified to the competent local authority included some, such as SARS, which might undoubtedly give rise to an epidemic; and furthermore Notifiable Diseases as defined in Extension vii were not a closed category, and clearly a newly occurring disease giving rise to an epidemic would be likely to lead to a requirement for notification, as indeed happened with COVID-19. Accordingly, there is cover under Extension vii for certain losses following an epidemic, and for the same reasons as apply to “poisoning” and “disease” we find it impossible to accept that the general exclusion was intended to withdraw such express cover.

We consider that the correct construction of the policy is that sub-clause L(b) (bis) must be understood as meaning that the terms of the exclusion were not intended to override express grants of cover, or at least was not intended to apply to the disease clauses in Extension vii. It seems likely that L(b)(bis) was included precisely to cater for the possibility that, without it, General Exclusion L might not fit well with other terms of the policy in which it was being included. L(b)(bis) is not confined to preserving the effect of other exclusions in the policy, because the first part of the clause applies to “all other terms and conditions of the Policy”, and the second part of the clause applies to a particular class, namely exclusions, which are “especially”, which connotes not exclusively, not to be superseded. It is therefore wide enough, in our view, to apply to cases such as the present, in order to avoid the absurdity, and potential unfairness, of an express grant of cover being eliminated by a General Exclusion dealing with a range of matters most of which are irrelevant to the cover involved.

Finally, under this heading we should say, while we do not consider it necessary to resort to it, this would be one of the few cases in which it would be appropriate to apply a principle of contra proferentem.

RSA’s third main argument relates to the “trends clause”. It contends that the “trends clause” included after the “Definitions for Section 2” is applicable to the calculation of an indemnity under Extension vii. This is notwithstanding that the “trends clause” is couched in terms of “circumstances” affecting the Business either before or after “the Incident”, and “Incident” is defined as loss, destruction or damage to physical property. RSA contends that this should, in effect, be manipulated, so that it is applicable to nondamage perils insured under the extensions to Section 2. The FCA contends, by contrast, that the indemnity under Extension vii is not subject to any of the contractual quantification machinery, and the “trends clause” is relevant only if that machinery is applicable.

In our view, though the policy is not well drafted, it should be construed as meaning that the contractual quantification provisions are applicable for the purposes of the calculation of the indemnity under Extension vii. It would, in our view, be commercially surprising if the parties intended that the provisions relating to quantification of business interruption claims arising out of damage should not apply, with appropriate adjustments, to claims in respect of non-damage perils and the parties should be left to debate the correct approach to the quantification of such claims notwithstanding their agreement on the principles relating to claims for business interruption following damage. If this is right, then there is also no good reason why the “trends clause” should not apply to such non-damage claims, though it would necessarily have to be manipulated to make it applicable. That manipulation would involve reading the clause as saying “… adjustments shall be made as may be necessary to provide for the trend of the Business and for variations in or other circumstances affecting the Business either before or after the occurrence of the insured peril or which would have affected the Business had the insured peril not occurred so that the figures thus adjusted shall represent as nearly as may be reasonably practicable the results which but for the insured peril would have been obtained during the relative period after the occurrence of the insured peril”.

There are two important related preliminary points about the “trends clause” in this wording, which are equally applicable to all the trends clauses and provisions which we are considering. First, it is in the quantification machinery for a claim, so that it is not part of the delineation of cover, but part of the machinery for calculating the business interruption loss on the basis that there is a qualifying insured peril. Where the policyholder has therefore prima facie established a loss caused by an insured peril, it would seem contrary to principle, unless the policy wording so requires, for that loss to be limited by the inclusion of any part of the insured peril in the assessment of what the position would have been if the insured peril had not occurred. Second, subject to the particular wording providing for something different, the object of the quantification machinery (including any trends clause or provision) in the policy wording is to put the insured in the same position as it would have been in if the insured peril had not occurred.

Therefore, in applying this clause, as manipulated, the insured peril would need to be recognised as the interruption or interference with the business following the occurrence of a Notifiable Disease within 25 miles. Given that the “trends clause” is intended simply to put the insured in the same position as it would have been had the insured peril not occurred, and given the construction which we have found to be correct in relation to the ambit of the insured peril under Extension vii (a)(iii), what this means is that one strips out of the counterfactual that which we have found to be covered under the insuring clause. This means that one takes out of the counterfactual the business interruption referable to COVID-19 including via the authorities’ and/or the public’s response thereto. The relevant Indemnity Period, however, only starts with the first occurrence of a Notifiable Disease within the 25 mile radius, because that is provided for in the special definition of Indemnity Period contained within Extension vii itself. Accordingly, to the extent that there was business interruption or interference related to COVID-19 before that date the insured could not claim for it.

RSA 4 involves two wordings entitled “Material Damage and Business Interruption Policy” and branded “Resilience”. The lead wording bears the logo of Marsh Limited and the other wording that of Jelf Insurance Brokers Limited, another entity in the Marsh & McLennan group. The wording was provided by Marsh/Jelf to RSA and other insurers, although one of the General Conditions stipulates that the wording is “accepted by and adopted as [that] of the Insurer”. Cover on RSA 4 terms was placed by Marsh/Jelf acting on behalf of policyholders and was used for both SMEs and wholesale insureds.

The structure of RSA 4 is significantly different from that of the other lead policies in the test case. It sets out the perils insured (both Material Damage and BI), followed by Extensions, Exclusions, Claim Conditions, General Conditions, and then a lengthy and detailed definitions section.

The relevant clauses for the purposes of the present consideration of what has been described as RSA 4’s “disease cover” or “notifiable diseases peril” are as follows.

The relevant insuring clause is within the “Specified Causes” under Clause 2.3. So far as material it provides:

“In the event of interruption or interference to the Insured’s Business as a result of:

…

viii.

Notifiable Diseases & Other Incidents:

a)

discovered at an Insured Location;

b)

attributable to food or beverages supplied at or from the

Insured Locations;

c)

which are reasonably likely to result from an organism discovered at an Insured Location; and/or

d)

occurring within the Vicinity of an Insured Location, during the Period of Insurance

…

within the Territorial Limits, the Insurer agrees to pay the Insured the resulting Business Interruption Loss.” 127. The Definitions included the following:

“9.

Business Interruption Loss means:

i.

the Reduction in Turnover; ii. Increased Cost of Working; and/or iii. Research & Development Expenditure.

…

17.

Covered Event means the events as described in Insuring Clause 2.1, 2.2 2.3 or 2.4 or any applicable Extension.

…

42.

Indemnity Period means the period of time during which interruption or interference to the Insured’s Business occurs as a consequence of the Covered Event beginning with the occurrence of the Covered Event and ending not later than the end of the Maximum Indemnity Period thereafter.

…

The FCA contends that there will be cover under Clause 2.3(viii)(d) of RSA 4 for any interruption or interference with an insured’s business as a result of COVID-19, including the government’s and the public’s response thereto, as from 31 January 2020, when there were the first diagnosed cases of COVID-19 in the UK. The FCA’s contention is that, as from that date, COVID-19 had occurred within the “Vicinity” (as defined in RSA 4) of all premises in the UK. The FCA calls attention to the particular terms of the definition of “Vicinity” and argues that, as a notifiable disease with a national impact, and one which had already become a national epidemic in other countries, the occurrence of COVID-19 anywhere in the UK could reasonably be expected to have led to a national response, and all businesses in the UK would reasonably be expected to be impacted by such measures. Its alternative case is that the Vicinity requirement was satisfied when COVID-19 occurred in a more localised area of the insured premises, this being a question of fact to be determined in each case. But, the FCA submits, (i) an occurrence of COVID-19 within at least the same city, town or village or other development is always likely to be within the Vicinity; and in any case (ii) that the area involved remains relatively wide.

The FCA contends that the causal connection required by the words “as a result” is one of proximate causation, but this does not import any requirement of “but for” causation. In particular the FCA contends that it cannot have been the intention of the parties that cover would be confined to the adverse effects of a response to a disease which occurred only inside the Vicinity, assuming that imports a relevant geographical constraint, and that there would be no cover if the disease was both within and outside the Vicinity and the response was to the disease as a whole. As the FCA put it, RSA

RSA’s submissions were that Vicinity, as used in the policy, has a “physically circumscribed (as opposed to an expansive)” meaning. It contended that the words “area surrounding or adjacent to” suggest “close spatial proximity to the Premises”. The FCA’s case that Vicinity could extend to the whole of the UK was incorrect. RSA accepted that what constituted the Vicinity was fact specific in each case, but denied that Vicinity would always include the same city, town, village or other development. RSA further contended that the only causally relevant incidences of a notifiable disease were those within the Vicinity of the premises. The relevant insured event, it submitted, “must be one specific to the Vicinity” and cannot be one which takes place over a wide area which encompasses the Vicinity. What falls within the insured peril must happen inside, and not beyond, the Vicinity, and it is only that which happens inside the Vicinity which must proximately cause the relevant business interruption for there to be cover. Any interruption or interference with the insureds’ businesses would have been caused either by the general presence of COVID-19 in the country as a whole or by the closure measures and/or the social distancing measures, and none of those were caused by the occurrence of a notifiable disease within the Vicinity of insureds’ premises.

Mr Turner QC elaborated on these points, arguing, by reference to the definition of Covered Event that the cover was against events, not states of affairs, and that COVID19 was not an event. He further submitted that what has actually occurred could not have been reasonably expected, and that this was relevant to the application of the definition of Vicinity.

We also had the advantage of detailed submissions from the HIGA Interveners on RSA 4, including on the Notifiable Diseases aspect of its cover. The HIGA Interveners emphasised in particular the following matters. The definition of Notifiable Diseases & Other Incidents includes a list of 33 illnesses, which largely, but not entirely, mirrors the list of notifiable diseases in Schedule 1 of the 2010 Regulations. One of those listed diseases was SARS which emerged in China and spread to various countries in 20022003, and which in certain places led to stringent control measures. The policy thus specifically envisaged that there would be cover for business interruption as a result of SARS, a disease capable of causing an epidemic or pandemic, occurring in the Vicinity of the insured’s premises. Furthermore, the definition of Notifiable Diseases & Other Incidents meant that there would be cover for any additional disease which became notifiable under the 2010 Regulations. A disease included in that category might well be highly contagious with the potential to cause an epidemic or pandemic: a highly contagious, fast-spreading, serious disease is exactly the sort of disease which would be expected to be made notifiable.

Secondly, the HIGA Interveners stressed that the use of the term Vicinity meant that RSA 4 was unlike every other policy in the test case. The relevant area in which the disease must have “occurred” was not expressed in terms of a fixed, measurable distance from the insured location. Under the RSA 4 definition, while the Vicinity is an area surrounding or adjacent to an Insured Location, that did not place a limit on the size of the area. The second part of the definition was therefore of key significance, and what that involved was identifying the area surrounding the location in which events occurring would reasonably be expected to have an impact on the insured or its business. In that part of the definition, “events” must refer to the relevant triggers within the insured perils. What this definition entails is that the area constituting the Vicinity could depend on three matters, namely (a) the nature of the “events”, (b) the nature of the insured’s business, and (c) the location of the Insured Location. In the case of a new notifiable disease, the area constituting the Vicinity would depend on its nature. In the case of COVID-19, a highly contagious, and often fatal, disease, the relevant area would be the entirety of the UK because one would reasonably expect a national response which might affect the insured’s premises wherever they were. If that were correct then RSA had no remaining basis for denying cover, as any interruption or interference which resulted from the presence of COVID-19 in the country, including the national response thereto, would unquestionably be the result of a notifiable disease within the Vicinity.

Even if Vicinity were given a more restricted meaning, and assuming it could be established that there was an occurrence of COVID-19 within whatever constituted the Vicinity, there would still be cover. In particular cover was not precluded by the requirement that the interruption or interference with the business should be “a result of” the notifiable disease within the Vicinity. In this context Mr Philip Edey QC, for the HIGA Interveners, submitted that “as a result of” did not import a requirement of proximate causation. That test was apposite between the loss and the peril. Here the peril was business interruption or interference resulting from a notifiable disease in the Vicinity. The phrase “as a result of” was accordingly a link within the insured peril; there was no presumption that it meant proximate causation; and meant something looser than that. But even if that was wrong, any causal requirement was satisfied in this case, when the nature of the peril was understood, because, given the nature of the peril, the cause should be regarded as one indivisible thing, namely COVID-19, of which the local cases were a part, or each was a different equally effective cause.

We should finally mention a point made by Ms Susannah Jones, for the HIGA Interveners, in reply. She drew attention to the fact that one of the HIGA Interveners, insured under the equivalent of the RSA 4 wording, though not in fact by RSA, has 340 locations in the country, and the vicinity of those locations (on any view of that term) covers a large swathe of the country. On RSA’s argument, that insured could not establish that interference with its business was the result of the notifiable disease within the Vicinity of any of those 340 locations, because it would have resulted from the occurrence of the disease elsewhere. She submitted that that was an absurd result which demonstrated the fallacy of RSA’s case.

We begin with a preliminary matter, which was at issue between the parties, but which, as we understood it, does not have a significant effect. This is the question of the date on which COVID-19 is deemed to have been a notifiable disease. There is no dispute that COVID-19 is a notifiable disease in the sense of Definition 69 (ii). What is in issue is the relevant date of its “initial outbreak”, and thus the deemed date of its being a notifiable disease in accordance with that Definition. The FCA contends that the “initial outbreak” was on 31 December 2019, when the first cases in Wuhan were confirmed. RSA contends that the date of the “initial outbreak” is 31 January 2020, when the first positive test for COVID-19 occurred in England. We consider that the FCA is correct in relation to this point. The Definition does not refer to an initial outbreak “within the Territorial Limits” or within England and Wales or within the UK.

We turn therefore to what may constitute the Vicinity. We fully accept that the ordinary meaning of “vicinity” is a near area, and that the ordinary meaning of a word, even if the word is contractually defined, is some guide to what it is intended to mean as defined. We accept that it would not be normal parlance to refer to the whole of England and Wales as within the “vicinity” of any part of England and Wales. Nevertheless, given the particular terms of the definition of Vicinity in RSA 4, the perils insured under the policy, and the role that the defined term has in relation to those perils, we consider that the FCA and the HIGA Interveners are correct to say that, in relation to a disease such as COVID-19, an extensive area, possibly embracing the whole country, can be regarded as the relevant Vicinity for the purposes of this wording.

We say this because it appears to us to be correct that, given the terms of the definition, the extent of the Vicinity depends in part on the nature of the event which may occur within it. “Events” in this context must include the occurrence of any of the notifiable diseases within Definition 69 (i), (ii) and (iii). The occurrence of some of those diseases, anywhere in the country, might very well, possibly depending on the nature of the insured’s business, reasonably be expected to have an impact on it. One example given by the FCA is an outbreak of foot and mouth disease in sheep and cattle which, depending on the insured’s business, might reasonably be expected to have an impact on it, even though the outbreak was geographically very distant. More similar to the case of COVID-19 is SARS, which as we have set out was named in the policy as a notifiable disease. Given the severity of the SARS outbreak in 2002-2003 in various countries, it would in our view have been reasonably expected, at the time of conclusion of the contracts of insurance, that a significant outbreak of a SARS-like disease anywhere in the UK would have “an impact” on an insured or (depending in part on what that business was), its business. That is not to say that the extent of the measures in the event taken in relation to COVID-19 or the extent of its effects on the population and the economy would reasonably have been expected, but some impact would be. A similar point can be made in relation to malicious deposits of radioactive isotopes or of biological or chemical materials within Definition 69 (iv). It is, and we consider would have been at the time of the conclusion of the policies, reasonably expected that some such incidents would have a very wide-ranging, and possibly national, impact.

RSA’s response is that to concentrate on the potential width of the area which might be impacted by a particular event is to give no or insufficient weight to the first part of the definition of Vicinity, namely “an area surrounding or adjacent to an Insured Location”. We do not consider that this is the case. That the area must be surrounding or adjacent to the Insured Location does not of itself say much about the size of the area. The part of the definition which does dictate the size of the area is the second part

We therefore consider that the reasonable person would have understood the parties, in using the defined term Vicinity to mean an area whose extent would depend in part on the nature of the relevant “event”, but which could be very extensive, and indeed nationwide. Furthermore, if that is right, then we consider that it can properly be said that, when COVID-19 occurred, it was of such a nature that any occurrence in England and Wales would reasonably be expected to have an impact on insureds and their businesses, and therefore that all occurrences of COVID-19 were within the relevant “Vicinity”.

If we are wrong in rejecting RSA’s case as to the meaning of Vicinity, then the Vicinity will be an area which is fact sensitive, and depends, at least in part, on the nature of the insured’s business and its location, but that there is an overriding criterion of “close spatial proximity”. What that area might be is not something which we can determine and would need to be addressed on a case by case basis. It will only be material to do so, however, if RSA is right to contend that, if the Vicinity is confined as it alleges, then even if there was a case or cases of COVID-19 within the relevant Vicinity, because there will also have been the occurrence of the disease outside that Vicinity, the cover under 2.3 is not intended to apply to such a situation and/or that the cases outside the Vicinity will have been the cause of the interruption to or interference with the business. Those points we can deal with and should do so in case this matter goes further. These points raise similar issues to those which we have considered in the context of RSA 3.

In our judgment, it is not the correct construction of the coverage provision in Clause 2.3 (viii)(d) that cover is confined to what can be shown to be the business interruption or interference resulting from the occurrence of the disease within the Vicinity and does not extend to the business interruption or interference resulting from the disease being both inside and outside the Vicinity. We do not consider that that construction is dictated by the words used. The cover is not confined to the effects of a disease occurring only within the Vicinity, nor in our view is it expressly or implicitly confined to the effects only of the cases of the disease within the Vicinity. Instead the words used in Clause 2.3 are consistent with there being cover for the effects of a Notifiable Disease which has come within the Vicinity.

That this is the correct construction is, we consider, supported by a consideration of the nature of the insurance provided. It is for business interruption resulting from the occurrence within the Vicinity of, inter alia, certain diseases including the notifiable diseases under the 2010 Regulations. The regime under those Regulations was introduced to control epidemic, endemic and infectious disease. The nature of some of those diseases is that they may very well spread over a significant, and difficult to predict, area. SARS, which is specifically included in the table in the policy, would be an example of where that would be the case. It is also of the nature of such diseases

Consistently with this, we do not consider that it is the proper construction of the policy that cases of the notifiable disease within the relevant area are independent of, and can be regarded as a separate cause from, cases outside the area. The nature of highly infectious or contagious notifiable diseases is that cases in different areas will be related to each other, having spread from one source to multiple individuals in different places. As people move, the disease moves with them. Given the nature of notifiable diseases, which was a matter which was or is taken to be known to the parties when agreeing the insurances, we consider that the correct construction of the cover provided by Clause 2.3 and in particular by (viii)(d), is for the business interruption or interference caused by a notifiable disease, if that notifiable disease occurs within “the Vicinity”. The occurrence of at least a case of the disease within the Vicinity is required for there to be cover, which is a restriction which means, on the interpretation of Vicinity now under consideration, that there is no cover for diseases which are only geographically remote.

We have not overlooked the fact that Mr Turner QC argued that the RSA 4 cover was for “events” and not for “states of affairs”, which pandemics would be. As we understood it, this argument was principally directed to supporting RSA’s argument that cover was provided only for the occurrence of a relevant insured event, here a disease, specific to the Vicinity, and not in respect of one which occurs in a wide area encompassing the Vicinity. We did not consider that this point had force in relation to RSA 4. It is true that Definition 17 is of “Covered Event”, and that it refers to the matters in the Insuring Clauses as “events”. That appears simply as a shorthand for what is in the various insuring clauses, some of which (as in Clause 2.3 itself) are more specifically identified as “specified causes”, while Clause 2.4 is by contrast called a “Cyber Event”. Certainly, we do not regard Definition 17 as importing any additional requirement or re-definition of what constitutes an insured peril beyond what is set out in Clause 2.3 itself. Moreover, a consideration of the matters within Clause 2.3 indicates that the insurance in that clause is not simply against the effects of a matter which may happen at one time and be swiftly over. The cover in respect of the deposit of radioactive isotopes etc, which is part of the cover under Clause 2.3 (viii), and Definition 69 (iv), is an example. While the deposit may occur at a particular time, the cover will extend to the effect of the presence of the radioactive isotopes over a period of time.

While in the context of RSA 4, as elsewhere, arguments were addressed in terms of causation requirements, we consider that the issues of causation are largely answered by the construction to be placed on the cover provided. If we are correct in what we have said above, it clearly makes no sense to say that there is no cover because it cannot be shown that the cause of the business interruption was the disease insofar as it was outside and not within the Vicinity. Given the specific arguments which were addressed on the causal requirements of RSA 4 we will, however, briefly set out our conclusions.

We consider that Mr Edey QC was correct to say that the insured peril is “interruption or interference to the Insured’s Business as a result of [one or more of (i) to (xiii)]”; that the phrase “as a result of” is accordingly a link within the definition of the insured peril; and accordingly not between the loss and the insured peril, which is the link which is ordinarily characterised by a requirement of proximate causation. We consider that, as such, the causal connection which the phrase imports must be one which gives effect to and does not defeat the intentions of the parties in agreeing the cover. We agree with Mr Edelman QC that the phrase “as a result of” requires that the matter within one of (i) – (xiii) of Clause 2.3 should be an effective cause of the interruption or interference to the Insured’s Business. Given the nature of the insurance, which we have addressed above, that requirement is to be regarded as fulfilled on the basis either that the occurrence of the disease within the Vicinity was part of an indivisible cause, namely COVID-19 and the governmental and public reaction thereto, or was one of many equally effective causes.

RSA relies on the terms of the definition of Standard Turnover as the equivalent of a “trends clause” and as another route to saying that there will be no cover to the extent that business interruption would have arisen by reason of the occurrence of the disease, and the reaction thereto, outside the Vicinity. If we are right in relation to Vicinity, or if we are right in rejecting RSA’s case that, if Vicinity is given a narrower meaning, the occurrence of the disease outside the Vicinity constitutes a separate cause which can be relied upon as showing that the business interruption would have occurred even without the occurrence of the disease within the Vicinity, then there is no further point which arises in relation to the definition of Standard Turnover. On these bases the “Covered Event” will include the effects of COVID-19 on the business, and that is what will need to be removed for the purposes of the Standard Turnover definition. On our preferred analysis of what is meant by Vicinity, there will have been a Covered Event from the date of the occurrence of the disease in England and Wales. If Vicinity has a more confined meaning, then there will have been a Covered Event from the date on which there was the occurrence of a case of COVID-19 within whatever was the relevant

Two Argenta policy wordings have been included within the representative sample of policy wordings. They are the Holiday Insurance Underwriting Agencies (or “HIUA”) Guest House and B&B Insurance policy wording, and the HIUA Holiday Home and Self-Catering Accommodation policy wording. The BI section of each is materially identical and it is the HIUA Guest House and B&B Insurance policy which has been selected as a lead policy wording (Argenta 1) for the purposes of the test case. All of Argenta’s policyholders fall within Category 6.

Argenta 1 has a section providing BI cover. The basic insurance is dependent on damage insurable under the Buildings Insurance or Contents Insurance sections. There

The Definitions for the purposes of the BI section of the insurance included the following:

“NOTIFIABLE HUMAN DISEASE

illness sustained by any person resulting from

(a)

food or drink poisoning or

(b)

any human infectious or human contagious disease an outbreak of which the competent local authority has stipulated shall be notified to them excluding Acquired Immune Deficiency Syndrome (AIDS) or an AIDS related condition.

STANDARD GROSS INCOME

the GROSS INCOME during that period in the twelve months immediately before the date of the DAMAGE which corresponds with the INDEMNITY PERIOD to which such adjustments will be made as necessary to take account of the trend of the BUSINESS and of the variations in or other circumstances affecting the BUSINESS either before or after the DAMAGE or which would have affected the BUSINESS had the DAMAGE not occurred so that the figures thus adjusted will represent as nearly as may be practicable the results which but for the DAMAGE would have been obtained during the relative period after the DAMAGE”

The “Basis of Settlement” clause provides in part:

“The COMPANY will pay as indemnity the amount of the loss sustained by the INSURED as follows

A)

In respect of the reduction in GROSS INCOME

the amount by which the GROSS INCOME during the INDEMNITY PERIOD falls short of the STANDARD GROSS INCOME due to the DAMAGE.”

The FCA contends that there was relevant cover under sub-clause (d) of the “Defective Sanitation, NOTIFIABLE HUMAN DISEASE, Murder or Suicide” extension. It contends that COVID-19 was a Notifiable Human Disease for the purposes of the definition in the policy; and that there was an occurrence of an illness sustained by a person resulting from COVID-19 within a radius of 25 miles of the premises when a person or persons who had contracted COVID-19 was or were within that radius. It argues that the reference in the extension to the BI section to “such interruption” is a reference to the primary, damage-based BI insuring clause, which refers to “BUSINESS at the PREMISES is interrupted”. The FCA submits that all Argenta’s relevant insureds, and all Category 6 businesses, were interrupted from 16 March 2020, and thereafter by (i) the social distancing guidance including to stop all unnecessary travel of 16 March 2020; (ii) the 21 March Regulations closing bars and pubs to the extent those were run by Category 6 businesses; (iii) the 24 March 2020 statement on holiday accommodation closure; and (iv) the 26 March 2020 Regulations closing holiday accommodation.

The FCA contends that the exclusion in relation to “premises not directly affected” is inapplicable. As it puts it, that exclusion excludes losses to premises more than 25 miles away from the disease, not losses to premises less than 25 miles away from the disease.

In relation to whether there was “interruption as a result of … any occurrence” of COVID-19 within 25 miles, the FCA notes that the connection is between the disease and the interruption. There is no required element of authority action, though such action may well be the means by which the disease results in the interruption. The FCA contends that COVID-19 within the 25 mile radius was at least an effective and proximate cause of the interruption. The government was responding to the presence of COVID-19 around the country, and if there was a case of the disease within 25 miles of the premises that was a part of what the government and the public were responding to. The parties had clearly not intended that cover could be denied on the basis that the causal connectors or a “but for” test could defeat a claim when a disease was both inside and outside the 25 mile radius. The correct interpretation was that the disease occurring elsewhere was not “another matter”: it was the same disease as occurred within the 25 mile radius.

Argenta took issue with the FCA on three particular matters. In the first place, it submitted that an insured would have to demonstrate that any interruption with its business had been proximately caused by the occurrence of COVID-19 within the 25 mile radius. Only the consequences of such a “local” occurrence were insured. There was no cover provided for losses caused by measures taken in response to occurrences outside the 25 mile radius or the response of the public to such occurrences. While Argenta accepts that there may be some cases where the occurrence of COVID-19 has caused some covered loss, for example where customers cancelled bookings prior to 16 March 2020 due to a specific occurrence of COVID-19 within 25 miles of the premises, or as a result of targeted lockdowns like the one in Leicester, these represent a tiny proportion of the COVID-19 claims received by Argenta. Secondly, Argenta contends that sectional exclusion (iii), namely for losses arising from premises not “directly affected by the occurrence” was applicable. Thirdly, Argenta contends that the correct application of the “trends clause” contained in the definition of “Standard Gross Income” was that there should be removed from the counterfactual only the occurrence of COVID-19 within the 25 mile radius. On this basis policyholders would in all or almost all cases have suffered the same loss in any event because of the occurrence of the pandemic elsewhere and the governmental and international reactions to it. In particular Argenta contends that there is no cover for any loss that would have been sustained in any event as a result of anyone complying with the UK Government’s advice on 16 March or 24 March, because such losses would have been sustained anyway, or as a result of the requirements to close holiday accommodation imposed on 26 March in England, Wales and Scotland, again because they would have been sustained absent any “local” cases.

We will begin by considering the first group of Argenta’s arguments. They substantially overlap with those of RSA, and of MS Amlin and QBE to which we will come, in relation to other “disease clauses”.

The argument proceeds against a background of common ground that COVID-19 was added to the list of notifiable diseases set out in Schedule 1 to the 2010 Regulations as from 5 March 2020, and that it thereby became a Notifiable Human Disease as defined in the Argenta lead policy. Further, it is common ground between the FCA and Argenta that an “occurrence” of COVID-19 for the purposes of Extension 4(d) requires there to be at least one person within the relevant 25 mile zone on the relevant date who has contracted COVID-19 such that it is diagnosable, whether or not it has been verified by medical testing and whether or not it is symptomatic. Argenta accepts that there will have been an occurrence of COVID-19 within 25 miles of many of its policyholders by at least the end of April 2020, but its position is that it will be a matter for the policyholder to prove in each case. Argenta accepts that this may be proved by reference to the best available scientific evidence at the time the issue is decided and does not dispute that inferential proof may be appropriate in relation to this issue. We return to these latter matters when considering “Prevalence”, in Section H below.

Argenta’s case, as with that of other insurers where their policy contains what was called by Mr Edelman QC a “relevant policy area”, was that the policyholder needed to demonstrate that business interruption was the result of the occurrence of the disease within that area, and not its occurrence elsewhere. In our consideration of the RSA policies, we have already given our reasons why we did not consider that that argument was correct in relation to the cover provided under those policies. Equally, we do not consider Argenta’s similar argument to be correct in relation to its relevant policies, and for essentially the same reasons. In brief, we do not consider that Argenta’s argument accurately reflects the nature of the cover provided by an agreement in the terms of its policies.

Critical here again is the fact that Extension 4(d) does not say “any occurrence of a NOTIFIABLE HUMAN DISEASE only within a radius of 25 miles of the PREMISES” or anything which dictates such a reading. Essential also is that what was being insured under Extension 4(d) was business interruption resulting from Notifiable Human Diseases, and that it was known at the time of the conclusion of the contracts that such diseases embraced the list to which we have already referred, including SARS and other highly contagious or infectious diseases, which were required to be notified precisely in order to permit a response from the authorities, and some of which were capable of spreading over large areas, as infected people moved around. Furthermore, the nature of the definition of Notifiable Human Diseases in the policy meant that newly occurring diseases, if made notifiable by the relevant authorities, would count, even if they had not existed or been notifiable at the outset of the policy. The potentially widespread effects of the diseases in question were recognised by the fact that the “relevant policy area” was stipulated to be a radius of 25 miles, which, as the FCA reminded us on a number of occasions is an area of about 2,000 square miles, or as we were told, an area the size of Oxfordshire, Berkshire and Buckinghamshire combined. The ways in which a disease could have such an effect must have been recognised as including via the reaction of the authorities and / or the public. The parties thus knew or must be taken to have known that what was being insured under Extension 4(d) was business interruption deriving from a range of diseases some of which might spread over a wide and unpredictable area, and which might have an effect at a considerable distance from a particular case, including through the reaction of the authorities; and where it might well be impossible to distinguish whether that reaction was to the disease within or outside the relevant policy area.

In those circumstances, we consider that the proper construction of the agreement is that the parties were not agreeing that it was the business interruption consequences of a notifiable disease only insofar as it was within the “relevant policy area” that was being insured, but the business interruption arising from a notifiable disease of which there was an occurrence within the relevant policy area. We consider that this is consistent

In the context of Argenta’s submissions, two particular anomalies of its position appeared to us to be striking. The first relates to its case that there would not be any cover for any loss which would have been sustained as a result of anyone complying with the UK Government’s advice on 16 March as to avoidance of unnecessary travel, or of 24 March that Category 6 businesses should remain closed for leisure-related stays, or with the 26 March Regulations when businesses providing holiday accommodation in most of the UK were required to close their premises. This case is that, even if the policyholder could show that there had been a case of COVID-19 in the relevant policy area by any of these dates, there will be no cover because these steps would have been taken anyway. However, as the FCA pointed out, in the particular context of holiday accommodation, whether that accommodation was within an area of some 2,000 square miles with no COVID-19 or alternatively was in an area with COVID-19, would have been relevant to decisions as to where to go on holiday. Had there been no government action which closed down holiday accommodation, the occurrence of the disease in the relevant policy area would have been likely to affect the holiday businesses in that area. But, on Argenta’s case, because the government acted nationally and thus forestalled individual decisions based on or influenced by local incidence of the disease, there is no cover because that national response is to be regarded as an independent and the true proximate cause. The insured would thus have been in a better position under its insurance if the government had not taken steps to prevent the spread of the disease, including restricting its spread within the relevant policy area. It appears to us that such results cannot have been intended and an interpretation of the policy which gave rise to such results is not that which a reasonable person would have put on the language they used.

The other case which was the subject of argument in the context of Argenta was a “local” lockdown of the sort which, at the time of the hearing, had been introduced in Leicester, and which has subsequently been imposed or mooted elsewhere. Argenta accepts that in the case of the imposition of such localised lockdowns there would be cover for business interruption resulting for businesses within the relevant policy area. However, it is easy to imagine a case in which there were two local outbreaks (A and B), perhaps rather more than 25 miles apart, and the decision was taken on the basis of the occurrence of each to impose a lockdown on a wider area, which embraced both (the example given in argument was a lockdown of the East Midlands). We consider that it would not be open to an insurer of a business in the relevant policy area of outbreak A to contend that the lockdown would have been imposed in any event because of outbreak B. The reason, as we see it, is that such an argument would be inconsistent with the nature of the cover granted under Extension 4(d).

As with other insurers, much of this argument was put in terms of causation. As we have said in the context of RSA, we consider that questions of causation are largely answered by the issue of construction, because it determines what can and what cannot be regarded as independent causes. Thus, to take the example given above, the authorities’ response to outbreak B could not be regarded as an independent cause, which could be said by insurers to be the true, or a sufficient cause of the lockdown and

In relation to Argenta’s wording, as with others, we consider that the insured peril is properly to be regarded as business interruption at the premises (“such interruption”) as a result of one of Extensions 1-6. The link “as a result of” is thus within the composite insured peril. We nevertheless agree with Ms Leigh-Ann Mulcahy QC for the FCA that it imports a requirement that one of Extensions 1-6 must be an effective cause of the interruption; but that the application of the causal test must give effect to and not thwart the intention of the parties. We agree that the test can be regarded as satisfied on the basis that the occurrence of the disease within the area was a part of an indivisible cause, constituted by COVID-19. Or alternatively, that each of the cases of the disease was an independent cause, and they were all equally effective in producing the government response.

While expressed in negative rather than positive terms, this clause is in effectively the same terms as Clause 4 of the Additional Definition in respect of Notifiable Diseases in RSA 3. What we have said above in relation to that clause is equally applicable to this clause of the Argenta policy. It indicates that, in the case of multiple premises of the insured, it is only losses arising from the premises which are directly affected by the occurrence, discovery or accident (i.e. whichever of Extensions 4(a) to (e) is applicable) which are covered. In relation to Extension 4(d), its main significance is likely to be to eliminate losses arising from premises which are not within a radius of 25 miles of an occurrence of a notifiable disease. On our analysis, set out above, however, if there was the occurrence of a case of COVID-19 within the 25 mile radius of an insured’s premises, then we consider that those premises count as “premises directly affected by the occurrence”.

It is common ground between the parties that, though the Basis of Settlement provisions refer to and require “DAMAGE”, and though DAMAGE is defined as “accidental loss damage or destruction”, those provisions are intended to be applicable to the nondamage business interruption extensions and must be “made to work” in relation to those covers. That involves a degree of verbal manipulation.

We consider that the manipulation required is essentially the same as that which we have set out above in relation to the “trends clause” in RSA 3. The reference to DAMAGE in the definition of Standard Gross Income and in clause (A) of the Basis of Settlement clause should be understood as a reference to the insured peril. Given that the “trends clause” is intended simply to put the insured in the same position as it would have been had the insured peril not occurred, and given the construction which we have found to be correct in relation to the ambit of the insured peril under Extension 4(d), what this means is that one strips out of the counterfactual that which we have found to be covered under the insuring clause. This means that one takes out of the counterfactual the business interruption referable to COVID-19 including via the authorities’ and/or the public’s response thereto.

The parties differ to some extent as to what can be decided at this stage as to the extent of any interruption with the businesses of Argenta’s policyholders.

There is no significant issue between the FCA and Argenta as to what can constitute an “interruption” to business. Specifically, Argenta does not contend that “interruption” for the purposes of the BI section of its relevant policies requires “a complete cessation of business”. We consider that concession to be correctly made.

There is also no significant issue in relation to the effect of the 26 March Regulations. Argenta accepts that those regulations did cause an “interruption” in the business of policyholders located in England, insofar as such businesses were otherwise continuing, and insofar as bookings did not fall within any of the exceptions. Similarly, Argenta accepts that the equivalent regulations in Wales and Scotland and Northern Ireland caused an “interruption” in the business of the relevant policyholders, insofar as those businesses were otherwise continuing.

What appeared, at least on the skeleton arguments, to be in issue is whether the Court can decide that the effect of the 16 March advice and recommendations was to interrupt relevant businesses; and whether the 21 March Regulations did so if the guesthouse or holiday accommodation operated a bar or restaurant. Argenta’s position is that the former were capable of causing an interruption to many businesses, including businesses operating holiday accommodation in the UK; and that the latter were likely to have caused an interruption to businesses of policyholders insofar as they operated a bar and/or restaurant; but that whether there was actually an interruption was a matter of fact to be determined in each case. We consider that Argenta’s position is correct. Argenta: Pre-notifiability losses

An issue arose between the FCA and Argenta as to whether the Court could or should make any finding in relation to whether losses could be claimed by policyholders before COVID-19 became a notifiable disease. Argenta contends that such losses could not, on any view, be recoverable, because there could not have been an occurrence of a Notifiable Human Disease, which means a disease “an outbreak of which the competent local authority has stipulated shall be notified to them”, within the 25 mile radius, before COVID-19 had been made notifiable under the 2010 Regulations. Mr Edelman QC said that this was not a point which the FCA was arguing in this test case, and that the Court should avoid expressing a view in relation to it, so as not to prevent its being raised by policyholders in “other fora” if they saw fit.

Given the submissions which Mr Edelman QC did make on the subject, including about

There are three lead wordings of MS Amlin Underwriting Limited (“MSA”) before the Court. Each of MSA 1 and 2 has a “disease clause”. They are in very similar but not identical terms. We will consider them together.

MSA 1 is a Commercial Combined Policy purchased predominantly by policyholders in Categories 3 and 5, essential shops etc., and other businesses never required to close pursuant to any government regulations.

Insofar as relevant at this point, the MSA 1 wording provided for Optional BI insurance, in Section 6. The basic insuring clause in Section 6 provided as follows:

“For each item in the schedule, we will pay you for any interruption or interference with the business resulting from damage to property used by you at the premises for the purpose of the business occurring during the period of insurance caused by an insured cover and provided that damage is not excluded under section 1.”

“Damage” is defined as “Loss or destruction of or damage to the property insured as stated in the schedule and used by you in connection with the business.” Thus, the primary insuring clause of Section 6 covers business interruption as a result of physical damage.

Section 6 also has certain “Additional covers” which are said to be “provided as standard”. They include a range of covers, including failure of utilities, and lottery win by employees. They also include the following:

“We will pay you for:

...

6.

Notifiable disease, vermin, defective sanitary arrangements, murder and suicide

Consequential loss as a result of interruption of or interference with the business carried on by you at the premises following:

a)

i. any notifiable disease at the premises or due to food or drink supplied from the premises;

ii.

any discovery of an organism at the premises likely to result in the event of a notifiable disease;

iii.

any notifiable disease within a radius of twenty five miles of the premises;

b)

the discovery of vermin or pests at the premises which causes restrictions on the use of the premises on the order of the competent local authority;

c)

any accident causing defects in the drains or other sanitary arrangements at the premises which causes restrictions on the use of the premises on the order of the competent local authority; or

[d)] any murder or suicide at the premises.

The maximum we will pay for any one loss will not exceed £100,000.

Conditions

1.

For the purpose of this additional cover premises will mean only those locations stated in the premises definition. If this policy includes an additional cover which deems damage at other locations to be insured, the additional cover will not apply to this additional cover.

2.

We will not be liable for any costs incurred in the cleaning, repair, replacement, recall or checking of property.

3.

We will only be liable for loss arising at those premises which are directly affected by the loss, discovery or accident.

…”

The definition of “consequential loss” in the General Definitions of the policy was “Loss resulting from interruption of or interference with the business carried on by you at the premises in consequence of damage to property used by you at the premises for the purpose of the business.”

The Additional definitions for the purposes of Section 6 of the policy included the following:

“Indemnity period

The period beginning with the loss and ending not later than the maximum indemnity period after that during which the results of the business will be affected following the loss’

However for the Notifiable disease additional cover the following definition applies:

The period during which the results of the business will be affected following the loss, discovery or accident beginning:

a)

In the case of 1 and 4 [by which is meant (a) and (d) in the version quoted above] with the date of the loss or discovery; or

b)

In the case of 2 and 3 [i.e. (b) and (c)] with the date from which the restrictions on the premises are applied and ending not later than the maximum indemnity period after that.

…

Maximum indemnity period

The indemnity period stated in the schedule, other than under the Notifiable Disease and the Lottery win by your employees additional covers where it is 3 months.

Notifiable disease

Illness sustained by any person resulting from:

a)

food or drink poisoning; or

b)

any human infectious or contagious disease (excluding Acquired Immune Deficiency Syndrome (AIDS)) an outbreak of which the competent local authority has stipulated will be notified to them.”

The bases of settlement clauses in Section 6 provide that the insurer will pay for loss of gross profit, additional cost of working, rent receivable, book debts and accountant’s charges. The loss of gross profit clause provides that the insurer will pay for “reduction in turnover”, being “the sum produced by applying the rate of gross profit to the amount by which the turnover during the indemnity period will following the damage fall short of the standard turnover”. “Standard turnover”, is defined as:

“The turnover during that period in the 12 months immediately before the date of the damage which corresponds with the indemnity period to which adjustments will be made as necessary to provide for the trend of the business and for variations in or other circumstances affecting the business had the damage not occurred, so that the figures adjusted represent as nearly as may be reasonably practicable the results which but for the damage would have been obtained during the relative period after the damage.”

MSA 2 is called a “Retail Insurance Policy”. MSA 2 was issued in several forms to suit Retail (including Categories 3 and 4) Leisure (Categories 1 and 2) and Office and Surgery (including Categories 3 and 5). At sub-section 2 of “Section A, Automatic cover”, it provides BI cover.

The relevant terms of MSA 2 are materially identical to those of MSA 1 save that the cover in respect of “Notifiable disease, vermin, defective sanitary arrangements, murder and suicide” is expressed as follows:

“We will pay you for:

…

6.

… consequential loss following:

a)

…

iii.

any notifiable disease within a radius of twenty five miles of the premises

…”

The definition of “consequential loss” appeared in the Additional definitions, in the BI sub-section, and was as follows: “Loss resulting from interruption of or interference with the business carried on by you at the premises following damage to property used by you at the premises for the purpose of the business.”

The FCA contends that there is clearly cover for insureds who suffered business interruption as a result of COVID-19 and the governmental and public response thereto, after COVID-19 became a notifiable disease on 5 / 6 March 2020 in England and Wales, and after there had been a person or persons who had contracted COVID-19 such that it was diagnosable, within a radius of 25 miles of the premises, whether or not it had been verified by medical testing. The FCA contends that the “disease clauses” in MSA 1 and 2 are the most straightforward such clauses in this test case. This is because, as the FCA puts it, they require only (1) loss “resulting from” or “as a result of” interruption or interference with the business (2) “following” any COVID-19 within 25 miles of the premises.

The FCA accepts that the definition of “consequential loss”, which itself refers to “damage” must be manipulated in order to apply to the “Notifiable disease” covers in MSA 1 and 2, for otherwise there could be no recovery in respect of those non-damage perils. It contends however that the quantification machinery involved in the bases of settlement clauses, and in particular the definition of “standard turnover”, have no application to the extensions, and apply only to cases of property damage.

MSA disputes the FCA’s case on a number of grounds, which overlap with those relied upon by other insurers. It accepts that from 5 March 2020 in England and from 6 March 2020 in Wales COVID-19 was a notifiable disease for the purposes of the policies. It does not accept that it could be shown that there was a notifiable disease within the relevant area if the case was not diagnosed and was asymptomatic. In any event it contends that the insured would have to establish that any business interruption was caused by the illness of a person or persons within the 25 mile radius from the premises. And MSA further contends that this can be tested by assuming that the person(s) in the radius who had the disease, did not have it. Would that have made a difference? MSA suggests that the answer will be that the government’s response would have been no different.

MSA accepts that the terms of the definitions of “consequential loss” in MSA 1 and 2 have to be manipulated to extend their application to the non-damage based covers. It contends, however, that the bases of settlement provisions apply as much to the Insuring Clause and the Additional covers, including the “disease clause” covers in

The issues which arise in relation to the cover provided under the “disease clauses” in these two MSA wordings are, in essence, the same as some of those included in our discussion of RSA 3 above.

In particular these MSA “disease clauses” constitute agreements to pay losses resulting from business interruption at the premises “following” one of a number of matters. The MSA disease clauses are, as the FCA submits, in one respect, even simpler provisions, in that, unlike the clauses in RSA 3 and Argenta 1, they are not expressed in terms of the “occurrence” of the notifiable disease within 25 miles of the premises, but simply in terms of consequential loss “following” a notifiable disease within 25 miles of the premises.

We have carefully considered whether and the extent to which what we have said in relation to RSA 3 above applies in relation to each of the MSA covers now under consideration, conscious of the fact that they are separate wordings and have to be construed in their own right and in full. We have concluded that our conclusions in relation to RSA 3 as to what is covered by the insuring clause, as to the meaning of “following”, as to the satisfaction of any causal requirements in the “disease clause” and as to the inconsistency with the insurance of an argument that the occurrence of the disease outside the radius, and the response thereto, constitute a separate cause which means that there is no “but for” causation and no cover, apply equally to the MSA covers.

There are, however, four matters which need to be addressed because they are specific or arise in a different way in relation to the MSA covers.

The first is MSA’s argument that “following” meant proximate and in any event at least “but for” causation because of the terms of the definition of “consequential loss” in

As was common ground, the definition of “consequential loss” is inappropriate to the Additional covers and has not been properly drafted with them in mind. It appeared to us that the appropriate manipulation of the consequential loss definition, to make it applicable to the cover in Additional cover 6, was that it should read “interruption of or interference with the business carried on by you at the premises following [one of (a) to (d)]”. Certainly, we see no reason to construe “following” in Additional cover 6 as having a meaning different from its natural meaning by reference to a different clause which is admittedly drafted without proper reference to the terms of the Additional covers. In any event, we would consider that, even if the definition of “consequential loss” is to be read, as manipulated, as including the words “in consequence of [the insured peril]”, this is a shorthand for the causal requirement in the Additional covers.

We observe that in the case of MSA 2, the definition of “consequential loss” itself uses the phrase “following” and accordingly the argument which we have just considered in relation to the definition of “consequential loss” in MSA 1 does not apply in the same way in relation to MSA 2.

Secondly, the link required in Additional cover 6 is with “any notifiable disease within a radius of 25 miles of the premises”. The MSA “disease clauses” are not expressed in terms of the “occurrence” or “manifestation” of the disease, but rather in terms of there being the disease within the 25 mile radius. That would embrace any case where a person has or persons have the disease within the radius. The definition of notifiable disease does not require that the disease should have been diagnosed. More generally, as the FCA submits, the fact that the relevant MSA clauses do not refer to an “occurrence” makes it, to our minds, relatively straightforward to conclude that the cover extended to the effects of a notifiable disease if and from the time it is within the 25 mile radius and is not limited to the specific effects only of the instances of the disease within the radius.

Thirdly, as the FCA pointed out, MSA’s interpretation of the provision in Condition 3 to Additional cover 6, namely that the insurer would only be liable for loss “arising at those premises which are directly affected by the loss, discovery or accident”, was in line with its own, namely that what it was aimed at was to exclude losses which might arise from the interruption of business at insured’s premises B, if there was a “knock on” effect on that business from a matter affecting the insured’s premises A. We consider that that is the correct construction of the provision. MSA was right, in our view, not to seek to raise an argument that that provision assisted its argument that there was no effective cover under the “disease clauses” because of the fact that the notifiable disease, here COVID-19, was present and had effects outside the stipulated radius.

Fourthly, there was an issue as to whether the “trends” provision in the definition of “standard turnover” in both MSA 1 and 2 was applicable to the covers under the “disease clauses”. We conclude that it is, notwithstanding the reference in it to “damage”. In our view the parties cannot have intended that the policy should be silent on how non-damage business interruption losses should be calculated given that they had made provision for the calculation of damage-based business interruption claims. Furthermore, there are particular definitions of “Indemnity period” and “Maximum indemnity period” for the “Notifiable Disease cover”. Those definitions are only relevant when assessing the insured’s losses in accordance with the basis of settlement provisions, because it is only those provisions that require, in relation to the gross profit basis of settlement that the insured’s claim for reduction of turnover be assessed by reference to the amount that the turnover “during the indemnity period” falls short of standard turnover. That appears to us to show that the parties intended the basis of settlement provisions to be applicable to the additional covers.

With that said, however, as with other “trends” provisions relied upon by insurers, we do not consider that the potential applicability of the “trends” provisions has a material

There are seven QBE wordings, which have been grouped into three wording types which have been called QBE 1, QBE 2 and QBE 3. There are what we consider to be significant differences between them. We propose to refer to the relevant terms of the three wordings; then to set out the parties’ cases on all of them; and then to give our conclusions in relation to each.

QBE 1 is a Business Combined Insurance Policy. A wide range of covers is available to the policyholder. One group is included in the “Property section”, another in the

“Business interruption section”. There are also groups of cover in, for example, the “Money section”, the “Employers’ liability section” and the “Goods in transit section”. The relevant provisions for present purposes are within Section 7, the “Business interruption section”.

In that section, by Clause 7.1.1, there is cover for “loss caused by the interruption of or interference with the business resulting directly from damage to property used by you at the premises within the territorial limits”, on the terms specified. “Damage” is defined (by Clause 23.25.1) as meaning “loss of, destruction of or damage to tangible property”. “Premises” is defined (by Clause 23.80) as “the buildings and land shown in the schedule being occupied by you for the purpose of the business”. “Business” is defined (in Clause 23.17) as “the business stated in the schedule”.

In Clause 7.3 there are set out certain “Extensions applicable to this section”. A range of covers is then set out. These include cover against loss: (a) resulting from business interruption or interference as a result of damage to the insured’s property whilst in transit or at a site other than “the premises” (Clause 7.3.2); (b) resulting from business interruption or interference as a result of damage to property at the premises of the insured’s direct customers and direct suppliers (Clause 7.3.3); (c) resulting from business interruption or interference as a result of denial of access to the premises, both when there has been damage to property and in certain circumstances where there has not (Clauses 7.3.4 and 7.3.5); (d) in consequence of loss of attraction by reason of damage to property within a one mile radius of the premises (Clause 7.3.7); (e) constituted by the additional costs arising from an employee or group of employees resigning as a result of winning the lottery or equivalent (Clause 7.3.8); and (f) resulting from business interruption or interference caused by failure of the supply of utilities (Clause 7.3.13).

Amongst these “Extensions” is Clause 7.3.9, which is the clause most directly relevant. It provides:

“Murder, suicide or disease interruption of or interference with the business arising from:

a)

any human infectious or human contagious disease (excluding Acquired Immune Deficiency Syndrome (AIDS) or an AIDS related condition) an outbreak of which the local authority has stipulated shall be notified to them manifested by any person whilst in the premises or within a twenty five (25) mile radius of it;

b)

actual or suspected murder, suicide or sexual assault at the premises;

c)

injury or illness sustained by any person arising from or traceable to foreign or injurious matter in food or drink provided in the premises;

d)

vermin or pests in the premises [sic];

e)

the closing of the whole or part of the premises by order of a competent public authority consequent upon defect in the drains or other sanitary arrangements at the premises.

The insurance by this clause shall only apply for the period beginning with the occurrence of the loss and ending not later than three (3) months thereafter during which the results of the business shall be affected in consequence of the damage [sic]”

It is also pertinent to note the terms of one of the exclusions to section 7, namely that in Clause 7.4.3, which is in these terms:

“… we will not indemnify you for any loss:

…

Off premises damage any loss caused by:

a)

acts of any civil, government or military authority caused by or following:

i)

conflagration; or ii) storm; or iii) earthquake; or iv) explosion; or

v)

impact by aircraft or other ariel [sic] or spatial device; or vi) flood; or

vii)

actual or suspected presence of any radioactive or toxic material (including “dirty bombs”); or viii) suspect packages

…’

The lead wording in QBE 1 contains bases of settlement providing for “Insurable gross profit”, “Gross fees”, “Gross revenue”, “Increased cost of working”, “Rent receivable”, and “Book debts”. All these refer to “damage”. These bases of settlement include adjustment language in their definitions. For example, the cover for reduction in gross revenue is the amount by which the “gross revenue” during the “indemnity period” will, in consequence of the “damage” fall short of the “standard gross revenue” (Clause

There are some variations between the language of the wordings as to bases of settlement between the four policies within the QBE 1 type.

QBE 2 is called a “NDML Nightclub and Late Night Venue Policy”. It has various sections providing different types of cover. Insured section B is the BI section. The basic BI cover is provided in Clause 3.1 and is for business interruption if the insured’s premises are damaged. Clauses 3.2 and 3.3 have a number of additional business interruption covers. Of particular relevance is Clause 3.2.4, which is in the following terms:

“3.2.4

Infectious disease, murder or suicide, food or drink or poisoning

Loss resulting from interruption of or interference with the business in consequence of any of the following events:

a)

any occurrence of a notifiable disease at the premises or attributable to food or drink supplied from the premises;

b)

any discovery of any organism at the premises likely to result in the occurrence of a notifiable disease;

c)

any occurrence of a notifiable disease within a radius of 25 miles of the premises;

d)

the discovery of vermin or pests at the premises which cause restrictions on the use of the premises on the order or advice of the competent local authority;

e)

any accident causing defects in the drains or other sanitary arrangements at the premises which causes restrictions on the use of the premises on the order of the competent local authority;

f)

any occurrence of murder or suicide at the premises; provided that the

g)

insurer shall not be liable for any costs incurred in cleaning, repair, replacement, recall or checking of property except as stated below;

h)

insurer shall only be liable for loss arising at those premises which are directly subject to the incident;

i)

insurer’s maximum liability under this cover extension clause in respect of any one incident shall not exceed GBP 100,000 or 15% of the total sum insured (or limit of liability) for this insured section B, whichever is lesser, any one claim and GBP 250,000 any one period of insurance.”

In QBE 2 there is no equivalent of the exclusion in respects of specific acts of any civil, government or military authority which appears in QBE 1.

QBE 2 contains the following definition of “indemnity period”:

“18.47.1

Indemnity period means the period beginning with the occurrence of the damage and ending not later than the maximum indemnity period thereafter during which the results of the business will be affected in consequence of the damage.

18.47.2

But for the purposes of clause 3.2.4 the indemnity period shall mean the period during which the results of the business shall be affected in consequence of the an [sic] event beginning in the case of:

3.2.4

a) and d) with the occurrence or discovery of the incident,

3.2.4

b) and c) above with the date from which the restrictions on the premises are applied,

and ending not later than twelve (12) months thereafter.”

QBE 2 contains a definition of “notifiable disease”, in Clause 18.67, as follows:

“Notifiable disease means illness sustained by any person resulting from:

18.67.1

food or drink poisoning, or

18.67.2

any human infectious or human contagious disease, an outbreak of which the competent local authority has stipulated shall be notified to them excluding Acquired Immune Deficiency Syndrome (AIDS), an AIDS related condition or avian influenza”

In Clause 3.4, under the heading “Business interruption limitations and exclusions”, appear various bases of settlement, including “gross profit/estimated gross profit”, “gross revenue/estimated gross revenue”, “increased cost of working”, “additional increased cost of working” and “rent receivable”. All these refer to “damage” as defined. As with the similar provisions in QBE 1, they incorporate trends clauses through definitions of terms such as “annual turnover”, which incorporate a requirement that they be “trend adjusted” and the “trend adjusted” Clause, Clause 18.119, is in the same terms as that in QBE 1.

QBE 3 is called a “Commercial Combined Insurance Policy”. Under Section 3 there is BI insurance. The basic cover, in Clause 3.1.1, is for business interruption “resulting directly” from accidental “damage”, which is defined to mean loss, damage or destruction of tangible property. Clause 3.4 contains a number of extensions to the BI cover. These provisions include the following:

“The insurer shall indemnify the insured for the following, if shown as insured in the schedule:

…

3.4.8

Notifiable disease, murder or suicide, food or drink poisoning

Loss resulting from interruption of or interference with the business as covered by this section in consequence of any of the following events:

a)

an occurrence of a notifiable disease at the premises or attributable to food or drink supplied from the premises;

b)

the discovery of any organism at the premises likely to result in the occurrence of a notifiable disease;

c)

an occurrence of a notifiable disease within a radius of one

(1)

mile of the premises;

d)

the discovery of vermin or pests at the premises which causes restrictions on the use of the premises on the order or advice of the competent local authority;

e)

an accident causing defects in the drains or other sanitary arrangements at the premises which causes restrictions on the use of the premises on the order or advice of the competent local authority;

f)

an occurrence of actual or suspected murder, suicide or actual or alleged sexual assault at the premises.

Provided that:

i)

the insurer shall only be liable for loss arising at those premises which are directly subject to the incident;

ii)

the insurer shall not be liable for any costs incurred in cleaning, repair, replacement, recall or checking of property except as provided for in the Property section.

‘Notifiable disease’ means illness sustained by any person resulting from any diseases that may be notifiable under the Health Protection (Notification) Regulations 2010.”

The term, “accident”, as used in Clause 3.4.8(e) is a defined term, meaning (by Clause

The bases of settlement provisions are very similar to those in QBE 1, and there is a similarly worded “trends adjusted” clause as Clause 25.180.

The FCA’s case in relation to QBE 1 is that there will have been a human infectious or human contagious disease an outbreak of which the local authority has stipulated shall be notified to them within a radius of 25 miles of the premises, for the purposes of Clause 7.3.9, when a person who had contracted COVID-19 was within that radius of the premises. It contends further that there will have been an interruption or interference with the business from 16 March 2020 given the government advice and lockdown restrictions from that date, alternatively (but only when the business was ordered on that date to close) from 20 March 2020; and that the interruption or interference arose from and was caused by the disease, if they would not have occurred had there been no COVID-19 outbreak or interventions by the government. The FCA further contended that the bases of settlement clauses and quantification machinery (including the “trends clause”) do not apply to claims under the relevant extension of cover in two of the four policies within the QBE 1 type (QBE 1 PBCC040120 and PBCC170619), though it accepts that the quantification machinery (and “trends clause”) is made applicable in relation to QBE 1 POFF180120. In relation to QBE 1 POFP040120 the FCA contends that while there is an incorporation of the contractual quantification machinery, and while the relevant definitions say that they are “trend adjusted” there is no definition of what that means, and it is ineffective.

The FCA’s case is that the “disease clause” in QBE 2 is not materially different in effect, save that QBE 1 has a three month maximum indemnity period, and QBE 2 excludes clean-up costs. It contends that no quantification machinery (and no “trends clause”) was incorporated into QBE 2.

Equally, the FCA contends that the “disease clause” in QBE 3 responds in materially the same way, though the relevant radius is specified as 1 mile, rather than 25 miles, which means that there will not be cover unless it can be shown that there was a case within a mile of the premises. The FCA accepts that, because of the inclusion in the introductory words to several extensions, including the “disease clause”, of the words “as covered by this section”, that the quantification machinery for the whole BI section (including the “trends clause”) is applicable to cover under those extensions.

QBE’s case, by contrast, is that none of QBE 1-3 provided cover and that they were not intended to provide cover in respect of a national pandemic or the Government response to an actual or feared national pandemic. It contends that it was a critical feature of the cover under each of the wordings that the insurance was in respect of a disease manifested by a person, or occurring, within a 25 mile (or 1 mile in the case of QBE 3) radius of the premises. The relevant insurances were to cover, only, the consequences of such manifestation / occurrence. That is their proper construction. Accordingly, QBE contends that the appropriate analysis as to whether there is cover in the circumstances of the present case is as follows:

Mr Mark Howard QC for QBE in his oral submissions elaborated on this argument. He submitted that the core issue was one of construction. Did the insurances have the meaning contended for by the FCA, such that they covered the effects of a notifiable disease, if it occurred within the relevant policy area, or did they cover only the consequences of such occurrence of the disease as there was in the relevant policy area, such that if an interruption in the business would have happened in any event by reason of the occurrence of the disease elsewhere, there was not cover? As he submitted, once that question of construction had been answered, whichever way it was answered, there would be little if anything left by way of issues of causation. He submitted that QBE’s construction was correct, and that the FCA’s approach to the clauses meant that the requirement of the manifestation or occurrence of the disease within the relevant radius became a matter of chance or happenstance, irrelevant to the purposes of the insurance, and reduced the insurance to “a game of lotto”.

We also received submissions on the QBE policies from the HIGA Interveners. The HIGA Interveners did not accept that the words in the QBE “disease clauses” which required that the business interruption or interference should “arise from” (QBE 1) or be “in consequence of” (QBE 2-3) the relevant fortuity denoted proximate causation.

The HIGA Interveners contended that the critical question which arose in relation to all the QBE wordings was the following: In the context of the notifiable disease cover undoubtedly provided by the policy, is the effect of the radius requirement also that where the notifiable disease is present beyond the radius, and there is a single response to the presence of the notifiable disease in that wider area which applies to that wider area (within which, of course, the insured premises sit), the cover which would otherwise have existed had the response been confined to the relevant area vanishes unless the insured is somehow and highly improbably able to divide up the effect of the single response as between its effect within the radius and outside it? The HIGA Interveners contended that the obvious answer to that question is “no”.

At the heart of the dispute in relation to these policies is the question of whether the insured is entitled to recover only for business interruption which can be shown to have been the consequence of the manifestation / occurrence of the disease within the particular radius as opposed to its manifestation / occurrence or the threat of its manifestation or occurrence outside that radius. This depends on a proper construction of the policies, and it is necessary to consider each separately and in turn.

One minor issue of construction should be addressed at the outset. This is what is meant by “manifested” as used within Clause 7.3.9(a). Clearly someone who is displaying symptoms of a disease can be said to “manifest” it. We consider that it would also be the case that a person “manifested” the disease if, though superficially asymptomatic, he or she was diagnosed with the disease, because the disease would have “manifested” itself to the diagnoser. We do not consider that it is possible to speak of someone who is asymptomatic and has not been diagnosed as having the disease as having “manifested” it.

We turn therefore to the main issues of construction of Clause 7.3.9. It is a clause which is expressed in a somewhat convoluted way. We understood it to be common ground, however, and in any event appears clear, that the phrase “an outbreak of which the local authority has stipulated shall be notified to them” simply identifies the relevant human infectious or contagious diseases as being notifiable diseases, including those notifiable under the 2010 Regulations. AIDS and AIDS-related conditions are not relevant. Accordingly, for the purposes of simplicity, Clause 7.3.9 can be shortened as follows: “interruption or interference with the business arising from: (a) any notifiable human infectious or contagious disease manifested by any person whilst in the premises or within a 25 mile radius of it…”. That is the relevant insured peril.

Focusing on the language and structure of Clause 7.3.9, we consider that, within the insured peril, the required causal link (“arising from”) is between the interruption or interference with the business on the one hand and the notifiable disease on the other, provided it has been “manifested” by a person within the 25 mile radius. We do not consider that the clause most naturally reads, or should be construed, as saying that the interference has to result from the particular case(s) in which the disease is manifested within the 25 mile radius. Instead the cover is for the effects of a notifiable disease if it has been manifested within the 25 mile radius. This appears to us to be apparent from the juxtaposition of the phrase relating to business interruption with that relating to notifiable disease, and the fact that the phrase “manifested by any person whilst in the premises or within a twenty five mile radius of it” is most naturally read as an adjectival clause limiting the class of notifiable diseases which, if they interfere with the business, will lead to coverage.

We consider that this makes good sense. On this interpretation, cover is provided for the effects of a notifiable disease on the business, if the disease has come to or within the specified distance of the premises; and is not provided solely for the consequences of the particular manifestation of the disease by one or more individuals who happen to be within the radius. On this basis, what would be significant is the fact of the disease having been (relatively) close, not whether it was the particular case(s) within the radius which had had any specific effect.

The clause does not spell out, or seek to limit cover by reference to, the ways - or what might be called the “transmission mechanisms” - by which the notifiable disease may cause interruption or interference with the business. It is obvious that one such way for business interruption or interference to arise is as a result of the response of governmental or local authorities, and of the public, to the existence of the notifiable disease. Consistently with this, there is no exclusion in Clause 7.4.3 of loss caused by civil, government or military authority caused by or following disease. If a notifiable disease manifested itself both within and outside the 25 mile radius it would be likely that there would be such governmental / public responses to the disease outbreak, rather than to specific cases of the disease, either those within or outside the radius. The construction we favour of the terms of Clause 7.3.9 avoids the anomaly of there being no effective cover in such a case.

On the basis of this construction no significant further questions of causation arise. The insurers clearly cannot, on this basis, contend that the occurrence of the disease elsewhere, or the reaction to it, are to be regarded as separate causes. There is to be regarded as sufficient causation of the business interference if the disease which has manifested itself in the radius is an effective cause of that business interference.

The wording of QBE 2 (and QBE 3 to which we will come) appears to us to be significantly different both from QBE 1 and from other “disease clauses” with which we are concerned. This is despite the fact that the language used is in various respects similar to other clauses which we have considered above.

Specifically, in QBE 2, there is a combination of factors which together, to our minds, indicate that the cover is indeed intended to be confined to the results of specific (relatively) local cases. This combination of factors has meant that our first, and subsequent, impressions of the relevant “disease clause” is not as contended for by the FCA and the HIGA Interveners. In particular, the relevant clause has the following features. In the first place, the insuring clause itself identifies the matters in (a) to (f) as “events”. This indicates that what is being insured is matters occurring at a particular time, in a particular place and in a particular way: see the dictum of Lord Mustill in Axa Reinsurance v Field [1996] 1 WLR 1026 at 1035 as to the meaning of “event”. This is

This focus of the clause is then emphasised by the fact that in (h), it is stated that the insurer is only to be liable for loss arising at those premises which are directly subject to the “incident”, and that in (i) there is a limitation per “incident”. These uses of the word “incident” appear to us to reinforce the fact that the clause is concerned with specific events, limited in time and place. This is further emphasised by the terms of Clause 18.47.2, which reiterates that the cover is in respect of the effects on the business in consequence of “an event”; and, although the reference in Clause 18.47.2 to the subparagraphs of Clause 3.2.4 (at least in the version put before us) are inapposite, it is significant that there is a further reference to the “incident”. Clause 3.2.5, which provides that certain clean-up costs may be recoverable in relation to matters falling within Clause 3.2.4 contains a further reference to an “incident” (Clause 3.2.5(b)).

These indications from Clause 3.2.4 and related clauses appear of a piece with the drafting of other parts of the wording. In the General definitions, at Clause 18.2, there is an “accumulation limit”, which limits the maximum amount of benefits payable by the insurer “where a single event, or series of events in a twenty kilometres radius originating from the same proximate cause, occurs…”. This indicates that the draftsman, and the parties in agreeing the wording, distinguished between an event, a series of events, and a cause. When the term is used in Clause 3.2.4 it envisages that each of the matters within (a) to (f) will be an event, not a series of events or a cause.

We accept that, for the purposes of QBE 2, there will be an “occurrence” of COVID-19 within the radius when a person has the disease within the area, whether symptomatically or not, because that person has then “sustained” the illness within the definition in Clause 18.67. However, as we have said, the terms of Clause 3.2.4 show that there is cover only if there is business interruption as a result of the “event” of the person(s) sustaining that illness within the area. It is difficult to see how there could be such consequential interference if the disease was asymptomatic and undiagnosed.

Given our construction of Clause 3.2.4, the issues as to causation largely answer themselves. We accept that the words “in consequence of” imply a causal relationship. As we have found that this clause, unlike others we have considered, is drawing a distinction between the consequences of the specific cases occurring within the radius and those not doing so, because the latter would constitute separate “events”, we consider that insureds would only be able to recover if they could show that the case(s) within the radius, as opposed to any elsewhere, were the cause of the business interruption. In the context of this clause, it does not appear to us that the causation requirement could be satisfied on the basis that the cases within the area were to be regarded as part of the same cause as that causing the measures elsewhere, or as one of

While we consider that the correct construction of the “disease clause” in QBE 3 is not as clear as it is in relation to QBE 2, because QBE 3 does not have all the features which QBE 2 has and which we have referred to above, we consider that the correct construction is essentially the same for QBE 3 as for QBE 2.

The “disease clause” in QBE 3 shares with QBE 2’s the reference to cover being for the consequence of any of the “events” in (a) to (f), and the phrasing of (c) in terms of “an occurrence” of a notifiable disease. Proviso (i) to Clause 3.4.8 refers to an “incident”. On these bases we consider that this clause too is confining cover to the consequences of certain happenings, in particular specific occurrences of the disease within the radius, as opposed to other happenings or events, including instances of people contracting the disease outside the radius. The fact that the radius specified in QBE 3 is 1 mile makes it all the easier to accept that this is what the parties have agreed, because that reinforces the view that what is being contemplated is specific and localised events.

Given this construction of the “disease clause” the questions of causation fall to be answered in the same way as for QBE 2.

“Trends clauses”

As we have described there is an issue between the parties as to whether some, but not all, the QBE policies apply the contractual quantification mechanism to the relevant “disease clause” covers, including the “trends clauses”.

In our view all the QBE policies should be interpreted as applying the contractual quantification mechanism applicable to damage-related BI claims to the non-damage BI covers, including by manipulation of the requirement for “damage”, save insofar as inconsistent with more specific provisions as to quantification.

As will be apparent from what we have already said, however, we do not consider that it makes any significant difference as to whether the “trends clauses” are incorporated or not.

Hybrid clauses

We turn to consider a number of clauses which have been termed “hybrid”. Again, this terminology is only to be taken as a convenient means of referring to certain policy terms, which refer both to restrictions imposed on the premises and to the occurrence or manifestation of a notifiable disease. The policies which contain such clauses are those which have been called Hiscox 1–4 and RSA 1. There is also a further coverage provision in RSA 4, which it is convenient to deal with in this section. We will commence by considering the Hiscox policies.

We were informed that about 65% of the Hiscox insureds who have policies affected by the issues are in Category 5, and typically comprise small accountancy, consultancy, legal and other similar professional businesses. Of the 35% of insureds not within Category 5 considerable numbers are in Category 2, namely indoor and outdoor leisure activities, which were required to close or cease business by the 21 and 26 March Regulations or Category 4, namely non-essential shops, required to cease business by the 26 March Regulations, save for making deliveries or otherwise providing services in response to orders received on the web, by telephone or post. A small percentage of Hiscox insureds are in Category 1.

A total of 31,171 Hiscox policyholders had live policies in one of the four groups into which the FCA split the wordings (i.e. those referred to as Hiscox 1–4). 16,181 policyholders have policy wordings in group 1, which contains 8 different wordings. The lead wording chosen by the FCA is the Retail BI-16105 wording, but this is only the third most common. According to Hiscox, the wording in Hiscox 1 with most policyholders is a Professions BI wording used to insure professional service businesses including small accountancy firms and solicitors, virtually all of which are in Category 5. There are 10,872 policyholders with policies within group 2, which contains 23 different wordings. Hiscox 3 embraces five different wordings, and there are 582 policyholders within this group. There are four different wordings within Hiscox 4 with 3,536 policyholders.

The most relevant clauses in the lead policies are as follows.

The lead policy in Hiscox 1, contains a section of cover in respect of “Property – business interruption”. That section provides in part:

“What is covered We will insure you for your financial losses and other items specified in the schedule, resulting solely and directly from an interruption to your activities caused by:

…

Public authority 13. your inability to use the insured premises due to restrictions imposed by a public authority during the period of insurance following:

a.

a murder or suicide;

b.

an occurrence of any human infectious or human contagious disease, an outbreak of which must be notified to the local authority;

c.

injury or illness of any person traceable to food or drink consumed on the insured premises;

d.

defects in the drains or other sanitary arrangements;

e.

vermin or pests at the insured premises.

…

How much we will pay We will pay up to the amount insured …

…

Loss of income The difference between your actual income during the indemnity period and the income it is estimated you would have earned during that period or, if this is your first trading year, the difference between your income during the indemnity period and during the period immediately prior to the loss, less any savings resulting from the reduced costs and expenses you pay out of your income during the indemnity period. …

Loss of gross profit The sum produced by applying the rate of gross profit to any reduction in income during the indemnity period plus increased costs of working and alternative hire costs less any expenses or charges which cease or are reduced.

…

Business trends Provided that you advise us of your estimated annual income, or estimated annual gross profit if applicable, at the beginning of each period of insurance, the amount insured will automatically be increased to reflect any special circumstances or trends affecting your activities, either before or after the loss. The amount that we will pay will reflect as near as possible the result that would have been achieved if the insured damage had not occurred.”

The lead policy in Hiscox 2 contains a “Special definition” for the Property – business interruption section, as follows:

“Notifiable human disease Any human infectious or human contagious disease, an outbreak of which must be

notified to the local authority.”

Hiscox 2 contains a number of covers in the Property – business interruption section, including the following:

“We will insure you for your financial losses and any other items specified in the schedule, resulting solely and directly from an interruption to your business caused by:

…

Public authority 5. your inability to use the salon due to restrictions imposed by a public authority during the period of insurance following:

a.

a murder or suicide;

b.

an occurrence of a notifiable human disease;

c.

injury or illness of any person traceable to food or drink consumed on the premises;

d.

defects in the drains or other sanitary arrangements;

e.

vermin or pests at the premises.”

The quantification provisions of Hiscox 2 were substantially the same as those for Hiscox 1. The “Business trends” clause was, however, in the following terms:

“Business trends The amount we pay for loss of income or loss of gross profit will be amended to reflect any special circumstances or business trends affecting your business, either before or after the loss, in order that the amount paid reflects as near as possible the result that would have been achieved if the insured damage or restriction had not occurred.”

In Hiscox 3, the most relevant provisions were as follows:

“We will also insure you for your loss of gross profit up to the limit stated in the schedule as applicable resulting solely and directly from an interruption to your business caused by the following:

…

Public authority d. your inability to use the business premises due to restrictions imposed by a public authority following:

i.

a murder or suicide;

ii.

an occurrence of any human infectious or human contagious disease an outbreak of which must be notified to the local authority;

iii.

injury or illness of any person traceable to food or drink consumed on the premises;

iv.

vermin or pests at the premises.”

The “business trends” clause in the lead Hiscox 3 policy was similar to that in the lead Hiscox 2 policy, save that it contained no reference to “the restriction”.

Hiscox 4 contained an additional requirement within the relevant insuring clause, which was as follows:

“We will insure you for your financial losses and any other items specified in the schedule, resulting solely and directly from an interruption to your business caused by:

…

Public authority 7. your inability to use the business premises due to restrictions imposed by a public authority during the period of insurance following:

a.

a murder or suicide;

b.

an occurrence of a notifiable human disease within one mile of the business premises;

c.

injury or illness of any person traceable to food or drink consumed on the premises;

d.

defects in the drains or other sanitary arrangements; or

e.

vermin or pests at the premises.”

“Notifiable human disease” was defined in Hiscox 4 as “Any human infectious or human contagious disease, an outbreak of which must be notified to the local authority”. The “Business trends” clause in the lead Hiscox 4 policy was in the following terms:

“Business trends The amount we pay for loss of gross profit will be amended to reflect any special circumstances or business trends affecting your business, either before or after the loss, in order that the amount paid reflects as near as possible the result that would have been achieved if the insured damage, insured failure, cyber-attack or restriction had not occurred.”

The FCA contends that there was cover under the “public authorities” clause of each of the Hiscox 1-3 lead policies for all businesses from 16 March 2020. From that date, owners, employees and customers could not use the business premises for their intended purposes due to restrictions imposed by the UK Government as to nonessential travel or contact, social distancing, and so on, which followed an occurrence, indeed numerous occurrences, of COVID-19. Alternatively, and in any event, the FCA contends that the “public authorities” clause in Hiscox 1-3 was satisfied between 20 and 26 March 2020 for businesses in Categories 1, 2, 4, 6 and 7 from the relevant date in March when they were required to close (in whole or in part) by the Regulations or other Government announcements, when owners, employees and customers could not use those premises due to the restrictions.

In relation to Hiscox 4, the “public authorities” clause was satisfied from the same dates and in the same circumstances as in relation to Hiscox 1-3, provided that an insured can establish the presence of COVID-19 within a one mile radius of the premises preceding the relevant authority action.

The FCA contended, in particular, that the outbreak of COVID-19 in the UK constituted the “occurrence” of a notifiable disease. The word “occurrence” in the Hiscox wordings did not mean something which was small scale or local. In relation to Hiscox 4, there will have been an “occurrence” of COVID-19 within the one mile radius, whenever and wherever a person had contracted COVID-19 such that it was diagnosable. It did not need to have been medically verified.

The FCA contended that “your inability to use” meant an inability to utilise or employ the premises for or with their intended aim or purpose. That inability might be partial or total. The insured’s inability to use the premises could be caused by restrictions imposed on customers or other people who would be needed to attend the premises if it was to function for its intended purpose.

Further the FCA contended that the “restrictions imposed by a public authority” did not have to involve mandatory requirements, having the force of law. It would be enough if an authority promulgated something that it “requires or expects” to be followed. Accordingly, the stay at home instruction given on 16 March 2020 amounted to the imposition of restrictions leading to inability to use the premises. As the FCA put it, “these statements were received by the populace as instructions and acted on as instructions [and were] restrictions imposed…”

The FCA contended that an interruption to the insured’s activities did not require a cessation or stop. Instead, there will be an interruption to an insured’s activities if there has been a material prevention of use of the premises. The requirement that the losses should be “solely and directly” the result of the interruption, was relevant only to the relationship of the losses to the interruption, not to the causes of the interruption. As to the requirement that the interruption should be “caused by” the restrictions imposed by the public authority, there could be little dispute as to this, if the FCA’s case on the meaning of “interruption”, of “restrictions imposed” and of “inability to use” were accepted. While the FCA accepted that “following” had a causal element, it was a requirement of a looser connection than proximate causation.

The FCA contended that there was no relevant “trends clause” in those cases in which there had not been a reference to “restriction” in the clause itself. In those policies where the “trends clause” was in the same form as that in the lead Hiscox 1 policy, the trends clause was “upwards only and/or optional at the election of the insured”; and could only increase the amount insured. If not taken up by the policyholder then there can be no trends adjustment at all.

In order to quantify loss, or under any trends clauses, the FCA submitted that the proper counterfactual is a situation “in which there was no COVID-19 in the UK and no Government advice, orders, laws or other measures in relation to COVID-19, or alternatively in which such of these events as the Court adjudges to be interlinked had not occurred.”

In relation to the “public authorities” clauses in the Hiscox policies we had the benefit of submissions by the Hiscox Interveners. The Hiscox Interveners supported the contentions of the FCA. The Hiscox Interveners particularly stressed that the words “solely and directly” qualify only the requirement that losses must result from the interruption to the business. On the other hand, it is enough if one cause of the inability to use the insured premises was the restrictions imposed by the public authority. If all the elements of the insured peril ((1) occurrence of a notifiable disease, (2) following which, restrictions imposed by a public authority, (3) an inability to use the insured premises as a result, (4) causing an interruption in business activities, (5) resulting solely and directly in losses) were present, then Hiscox must indemnify the insured in respect of its business interruption loss. Insofar as there was a question of “but for” causation, the question was whether “but for” all the elements of the insured peril what would the insured’s position have been, i.e. “but for” the entire chain of events, including the occurrence of the disease. That also answered the counterfactual question posed by Hiscox: it was a world in which all the elements of the insured peril did not exist.

Mr Ben Lynch QC for the Hiscox Interveners further emphasised that, while in certain cases the language used in the Hiscox “public authorities” provisions was capable of the narrow meaning attributed to it by Hiscox, it was also capable of having a wider meaning, and where it was, it should be recognised as capable of covering a range of circumstances. In such cases, the right construction was the broader one. Further he submitted that it was impossible to read into the Hiscox policies the restriction of “occurrence” to a local matter specific to the premises. The clauses did not include any such limitation and could not be read as doing so.

Hiscox made a number of points or groups of points in relation to the “public authorities” clauses, which meant, it contended, that the FCA’s (and the Hiscox Interveners’) case in relation to those clauses was fundamentally incorrect. Mr Jonathan Gaisman QC elaborated on these arguments in oral submissions. Those points or groups of points can be summarised as eight-fold, as follows.

In the first place, Hiscox emphasised that the “public authorities” clauses referred to “your inability” to use the premises. Accordingly, it was only the inability of the insured to use the premises which was relevant and no one else’s. Secondly, inability to use denoted the insured being unable to use the premises, not something less. Thirdly, the “restrictions” referred to, particularly because of the use of the word “imposed”, must be mandatory. The only restrictions capable of fulfilling those requirements, and then only in respect of businesses covered by them, were Regulation 2 of the 21 March Regulations and Regulations 4 and 5 of the 26 March Regulations. Fourthly, the only “occurrence” contemplated by the clauses was one which was local and specific to one or more of the insured, its business or its premises. There was, Hiscox contended, no relevant “occurrence” here. Fifthly, the clause contemplated an “occurrence” that was identifiable, which, in the present context meant a case of COVID-19 which had been scientifically confirmed. An unidentifiable occurrence could not lead to public authority action. Sixthly, the word “following” required a causal connection and there was no such connection between any occurrence and the restrictions. Seventhly, interruption meant cessation and nothing less. Eighthly, that the correct counterfactual which must be posed for the purpose of causation and hence the assessment of loss was,

as a matter of general principle and/or (ii) under the “trends clauses” and/or (iii) by reason of the words “solely and directly resulting from” in the stem to the Hiscox wordings, one which assumes the existence of COVID-19 in the UK, its impact on the economy and public confidence, and the government measures falling short of the mandatory “restrictions”. If Hiscox was wrong in relation to any of its submissions on coverage, then the counterfactual would be altered accordingly, as the insured peril was widened.

We begin with the issue of what is meant by “restrictions imposed” by a public authority. The issue has to be addressed by considering the words used and the context in which they are used. In our view, what these words mean is something which is mandatory, and they do not include something which is less than mandatory. This is the natural meaning of “imposed”. Furthermore, these words are used in the context of a resulting inability on the part of the insured to use its own premises. That reinforces the conclusion that what is being referred to is something that has the force of law. Each of paragraphs (a) to (e) of the “public authorities” clause in – by way of example – Hiscox 1 is a case in which mandatory action can be taken by relevant authorities in respect of premises under identifiable legal or statutory powers, and the reference to “restrictions imposed” most naturally refers to the legally binding powers that can be exercised in relation to those situations.

What this means for present purposes is that the only relevant matters which constituted “restrictions imposed” are those which were promulgated by statutory instrument, and in particular Regulation 2 of the 21 March Regulations and Regulations 4 and 5 of the 26 March Regulations. Guidance, exhortation and advice given by the Government, including by the Prime Minister, including as to social distancing, do not count as “restrictions imposed” by a public authority. Nor would cases in which people decided not to visit premises, notwithstanding that they were legally permitted to do so, be caused by any “restrictions imposed”.

We turn then to the phrase “inability to use”. In our view this plainly does not embrace any and every impairment of normal use. “Unable to use” means something significantly different from “hindered in using” or similar. Furthermore, the phrase is used in a context which includes the various sub-clauses (a) to (e) (in Hiscox 1), in each of which situations restrictions amounting to a complete inability to use the premises for the purposes of the business (albeit typically for a limited time) are readily foreseeable. We agree with Hiscox that there will not be an “inability to use” premises merely because the insured cannot use all of them; and equally there will not be an “inability to use” premises by reason of any and every departure from their normal use. Hiscox accepted, however, in our view correctly, that partial use might be sufficiently nugatory or vestigial as to amount to an “inability to use” the premises. Whether that was so would depend on the facts of a particular case.

We were not, however, persuaded by Hiscox’s submission that the “restrictions imposed” contemplated by the “public authority” clause necessarily had to be directed to the insured, or to the insured’s use of the premises, although we accept that they would often be so directed in the various cases falling within sub-clauses (a) to (e) (of Hiscox 1). By way of example, it seemed to us that it might be the case that, by reason

But although we considered that there could be “restrictions imposed” which were not directed specifically at the insured or the insured’s use of the premises, we did not consider that it could be said that Regulation 6 of the 26 March Regulations amounted to a “restriction imposed” which could have led to an “inability to use” the premises of all insureds where that insured’s business had relied on the physical presence of customers. As we have said, “inability to use” premises means what it says and is not to be equated with hindrance or disruption to normal use. Given the exceptions to Regulation 6, which include the general exception of “reasonable excuse” and the specifically enumerated exceptions including travel for the purposes of work where it was not reasonably possible for the person to work from home, and given the possibility (and reality) that businesses could operate or come to operate by contacting customers at home, it appears to us that the cases in which Regulation 6 would have caused an “inability to use” premises would be rare. Whether there were such cases would be a question of fact.

We do not accept Hiscox’s case that the “occurrence” of an infectious or contagious disease referred to in the relevant sub-clause of the “public authority” clause (subclause (b) of Hiscox 1) has to be one which is “small scale, local and in some sense specific to the insured” as Mr Gaisman QC put it. We recognise that that is the typical case of what may fall within the sub-clause. However, we do not consider that the words used dictate that no other episode (to use what is intended to be a neutral term) can qualify, and nor is the context sufficient to mandate such a limited meaning. It is true that the phrase used is “an occurrence”, but in talking about “an occurrence” of a disease the clause is in our view plainly not equating “occurrence” with each case of the disease, such that if there were two or more cases it would necessarily follow that there was not “an occurrence” but multiple occurrences. We consider that there can be “an occurrence” of a notifiable disease for the purposes of the clause if there is an outbreak of such a disease. We therefore consider that the COVID-19 outbreak in the UK could qualify as “an occurrence” of a notifiable disease.

Hiscox, however, is right to say that in all of the wordings, there is a requirement that the restrictions imposed must “follow” the “occurrence” of the notifiable disease, and that this imports some sort of causal connection. In relation to Hiscox 1-3 this may not be of great significance in that, if we are correct to say that the outbreak of COVID-19 in the UK was “an occurrence” of a notifiable disease then any “restrictions imposed”, to the extent that there were any (as to which see above), “followed” that outbreak.

With more hesitation, we reach a similar conclusion in relation to Hiscox 4. For reasons which we have canvassed in relation to the “disease clauses” above, when one is considering notifiable diseases, it is not difficult to envisage that official responses will be to the full extent of an outbreak, and not necessarily specific to those in a given geographical area. In the circumstances, and unless the language otherwise dictates, it is appropriate to regard a response as having “followed” the local occurrence of the disease, provided that the response was temporally posterior, if it was a response to the outbreak of which the local occurrence formed a part. Were this not the case, similar anomalies to those we have considered in the context of the “disease clauses” would arise in relation to Hiscox 4. An example would be a case where there was a severe outbreak of a disease within the same town as the insured’s premises but just outside the one mile area, and also a few cases within the one mile area; and the outbreak outside the 1 mile area, and its potential to spread, was the reason which led to the local authority imposing the restrictions, the fact that there had actually been cases within the one mile area playing no part in the authority’s thinking. It would be surprising if there were not cover in such a circumstance, and while we accept that that result might be dictated by the terms of the clause, we do not consider that it is by the wording of Hiscox 4. The relevant clause would reasonably have been understood to be requiring the link to be between the restrictions imposed and a disease which has “occurred” within a one mile radius, rather than between the restrictions and the specifically local instances of the disease.

As to Hiscox’s argument that “interruption” must mean a cessation or stop, and nothing less, as we say below in the context of the Hiscox NDDA clause, there is much force in Mr Gaisman QC’s submission that the use of only “interruption” in the stem wording should not encompass “interference” as opposed to complete cessation of the business, since had it been intended that interference with the business would be covered, the wording could and would have said so, as other wordings we have to consider do. However, the stem wording has to be viewed in the context of the insuring clauses which follow. As we set out below, it seems to us clear from a number of those clauses, at least in the Hiscox 1 lead wording, that “interruption” in this wording is intended to mean “business interruption” generally, including disruption or interference, not just complete cessation.

The FCA did not accept that the “trends clause” was applicable to the business interruption claims made other than in those cases in which the “business trends” clause itself referred to “restrictions”. We consider that the trends clauses were intended to apply in all cases, and where there is no specific reference to “restrictions” the clause needs to be manipulated. There would be no good reason why the parties would have wanted the trends clause to apply differently as between claims involving material damage and claims involving non-damage business interruption. This is emphasised by the fact that such clauses can favour insureds on particular facts, as well as insurers on others, as Mr Gaisman QC pointed out. We consider that there is here an obvious error and that it can be corrected as a matter of construction, as we consider that it can be in relation to a number of other policies which we have had to consider where a similar mistake has been made.

What this means is that in those “business trends” clauses which do not refer to “restrictions”, they should be read as providing, in their concluding sentence, insofar as relevant to the present case: “The amount that we will pay will reflect as near as possible the result that would have been achieved if the insured damage had not occurred [or if you had not been unable to use the insured premises due to restrictions imposed by a public authority following an occurrence of any human infectious or human contagious disease, an outbreak of which must be notified to the local authority]”. That is, we should add, to write out in longhand what is conveyed by the reference to “restriction” in those “business trends” clauses in which that word appears.

We do not consider that a “business trends” clause in the form of that which we have quoted above from the Hiscox 1 lead policy (i.e. one which commences “Provided that you advise us…”) can be properly regarded as optional in any relevant sense. What might be described as conferring an option is the first sentence, which permits an increase in the amount insured. The second sentence (commencing “The amount that we will pay…”) is distinct, relates to the amount which will be paid by way of indemnity, and is not optional.

As to how the counterfactual is to be applied, whether it is being considered for the purposes of considering the losses which the insured can claim, either as a matter of application of the insuring clause, or pursuant to the “trends clause”, we consider that the exercise must give effect to the insurance effected. This means assuming that the insured peril did not occur. The insured peril is a composite one, involving three interconnected elements: (i) inability to use the insured premises (ii) due to restrictions imposed by a public authority (iii) “following” one of (a) to (e), relevantly (b) an occurrence of an infectious or contagious disease. What the insured is covering itself against is, we consider, the fortuity of being in a situation in which all those elements are present. In answering the counterfactual question as to what would have been the position of the insured’s business but for the occurrence of the insured peril, it is accordingly necessary to strip out all three interconnected elements, including in this instance the national outbreak of COVID-19.

We do not consider that it would give effect to the intentions of the parties for the assumption to be that there were no mandatory government restrictions and no inability to use the premises as a result specifically of such restrictions, but that the national outbreak of the disease and other governmental responses to it, and the economic and social consequences of these, were assumed to have been the same as occurred. That would not, in our judgment be how a reasonable person would understand what was agreed. It would involve an unrealistic and artificial exercise, and one which fails to recognise that the occurrence of the disease is an essential element of the insured peril, and of what the insured has covered itself against.

Moreover, and importantly, the effect of the imposition of restrictions of the sort involved in the Hiscox “public authorities” clauses will invariably, or almost invariably, have the result of preventing the insured from seeing what would have been the effect of the emergency (whatever it was) in the absence of the restrictions. That is part of the very nature of such restrictions. We regard it as not being a reasonable interpretation of the parties’ agreement that the insured could only recover to the extent that it could show what was the position which would have appertained without the restrictions, but on the basis that the situation which had led to those restrictions is assumed to have occurred, in circumstances where the insured has been put into an unusual situation by the emergency (and accordingly cannot rely simply on the ordinary results of its business) and where the effect of the restrictions will have been to render it impossible to say with any certainty what that position would have been.

In our view, the FCA effectively illustrated the fallacy and unreality of the insurers’ case in relation to the counterfactual, with particular reference to the Hiscox “public authorities” clauses, by focusing on the provision of cover in respect of an inability to use an insured’s premises, assumed to be a restaurant, due to restrictions imposed on them by the local authority following the discovery of vermin dislodged from a nearby building site. On the insurers’ case, the counterfactual involved stripping out the restrictions, but assuming the vermin were in the premises throughout, with whatever other consequences on the business the presence of vermin would have had. Thus, on insurers’ case the insured could not recover in respect of the period after the imposition of the restrictions, unless it could show that customers not coming to the restaurant was due to the restriction imposed rather than due to the vermin, and that if the insured could not demonstrate that customers would have come despite the presence of vermin, it could not recover. As the FCA submitted, this would render the cover largely illusory, as insurers would argue that, as no one is likely to want to eat at a restaurant infested by vermin, all or most of the business interruption loss would have been suffered in any event. Such illusory cover cannot have been intended and is not what we consider would reasonably be understood to be what the parties had agreed to.

Further and more specifically, not least of the difficulties with the insurers’ case is that they did not adequately explain how, in the example we have just mentioned, the insured would demonstrate the reason why customers had not come, in circumstances where the effect of the restrictions was that the restaurant was closed and they could not come. It was suggested on behalf of insurers that it would be done by cross-examining customers as to their motives in not coming. Quite apart from the impracticality of such an exercise if dealing with a large number of customers, insurers provided no answer to how all the customers who had not come could be identified, given that, ex hypothesi, they had not come, and the restrictions had made it impossible for them to do so.

For these reasons, we considered that the FCA’s case in this regard was to be accepted, and that the correct application of the counterfactual in the current case is to compare the actual performance of the business with that which the business would have achieved in the absence of the COVID-19 outbreak which led to restrictions (as understood in the sense we have given above) and the inability to use the premises. As we explain elsewhere, however, the counterfactual can only assume that the insured peril applies from the time that the restrictions are imposed, and only for as long as they are imposed.

RSA 1 is known as “Cottagesure”. It proclaims itself to be “The Holiday Cottage Owners’ Insurance Policy”. It provides for a number of different types of cover, including Business Interruption Insurance. There are a number of so-called “Extensions to Cover”, which are introduced by the words “This Insurance Also Covers”.

The extension which is relevant for present purposes is Extension 2, which is headed “Disease, Murder, Suicide, Vermin and Pests”, and which is in the following terms:

“Loss as a result of

A)

closure or restrictions placed on the Premises as a result of a notifiable human disease manifesting itself at the Premises or within a radius of 25 miles of the Premises.

B)

injury or illness sustained by any customer or Employee arising from or traceable to foreign or injurious matter in food or drink sold from the Premises.

C)

closing of the whole or part of the Premises by order of the Public Authority for the area in which the Premises are situate as a result of defects in the drains and other sanitary arrangements at the Premises

D)

murder, rape or suicide occurring at the Premises.

E)

closure or restrictions placed on the Premises on the advice or with the approval of the Medical Officer of Health or the Public Authority as a result of vermin and pests at the Premises.”

There is a sub-limit of £250,000 for this extension, and an exclusion for any amount of the loss which continues more than twelve months after the occurrence of the loss.

The definitions in the policy wording include definitions of Gross Revenue, and Loss of Gross Revenue. The latter is as follows: “The actual amount of the reduction in the Gross Revenue received by You during the Indemnity Period solely as a result of Damage to Buildings…”

The FCA contends that there was clearly a “closure or restrictions placed on the premises” from at least 26 March 2020, when Category 6 businesses were required to close by the 26 March Regulations. The FCA submits, however, that there was such “closure or restrictions” at an earlier date, and in particular from 16 March, by reason of the stay at home and social distancing guidance given on that date. The FCA contends, in this regard, that the clause contains no requirement for a closure to originate from any authority, official or otherwise. “Closure” means prevention of access, total or partial; and “restrictions” would include any measure which hindered or prevented (directly or indirectly) access to or use of the premises. The “placing” of those restrictions on the premises can be indirect as well as direct.

The FCA contends further that whenever the policyholder can prove that a person had contracted COVID-19 such that it was diagnosable, then the disease had been “manifested by any person” in a particular place. For reasons developed in the context of the “disease clauses” the FCA’s case was that the requirement that the closure or restrictions should be “as a result of” a disease manifesting itself in the 25 mile radius imported no special requirement other than proximate causation which was met here because the Governmental actions and advice were the response to the occurrence of COVID-19 nationwide.

The FCA’s case is that the contractual quantification machinery, and trends clause, are inapplicable to the extension. The loss under the Extension in question is simply the “loss as a result of” the closure or restriction. The FCA’s case on the appropriate counterfactual for the purposes of assessing loss was the same as it made elsewhere.

RSA’s submissions in relation to RSA 1 were as follows. In relation to the requirement of “closure” or “restrictions”, it accepted that this was satisfied by the measures imposed by Regulation 5(3) of the 26 March Regulations on holiday cottages with effect from 1 pm on 26 March 2020. But RSA does not accept that there was any “closure” or “restrictions” placed on cottages prior to 26 March 2020. A “closure” or “restrictions” could only be placed on the premises by some form of legal order or regulation impacting the premises and specific to the premises or premises of that type. There was nothing prior to 26 March which fell into that category.

In relation to the requirement that a notifiable disease should have “manifested” itself within the 25 mile radius, RSA submitted that a disease could not have manifested itself unless the disease was apparent, and so would not include asymptomatic and undiagnosed cases. Further, it was only the cases which manifested themselves within the 25 mile radius which had any causal relevance. The policy would respond, if at all, only to the consequences of events within the prescribed radius, rather than to the consequences of events taking place in a wider area. None of the government’s measures, including the closure measures of 26 March, had been caused by the presence of COVID-19 specifically within the relevant radius of each premises, but as a result of the occurrence and anticipation of a nationwide pandemic.

In any event, as a result of general principle, or the terms of the “Loss of Gross Revenue” clause, the policyholder would have to demonstrate that the relevant losses would not have been suffered but for the operation of the insured peril; and the correct counterfactual was one in which the insured peril was absent but everything else was the same. If the social distancing measures were a part of the “restrictions placed on the premises” then the correct counterfactual involved an assumption that the premises could have remained open, there would have been no legal prohibition against customers travelling to or staying at the premises, but there would still have been a COVID-19 epidemic in the country and all other businesses in the vicinity of the premises whose closure was mandated by the 26 March Regulations would have been required to stay closed. If the social distancing measures were not part of the peril insured then the correct counterfactual assumes that the premises would not have been closed from 26 March 2020 but that customers would still have been prohibited, by Regulation 6 of the 26 March Regulations, from travelling to or staying in the premises. On either analysis, there would have been no difference in outcome.

While it is not specified who may “close or place restrictions” on the premises, it is in our view clear that this must be by an authority having power to do so. We consider that this clause requires that the closure or restrictions should be mandatory: it is only such mandatory restrictions which would ordinarily be described as “closing” or being restrictions “placed on” the premises. Accordingly, we do not consider that there was any closure or restrictions placed on the relevant type of premises until 26 March. It is common ground that at that point there was.

As to when a notifiable disease may be said to have “manifested itself”, as we said in the context of QBE 1, we consider that there would be no manifestation of the disease by someone who was asymptomatic and undiagnosed. But if a person displayed the symptoms and/or was diagnosed, then there would be “manifestation” of the disease.

The argument that it is only the local cases of the disease which have causal significance in relation to the closure or placing of restrictions on the premises, is one which we have considered in relation to other wordings. In our view, on this clause, that is too narrow a construction. The phraseology used is that the closure or restriction must be as a result of the notifiable disease manifesting itself at the premises or within a 25 mile radius. The link is thus expressed in terms of whether the disease has shown or manifested itself close to the premises. Individual cases within the area are therefore treated as the demonstration of the presence of the disease at or relatively near the premises, rather than being focused on as being, in themselves, the cause of the closure or restrictions. For reasons which we have already canvassed this is not surprising. Notifiable human diseases may manifest themselves in areas which are not constrained by boundaries such as a 25 mile radius, and the response of the authorities is likely to be to the whole of whatever outbreak there is, rather than parts. In the same way as for a number of the “disease clauses”, we consider that there will be satisfaction of this requirement of the clause, if and from the time that there has been a case of the disease within the 25 mile radius, and this can be regarded as having led to (resulted in) the closure or restrictions placed on such premises on 26 March because it was part of one cause of those restrictions, which were imposed by the government as a response to a national picture which was made up of the individual local parts.

We consider that the contractual quantification machinery including the definition of Loss of Gross Revenue is intended to be applicable to heads of cover which do not involve physical damage and are to be read accordingly.

In relation to what the counterfactual should be, we consider that the position is similar to that in relation to the Hiscox “hybrid” (public authority) clauses, which we have discussed above. In our judgment the comparison which is to be made is between the actual performance of the business after the closure or restrictions which were imposed (which on our findings will not be before 26 March) and what would have been the performance assuming no restrictions and no COVID-19 which led to them (and no other response to it by the authorities or the public). We revert to this in Section G below on Causation.

We have already referred to the nature of the RSA 4 policy wording and have quoted the material terms. As set out above, one of the “Specified Causes” under Clause 2.3 is “Notifiable Diseases and Other Incidents”, and one of the “Notifiable Diseases and

The FCA identified the following as the requirements for cover under the “enforced closure” limb of the Notifiable Diseases and Other Incidents cover: (i) interruption or interference with the Insured’s business; (ii) “as a result of” “enforced closure of an Insured Location by any governmental authority or agency or a competent local authority for health reasons or concerns”; (iii) “occurring within the Vicinity of an Insured Location”.

The FCA contended that the actions of the UK Government were actions of “any governmental authority or agency”. It contended that government action from 16 March, including the advice and instructions as to social distancing, self-isolation, lockdown and restricted travel, constituted a relevant enforced closure of the premises. These measures and actions were imposed for “health reasons or concerns”, namely about COVID-19, and the “health reasons or concerns” occurred “within the Vicinity” of the insured premises, because there were concerns about everywhere in the UK. There was no need under this clause for the insured to show that anyone within a particular area had contracted COVID-19 at any particular time. “Health reasons or concerns” was a very broad concept indeed.

RSA did not dispute that the actions of the UK Government were actions of “any governmental authority or agency”. There would be “enforced closure” only if there were closure of the whole or part of the premises under legal compulsion. As Mr

In our judgment, there will only have been an “enforced closure” of premises, if all or a part of the premises was closed under legal compulsion. We agree with RSA that this would extend to closure which either is or is legally capable of being enforced. By “legally capable of being enforced” we include a case of where a governmental authority or agency or local authority directs that particular premises should be closed, and states that if they are not closed then a compulsory order for their closure will be obtained. But we consider that in that type of situation, there would have to be a clear direction by an authority which has the power to close premises that they should be shut failing which a compulsory order will be obtained. In the present case, we consider that the only “enforced closures” resulting from the actions of the government about which we have been addressed would be the closures of all or part of premises pursuant to the 21 and 26 March Regulations. To the extent that they required the closure of all or part of the premises of insureds under RSA 4, there will have been “enforced closure”. We do not, however, consider that advice or exhortations, or social distancing and stay at home instructions constitute “enforced closures”.

The 21 and 26 March Regulations were undoubtedly imposed “for health reasons or concerns”. The remaining question is whether the “enforced closure” had to be the result of “health reasons or concerns” which arose from matters occurring in the Vicinity, and if so, whether that requirement was satisfied. In our view, the correct reading of the clause is that it is the “health reasons or concerns” which need to “occur within the Vicinity”. The only other candidate is that the “enforced closure” must be within the Vicinity, but it would be unnecessary to say that. As we have already said above, we consider that the meaning of Vicinity in RSA 4, with its extended definition, could, depending on the facts, embrace a very extensive area, including possibly, in the case of a disease such as COVID-19, the whole country. If that is right, then we regard it as clear that the “health reasons or concerns” “occurred” within all relevant Vicinities. Even if Vicinity has a narrower meaning of close geographical proximity, however, we consider that there were “health reasons or concerns” which occurred in all such vicinities. This is because the clause does not require that there be particular cases of a disease in any vicinity at any given time. What is required is that there should be

“health reasons or concerns” in that vicinity. We think it clear that there were “health reasons or concerns” relating or applicable to, and thus, for the purposes of the clause, “occurring” in all vicinities, and that the enforced closures were imposed by the government “for” those “health reasons and concerns”.

In our judgment, and for reasons substantially the same as in relation to the Hiscox

Prevention of Access and similar wordings

A number of the wordings with which we are concerned provide cover where there has been a prevention or hindrance of access to or use of the premises as a consequence of government or local authority action or restriction. Since the scope of the cover provided is sometimes narrow and sometimes wider and all the wordings are different, it is necessary to consider each in turn, although in construing each wording, some common themes emerge, as will become apparent.

The Arch wording with which the Court is concerned is materially identical wording contained in three Arch policies: (i) the OGI Commercial Combined Policy; (ii) the OGI Retailers Policy; and (iii) the Powerplace (Offices and Surgeries) Policy. Of the categories of business identified by the FCA, the majority of Arch policyholders fall

The relevant BI extension in the Arch policies is the Government or Local authority Action (“GLAA”) Extension (Clause 7) which provides as follows (in the lead wording):

“We will also indemnify You in respect of reduction in Turnover and increase in cost of working as insured under this

Section resulting from…

(7)

Government or Local Authority Action

Prevention of access to The Premises due to the actions or advice of a government or local authority due to an emergency which is likely to endanger life or property. We will not indemnify You in respect of

(1)

any incident lasting less than 12 hours

(2)

any period other than the actual period when the access to The Premises was prevented

(3)

a Notifiable Human Infectious or Contagious Disease as defined in the current relevant legislation occurring at The Premises

The maximum We will pay under this Clause is £25,000, or the Business Interruption Sum Insured or limit shown in the Schedule, whichever is the lower, in respect of the total of all losses occurring during the Period of Insurance.”

In considering the effect of these provisions, it is important to identify at the outset what is the insured peril or risk and the extent to which it is common ground that the insured peril is triggered. On behalf of Arch Mr John Lockey QC submitted that, under this wording, the insured peril is prevention of access to the premises as a result of government actions or advice taken in response to an emergency likely to endanger life or property. The insured peril or risk is not as the FCA contends the “emergency”. We agree with Mr Lockey QC the insured risk is not the “emergency” but is the interruption of the business resulting from the composite peril as described by Mr Lockey QC.

Arch accepts that the COVID-19 pandemic is an “emergency…likely to endanger life” and that the government advice of 20 and 23 March and the 21 and 26 March Regulations were “the actions or advice of…government” within the meaning of the wording. Arch also accepts that, where the effect of the actions or advice was that businesses had to close or cease business, there was a prevention of access, in other words, it was not necessary that access be physically impossible or obstructed. Thus, in relation to the 20 March advice which was to the effect that nightclubs, theatres, cinemas, gyms and leisure centres should close and that pubs, bars, clubs and restaurants should not remain open for consumption of food and drink on the premises but could continue to provide take-out services, Arch accepts that there was prevention of access to businesses in Category 2 and that, in relation to businesses in Category 1 which did not previously provide take-away services, access was prevented because they could not be opened to customers without the policyholder making a fundamental change to the nature of the business. Arch also accepted that the effect of the Prime Minister’s announcement on 23 March was that there was a prevention of access to the premises of businesses in Category 4, non-essential shops, and Category 7, schools and places of worship if they closed pursuant to the advice (although as stated above, only two Arch policyholders fall into this latter category).

On behalf of the FCA, Ms Mulcahy QC noted the extent of this common ground. She also noted that Arch did not accept that there was any prevention of access in relation to Categories 3 and 5 (essential shops etc. and manufacturing and professional offices) as none of those businesses were ordered or advised to close their premises. Where pubs, cafes or restaurants in Category 1 had a pre-existing take-away business, Arch did not accept that there was any prevention of access. The premises could still be accessed for the purposes of carrying on the business even if part of the business (consumption of food and drink on the premises) was now prohibited and part of the premises (that part where food and drink were consumed on the premises) could not be used.

Ms Mulcahy QC submitted that the clause did not specify whose access must be prevented, so it could be anyone, including employees or customers, where it results in reduction of turnover or increased cost of working. She contended that the Arch approach that, if part of the insured business (for example a take-away service) could still be carried on from the premises, there was no prevention of access, was completely unrealistic. Likewise in relation to Category 4, non-essential shops, where the 26 March Regulations permitted premises to be accessed to carry on a mail order or online business, she submitted that it was unrealistic for Arch to adopt the approach that if part of the business already involved mail or online orders there was no prevention of access.

She also criticised Arch’s approach to Categories 3 and 5. In relation to Category 3, essential shops, she gave the example of a hardware store which was permitted to stay open for essential purchases, so a customer could go to the shop to buy something essential like light bulbs or batteries, but could not buy DIY products unless they were essential. Likewise, in the case of the service industries in Category 5, clients were only permitted to travel to the premises if it was essential to do so and she queried to what extent at the time of the lockdown, a visit to one’s financial adviser or accountant was essential. In each case there was a substantial reduction in turnover and she submitted that partial prevention of access to the premises for customers was still prevention of access for the purposes of this wording.

The other issue raised by the FCA in relation to the GLAA Extension concerns the date from which any coverage under the Arch policies is triggered. Ms Mulcahy QC submitted that this was 16 March when the Prime Minister’s advice was to stop nonessential contact and unnecessary travel and avoid going to pubs, clubs and theatres. That was “actions or advice” of government which the public followed so there was a prevention of access. Prevention was the whole point of the advice.

Arch’s case focused on the meaning of the words: “Prevention of access to The Premises”. It was common ground for the purposes of all the policies which we are considering where the word “access” is used that this is the means of entry or approach to the premises. Mr Lockey QC submitted that “prevention” bore its ordinary and natural meaning of something which stopped the means of access. In the context of the government actions or advice, only such actions or advice which recommended or required the closure of the premises qualified as “prevention of access” for the purposes of this provision. He submitted that the FCA’s case conflated “prevention of access” with restrictions on the use of the premises but this provision was limited to “access” not “use”. Furthermore, he submitted that the FCA case conflated “prevention” with “hindrance”. Hindrance meant that access to the premises was rendered more difficult, but prevention meant that access was stopped, effectively prohibited. Other provisions in the Arch wording drew a clear distinction between prevention and hindrance. Thus, Extension Clause (1) headed “Prevention of Access” covers reduction in turnover or increased cost of working resulting from:

“Damage to property in the vicinity of The Premises by any cause included under the Property Damage Section which hinders or prevents access to The Premises.”

Mr Lockey QC also submitted that the provision was concerned with “prevention of access to The Premises” not prevention of access to The Premises or any part thereof and with complete prevention, not partial prevention as the FCA contends, even if there could be such a concept as “partial prevention”. He submitted that the restrictions on free movement imposed by Regulation 6 of the 26 March Regulations on those who may use the premises, whether the owner of the business, the employees or customers and clients, are not directed at the means of access to the premises and do not prevent access to the premises. He accepted that the fact that prevention could be as a result of advice meant that the prevention of access did not need to be legally or practically impossible. It was sufficient if the advice recommended closure of the premises, but the recommendation had to be for total closure. He also accepted that access for a limited purpose which was not a breach of Regulations requiring closure, such as to switch off the electricity or to carry out urgent maintenance, would not preclude there being a prevention of access, since it is access to the premises for the purposes of carrying on the business which must be prevented.

He submitted that the FCA’s point that the provision does not say whose access must be prevented is a non-point, addressing the wrong question. It is access to the premises which must be prevented and social distancing advice or restrictions on free movement in the 26 March Regulations had no effect on the means of access to the premises and

In relation to the specific categories, as already noted, he said that Arch accepted that in relation to Category 1, if a pub or restaurant which had been obliged to close, started up a takeaway service during the lockdown, there was a prevention of access since that involved a fundamental change to the Business as defined in the policy schedule. However, where a pub or restaurant already provided a takeaway service as part of its Business, at least where that service was more than a de minimis part of the Business, and carried on providing that service during lockdown, Arch did not accept that there had been a prevention of access within the GLAA Extension. It is fair to point out that since, as already noted, Arch has very few policyholders in Category 1 in any event, this distinction may not be of much practical significance.

Mr Lockey QC did not accept that, as the FCA suggested, there was any inconsistency in the position of Arch in relation to Category 2, where Regulation 4(5) of the 26 March Regulations, whilst requiring theatres to close, permitted the broadcast of a performance for a remote audience, but Arch accepted that there was a prevention of access. There was no inconsistency because, as with the pub or restaurant which set up a takeaway service during lockdown, such remote performances were not part of the Business of the policyholder set out in the policy schedule.

Mr Lockey QC pointed out that 70% of Arch OGI Retailers policyholders operate essential businesses within Category 3 which were in Part 3 of Schedule 2 to the 26 March Regulations and thus permitted to remain open. He submitted that there could be no prevention of access to such businesses by reason of government action or advice where the relevant action or advice was to the effect that such businesses were expressly permitted to remain open. Furthermore, even if, contrary to Arch’s case, it was relevant to look at the position of the policyholder or employees or customers of such essential businesses, the Regulations expressly permitted all of them to work at or shop at essential businesses. The fact that some shops chose to close did not amount to a prevention of access within the GLAA Extension since such closure did not result from any government action or advice. The fact that businesses in Category 3 which stayed open had less footfall than usual or found it more expensive to operate because of social distancing advice did not amount to a prevention of access.

Category 5 businesses were also of considerable importance to Arch. Given that service businesses and manufacturers were not covered by any advice recommending closure or required to close by the Regulations, there was simply no prevention of access within the meaning of the GLAA Extension as the premises in question remained accessible throughout the lockdown.

Mr Lockey QC submitted in relation to Category 4, shops offering goods for sale or hire, that whether or not the GLAA Extension was triggered in any given case was fact sensitive. Regulation 5(1) of the 26 March Regulations required such businesses to cease business and close their premises save for the purpose of making deliveries in response to orders received online, by telephone or by post. A business which carried on a substantial proportion of that business online, for example a high street estate agent, was not required to close its premises if used for those purposes and, in that example, there would not be a prevention of access.

In relation to Category 7, Arch only has two policyholders. It accepted that the 23 March advice instructed the closure of all places of worship so that, where a place of worship was closed pursuant to that advice, there was a qualifying prevention of access. Nurseries and educational establishments were not ordered to be closed but permitted to remain open for teachers, vulnerable children and children of critical workers. If insured premises remained open for that purpose, there was no prevention of access to those premises.

We consider that Arch is right that, under the GLAA Extension, it is only where the premises closed pursuant to the 20 or 23 March government advice or the 21 or 26 March Regulations that there was a qualifying prevention of access. “Prevention” is to be contrasted with and is not synonymous with “hindrance”. In Tennants (Lancashire) Ltd v CS Wilson & Co Ltd [1917] AC 495, Lord Atkinson at 518 defined the distinction between prevention and hindrance in these terms:

““Preventing” delivery means, in my view, rendering delivery impossible; and “hindering” delivery means something less than this, namely, rendering delivery more or less difficult, but not impossible.”

Earl Loreburn at 510 said that “hindering” meant “interposing obstacles which it would be really difficult to overcome”.

The definitions of and distinction between prevention and hindrance recognised in that case have been approved and applied by the Court of Appeal in subsequent cases, culminating in Westfalische Central-Genossenschaft GmbH v Seabright Chemicals Limited (22 July 1980) (unreported). The FCA sought to persuade us not to adopt these definitions because they were inappropriate and overly strict, the cases in question being concerned with force majeure or other forms of exclusion clause. We were unimpressed by that argument. We have no sense that in any of these cases, the Courts gave to “prevention” or “hindrance” an unduly narrow meaning or anything other than a natural meaning in the particular commercial context. It seems to us that, albeit the definition of “prevention” requires some adaptation to take account of the fact that the clause we are considering concerns prevention of access to the premises as a result of government action or advice as opposed to prevention, for example in the delivery of goods, impossibility as the touchstone of prevention, as opposed to something being rendered more difficult, which is what “hindrance” connotes, is entirely apt.

Mr Lockey QC fairly and properly accepts that the impossibility in question does not need to be physical or legal given that it may result from “advice”, but we consider that, in the context of this clause, what has to result from the government action or advice is closure of the premises for the purposes of carrying on the Business as defined in the policy schedule. His concessions in relation to the pub or restaurant which started a takeaway service in lockdown or the theatre which started remote performances on the internet are well made since that entails a fundamental change from the business as described in the policy schedule and it is access to the premises for the purposes of carrying on the business described in the policy schedule which must be prevented.

However, where the pub or restaurant already had a takeaway service prior to the government actions or advice, it seems to us that the position is different. The paradigm example is that of the local restaurant which in addition to in-restaurant dining offered a takeaway collection or delivery service which may have formed a substantial part of its business. Whilst the government advice and the Regulations required the restaurant to close so far as consumption of food and drink on the premises is concerned, Regulation 4 of the 26 March Regulations did not require the restaurant to close the premises to the extent that they were used for the purposes of providing a takeaway service. It seems to us that Mr Lockey QC is right that, in such circumstances, there is not a prevention of access, as the restaurant owner policyholder and his employees are not prevented from accessing the premises for the purposes of carrying on that part of the existing Business which involves providing the takeaway service. They may be impeded or hindered in their use of the premises because they cannot operate the restaurant for in-house dining, but the Arch GLAA Extension does not insure against “hindrance” or against prevention or hindrance of “use” of the premises. If it had been objectively intended by the parties to provide cover against hindrance of access as well as prevention of access they could and would have used the words “hinders or prevents access to The Premises” as they did in Clause (1). We agree with Mr Lockey QC that the FCA’s argument that there is prevention of access even if part of the pre-existing business (like a takeaway service or a mail order part of the business) is still permitted to be carried on, conflates “prevention” with “hindrance” when the two are distinct concepts with different meanings, for the reasons we have given.

The same fallacy infects the FCA’s argument that the 16 March advice of the Prime Minister about working from home where possible, social distancing and avoiding going to pubs or clubs prevented access to insured premises. That advice did not in any sense cause a prevention of access to any premises, as can be demonstrated by the example of pubs. As we discussed with Mr Lockey QC in the course of argument, for the rest of the week after 16 March, there seems to have been something of a “boozefest” with people making the most of going to pubs before the anticipated instruction to close which came with the government advice on 20 March that pubs should close.

Likewise, we agree with Mr Lockey QC that the restrictions on free movement imposed by Regulation 6 of the 26 March Regulations did not in themselves prevent access to premises which remained open. To the extent that, in consequence of the Regulation, fewer people went to the relevant shop or office or only did so for the purposes of buying essential supplies or transacting business which could not be carried out remotely from home, it could be said that there was hindrance in the use of the premises in question for the purposes of the Business carried on there, but it is a complete misuse of language to describe that as “prevention of access to The Premises”. As we have said, that conflates prevention with hindrance and access with use.

There was only a prevention of access to the premises within the meaning of the GLAA Extension where the government actions or advice required or recommended complete, not partial, closure of the premises. Anything short of complete closure will not constitute prevention of access. Of course, as Mr Lockey QC sensibly accepted, the prevention does not have to have been physical, so that if the policyholder accessed the premises to carry out essential maintenance, that would not preclude there being a prevention of access within the meaning of the GLAA Extension, since access was still prevented for the purposes of carrying on the Business.

Turning to the various categories of Arch policyholder, our conclusion on Category 1, pubs, cafes, restaurants etc., will be apparent from what we have already said. Where those businesses closed completely in response to the 20 March advice or the 21 March Regulations, there was a qualifying prevention of access from the moment of closure. If the business then set up a takeaway service which it had not carried on before, there was still a qualifying prevention of access, since that takeaway business was fundamentally different from the Business described in the policy schedule. However, if the business had an existing takeaway service which it continued to operate from the premises, then for the reasons we have given, there was no prevention of access, save possibly, as Mr Lockey QC accepted, where the existing takeaway service was a de minimis part of the Business, which will depend on an analysis of the particular facts.

Likewise, in relation to Category 2 leisure businesses, cinemas, theatres, nightclubs, gyms and leisure centres, where those businesses closed completely in response to the 20 March advice or the 21 March Regulations, there was a qualifying prevention of access from the moment of closure. That prevention of access was not precluded if, for example, the business in question set up a remote or internet service which it had not previously provided, like the theatre which put on performances which could only be accessed remotely. As Mr Lockey QC said, that is because that “business” is fundamentally different from the business described in the policy schedule and there is no inconsistency in Arch’s position.

Category 3 is essential shops and businesses such as food retailers, pharmacies, petrol stations, banks and health centres, as listed in Part 3 of Schedule 2 to the 26 March Regulations. As the FCA accepts, they were permitted to carry on business by Regulation 5. Since none of them had to close due to government actions or advice, there is simply no qualifying prevention of access under the GLAA Extension. It is no answer for the FCA to say that there was a reduced footfall in such businesses because people only went out for essential supplies or, in Ms Mulcahy QC’s example, could only purchase essential items from the hardware store. That may amount to an impediment or hindrance in the use of the premises, but it is not in any sense a prevention of access to the premises. Where the policyholder chose to close down the business because of reduced footfall or for some other reason, that is not a qualifying prevention of access, because the closure was not due to government actions or advice, since the relevant actions or advice permitted the premises to remain open.

Category 4 is non-essential shops and businesses offering goods for sale or hire. Under Regulation 5 of the 26 March Regulations those shops and businesses had to cease trading and close their premises save for the purpose of making deliveries in response to orders received online, by telephone or by post. Where a business in this category closed completely pursuant to Regulation 5, there was a qualifying prevention of access from the moment of closure. In other cases, we agree with Mr Lockey QC that the question of whether coverage is triggered is fact sensitive. As with Category 1, if the relevant business, for example an independent bookshop, did not previously deliver pursuant to orders online or by telephone or by post, it seems to us that the setting up of such a service would not preclude there being a prevention of access, since that business would be fundamentally different from the business described in the policy schedule. In contrast, there would not be a prevention of access in the case of an estate agent which already conducted a proportion of its business online and which was permitted to continue using its premises to conduct that online business even though clients could no longer come to the premises in person. There will be variations according to the particular facts, for example if, before the closure was ordered the bookshop was already processing a handful of telephone and postal orders for elderly customers who were too infirm to come to the shop in person. Continuing that service after the Regulations came into force might be thought to be de minimis, thus not precluding there being a prevention of access, but whether there was coverage would depend on the particular facts.

Category 5 consists of businesses not listed in the 26 March Regulations such as manufacturing, accountants, lawyers, recruitment agencies and construction. These businesses were not ordered to close and were thus permitted to remain open. We agree with Mr Lockey QC that it cannot be said that there was a prevention of access within the meaning of the GLAA Extension in relation to any of those businesses. It is nothing to the point that clients or customers did not visit the offices of their accountant, lawyer or financial adviser because of the restrictions on movement imposed by Regulation 6 of the 26 March Regulations and conducted their business by Zoom or the like or by telephone. The offices were not required to close and at most there was an impediment or hindrance on the use of the premises, nothing which amounted to a prevention of access. To the extent that professional firms such as solicitors and accountants told their staff to work from home and not come into the office, that was due to the government advice to work from home where possible and due to the general concerns about the pandemic, not due to any actions or advice of government preventing access to the relevant premises. In other industries such as construction, the business in question was not ordered to close the construction site or sites and, if there were fewer workers than usual, that was due to general concern about the pandemic not due to any actions or advice of government preventing access.

There are no Arch policyholders in Category 6 and only two in Category 7. We were not told if these were places of worship or nurseries/educational establishments. However, Mr Lockey QC accepted that, in the case of places of worship, if the place of worship closed in response to the 23 March advice, there was a qualifying prevention of access. In relation to nurseries and educational establishments, they were permitted to remain open for the education and care of vulnerable children or the children of critical workers, so there cannot have been any prevention of access in relation to nurseries and schools which did remain open.

In common with the other wordings which the Court has to consider, the Arch wording contains a “trends provision” in the provisions for quantification of any business interruption loss. Like the other insurers, Arch raise a point of principle in relation to how the trends provision operates, which, in broad terms, is whether in assessing what the performance of the business would have been had the insured peril not occurred, what is “stripped out” in the counterfactual assessment is only the government restriction (in the case of Arch “actions or advice”) and its immediate effect (here the prevention of access) as the insurers contend, leaving as part of the counterfactual which would still have affected the performance of the business the coronavirus pandemic and its economic and social effect in the United Kingdom. The FCA, on the

The “trends” provision in the lead Arch wording is contained in the definition of Standard Turnover in the Business Interruption Section:

“Rate of Gross Profit and Standard Turnover may be adjusted to reflect any trends or circumstances which

(i)

affect The Business before or after the Damage

(ii)

would have affected The Business had the Damage not occurred.

The adjusted figures will represent, as near as possible, the results which would have been achieved during the same period had the Damage not occurred.”

“Damage” is defined as “Accidental loss or destruction of or damage to property used by You at The Premises for the purpose of The Business”. This definition tracked the definition of Damage in the Property Damage Section of the Arch policy: “Accidental loss or destruction of or damage to Property Insured”. The cover under the Property Damage Section was in effect all risks cover with certain exceptions, the Cover provision being: “We will indemnify You in respect of Damage occurring during the Period of Insurance at the Premises”.

“Cover” in the Business Interruption Section, under “(1) Gross Profit” provides as follows:

“In respect of each item in the Schedule We will indemnify You in respect of any interruption or interference with The Business as a result of Damage occurring during the Period of Insurance by

(1)

any cause not excluded by the terms of the Property Damage and, or Theft Sections of Your policy…”

In relation to the Arch trends provision, the FCA accepts that, because the opening words of the Extensions in the lead wording (what was described in relation to other wordings as “the stem”) state: “We will also indemnify You in respect of reduction in Turnover and increase in cost of working as insured under this Section resulting from…” (the opening words in the other Arch wordings being materially similar), the quantification machinery in the policy wording is being referred to and incorporated in the Extensions. Accordingly, unlike in relation to some of the other wordings before the Court, the FCA accepts that although the quantification machinery, including the trends provision, uses the word “Damage” which refers to physical loss and damage, the parties must have intended that the quantification machinery wording be adapted to the non-Damage situations covered by the Extensions.

As with the trends clauses or provisions in the other policy wordings which the Court is considering, the FCA’s central submission is that, given that the insured peril is a composite peril with three elements: the prevention of access (element 1) due to actions or advice of the government (element 2) due to an emergency likely to endanger life (element 3), it is necessary to remove all three elements in assessing what the position would have been “had the Damage not occurred”, in other words in order to properly assess what the position would have been had the peril insured by the relevant nonDamage extension not occurred, it is necessary to “strip out” not just the prevention of access and the actions and advice of the government, but the pandemic itself and assess what the results of the business would have been had there been no pandemic at all.

The position of Arch, as of the other insurers in relation to their trends clauses, is that whilst the emergency, here the pandemic, is a link in the chain of causation in the insured peril, it does not follow that the other effects of the emergency (other than the relevant government actions or advice), such as its economic or social effects, are excluded when examining what would be the position if the insured peril had not operated. Mr Lockey QC submitted that the FCA’s case involved the heretical proposition that any and all causes of the operation of an insured peril are to be excluded from consideration whether those causes are identified in the insuring clause or otherwise, which was not a point which could turn on whether this was a composite peril or not. He also submitted that the wording of the trends provision referring to “had the Damage not occurred” was expressly referring to “but for” causation and he adopted the general submissions on causation advanced on behalf of all the insurers by Mr Kealey QC.

Mr Lockey QC drew our attention to three practical points. First, on the agreed facts, many businesses did suffer a reduction in turnover in consequence of the pandemic, before any closure advice or the 21 and 26 March Regulations. The cause of any loss from that reduction in turnover was the emergency not the prevention of access due to government actions or advice. Second, even businesses permitted to remain open after 20 March suffered a loss of turnover compared to the same period last year because of the economic recession, the lack of consumer confidence and the restrictions on movement. Third, likewise, in the case of businesses required to close but which have since re-opened, they have also suffered a loss of turnover following the re-opening because of the economic recession and the lack of consumer confidence. He submitted that these were all issues which Arch was entitled to raise in any assessment or adjustment of the loss of any particular insured and the FCA was not entitled to declaratory relief which sought to rule out that entitlement.

We repeat the two preliminary points we made about trends clauses or provisions in relation to the RSA 3 wording in the “Disease clauses” section of the judgment. The current provision requires adjustments for any trends or circumstances which would have affected the business had the Damage not occurred. The definition of “Damage” and the Cover (1) Gross Profit provision make it clear that in the case of “standard” Business Interruption cover under this wording “Damage” is not damage in the abstract, but accidental damage occurring by any cause not excluded by the terms of the Property Damage Section of the policy, that being the relevant insured peril. As we note in the section of the judgment on Causation where we deal with the decision of Hamblen J in Orient Express, part of what we see as the fallacy in the reasoning in that case is that both the arbitrators and the judge proceeded on the basis that only the Damage in the abstract should be stripped out in assessing the counterfactual under the trends clause. However, we consider that, on a proper analysis, the insured peril in that all risks policy was not Damage in the abstract, but Damage caused by a fortuity, there the hurricane, so that what should have been stripped out in the counterfactual was not just the Damage but the Damage and the hurricane.

However, whether our analysis of Orient Express is right or not, in the present case, one has to substitute for the word “Damage” in the trends provision the insured peril under the relevant non-damage GLAA Extension so that, as manipulated, the trends provision reads as follows:

“Rate of Gross Profit and Standard Turnover may be adjusted to reflect any trends or circumstances which

(i)

affect The Business before or after the Prevention of access to The Premises due to the actions or advice of government due to an emergency which is likely to endanger life…

(ii)

would have affected The Business had the Prevention of access to The Premises due to the actions or advice of government due to an emergency which is likely to endanger life not occurred.

The adjusted figures will represent, as near as possible, the results which would have been achieved during the same period had the Prevention of access to The Premises due to the actions or advice of government due to an emergency which is likely to endanger life not occurred.”

It follows that upon the true construction of the Arch policy wording, the comparison required for the assessment or adjustment of the business interruption loss is between the performance of the business as a consequence of the prevention of access to the premises due to the actions or advice of the government due to the emergency and what the performance would have been had there been no emergency and thus no government actions or advice and no prevention of access to the premises. Of course, other trends of the business which would have operated had the insured peril not occurred will be taken into account. The example used by the parties during the course of argument is of the Michelin starred restaurant whose principal chef handed in his notice at the end of February to expire at the end of March. In all probability that business would have seen a downturn in business irrespective of the pandemic which must be brought into account.

In our judgment, this approach to the counterfactual question raised by the trends provision is not only correct on the true construction of the policy wording but accords with commercial and practical reality. We agree with Mr Edelman QC that the approach advocated by the insurers of stripping out the government restrictions etc. and their immediate effect, such as, in the case of the Arch wording, prevention of access, whilst leaving the pandemic and its economic and social effects is entirely artificial and ignores the inextricable connection between the various elements of the insured peril, both as a matter of legal analysis and as a matter of practical reality, given the nature of the pandemic emergency. For reasons elaborated in more detail in Section G of the judgment dealing with causation, we do not consider that “but for” causation, upon which insurers placed considerable reliance, requires a different result, if it is relevant at all, given our conclusion as a matter of construction of the policy wording.

That leaves the issue raised by Mr Lockey QC’s first practical point, the fact that many businesses had suffered a reduction in turnover as a consequence of the pandemic and its social and economic effects before the insured peril operated, here before the actions or advice of the government due to the emergency that businesses should close, thus preventing access. This was an issue raised generally by Mr Edelman QC in submissions on behalf of the FCA. In the Hong Kong Court of Final Appeal in New World Harbourview Hotel v Ace Insurance [2012] HKCFA 21; [2012] Lloyd’s Rep IR 537, a case concerned with a business interruption claim arising out of the 2003 SARS outbreak, Sir Anthony Mason NPJ (in a judgment with which the other members of the Court agreed) concluded that coverage was not available in respect of the period before the date when the disease became notifiable, because the cause of the loss under the policy wording had to be a notifiable disease and a disease did not become notifiable until it was required to be notified, which was on 27 March 2003. Before that date, any loss caused by SARS was caused by a loss which was not notifiable and hence not covered by the insurance.

Mr Edelman QC contended that even if that case was correct, it could be distinguished in relation to the trends clauses which we are considering because the whole emergency i.e. the coronavirus pandemic had to be taken out of the equation for the purposes of the counterfactual assessment, not just that part of the emergency which occurred after the date when the disease became notifiable. He argued that this was not to compensate the policyholder for loss prior to the disease becoming notifiable which was not permissible, but in relation to loss after notifiability, the effect of the pandemic on the policyholder’s turnover before it became notifiable should also be extracted. He submitted that if an insuring clause was contemplating insurance against a notifiable disease, it must encompass the emergence of a disease which, once the authorities get round to it, will be made notifiable.

Ingenious though this argument is, we consider that it is fallacious. Upon analysis, if it were correct, once an insured peril occurred, here the prevention of access due to government actions or advice due to the pandemic, the policyholder would in fact recover for its losses both before and after the occurrence of that insured peril, despite Mr Edelman QC’s attempts to contend that this was not the effect of his argument. In any event, in the case of the Arch policy wording, whatever the merits of the argument it is precluded by the express words of the trends provision. Any downturn in turnover before the date(s) when businesses closed pursuant to government advice or the Regulations was a trend or circumstance which affected the business before the Damage i.e. as manipulated before “the Prevention of access to The Premises due to the actions or advice of government due to an emergency which is likely to endanger life” within the meaning of (i) of the trends provision.

There are two Ecclesiastical Insurance Office plc (“EIO”) wordings to be considered by the Court. The “Parish Plus” policy under which most of the churches in the Church of England and their parish councils are insured has been taken as the lead wording for EIO Type 1.1 wordings. EIO Type 1.2 lead wording is Education Insurance insuring nurseries. The vast majority of their insureds are in Category 7, churches, charities and educational establishments, although they also have insureds in Categories 1 to 6.

There are slight differences between the two wordings. The provision under consideration in each case is one of the extensions to the standard business interruption cover. The introductory words in Type 1.1 are:

“The insurance by this section is extended to cover loss resulting from interruption of or interference with your usual activities as a result of the following…”

The introductory words in Type 1.2 are slightly different:

“The insurance by this section is extended to cover loss as insured hereunder directly resulting from interruption of or interference with the business carried on by you at the premises in consequence of the following…”

The Prevention of access provision in Type 1.2 reads as follows:

“1 Prevention of access

Access to or use of the premises being prevented or hindered by

(a)

damage to neighbouring property by any of the insured events

(b)

any action of Government Police or Local Authority due to an emergency which could endanger human life or

neighbouring property

Excluding

(i)

any restriction of use of less than four hours

(ii)

any period when access to the premises was not prevented or hindered

(iii)

closure or restriction in the use of the premises due to the order or advice of the competent local authority as a result of an occurrence of an infectious disease (or the discovery of an organism resulting in or likely to result in the occurrence of an infectious disease) food poisoning defective drains or other sanitary arrangements or vermin or pests Provided that

our liability in respect of any one occurrence shall not exceed the sum insured by the items or any limit of liability shown in the schedule”

In the Type 1.1 wording there are two separate provisions under the heading “What is covered”. Extension Clause 2 is headed: “Prevention of access-Damage” and corresponds, so far as relevant, with (a) in the Type 1.2 wording above. Extension Clause 3 is headed: “Prevention of access-Non-Damage” and corresponds with (b) in the Type 1.2 wording above. The exclusions in the Type 1.1 wording are set out under the heading “What is not covered” in terms materially identical to Type 1.2 above, but with closure or restriction in the use of the premises due to vermin being separated out as a separate exclusion (iv).

In considering the effect of exclusion (iii) (described by EIO as “the infectious disease carve-out”), which is one of the principal areas of dispute between the FCA and EIO, it is necessary to set out Extension Clause 6, the “specified disease” clause (Extension Clause 8 in the Type 1.2 wording). This is headed “Specified disease, murder, food poisoning, defective sanitation, vermin”. It begins with a list of specified diseases including cholera, dysentery, leprosy, malaria, measles, meningitis, mumps, plague, rabies, rubella, scarlet fever, tuberculosis, typhoid fever, viral hepatitis, and whooping cough. Obviously that list does not include COVID-19, or for that matter, SARS.

The clause then provides:

“What is covered

(a)

any occurrence of a specified disease being contracted by a person at the premises or within a radius of 25 miles of the premises;

(b)

any discovery of an organism at the premises likely to result in the occurrence of a specified disease being contracted by a person at the premises;

(c)

any injury or illness sustained by any person arising from or traceable to foreign or injurious matter in food or drink provided at the premises;

(d)

any accident causing defects in drains or other sanitary arrangements at the premises;

which causes restrictions in the use of the premises on the order or advice of the competent local authority. (e) any discovery of vermin at the premises; (f) murder, rape or suicide at the premises.

Special conditions

(i)

We shall only be liable for the loss arising at those premises which are directly affected by the occurrence, discovery or accident. In the event that the policy includes an extension which deems damage at other locations to be damage at the premises such extension shall not apply to this Extension.

(ii)

Indemnity period shall mean the period during which your results shall be affected in consequence of the occurrence, discovery or accident beginning with the date from which the restrictions on the premises are applied (or in the case of (f) above with the date of occurrence) and ending not later than three months thereafter.

(iii)

In respect of (e) you must obtain our consent before you restrict the use of the premises.

What is not covered

Costs incurred in the cleaning, repair, replacement, recall or checking of property.”

As Mr Edelman QC submitted, there is a tension between the various provisions. The insuring provision in the Prevention of access Extension talks of “any action of government, police or a local authority” whereas both exclusion (iii) and the specified disease clause talk about the “order or advice of the competent local authority”. The FCA’s case is that the answer to the issue of construction is not to be found, as EIO suggests, in an analysis of public health legislation, with which the FCA did not disagree. The reasonable reader of the words “competent local authority” when contrasted with “government, police or local authority” would conclude that the exclusion did not exclude orders or advice of the government. Mr Edelman QC submitted that when exclusion (iii) and (iv) are viewed together (and they are run into one provision in the Type 1.2 wording) they are all dealing with matters which are parochial and local, like food poisoning, defective drains and vermin on the premises, matters for which a local borough council would be responsible. The reference to “the occurrence of an infectious disease” is likewise to be construed as a local outbreak in relation to which such a local authority would give any orders or advice.

Mr Edelman QC accepted that it was certainly possible that the words “competent local authority” in the exclusion should bear the same meaning as in the specified disease clause but submitted that, on closer analysis, the latter clause was also focused on local outbreaks of disease. It did not refer to “notifiable” diseases and, notably, the list did not include the most recent epidemic disease of a type which could spread across the country, SARS.

On the assumption that the exclusion does not apply to the actions or advice of central government Mr Edelman QC noted what was accepted by EIO, namely that there was “an emergency which could endanger human life” from 12 March and that, although

In relation to churches, Mr Edelman QC noted that EIO accepted in its Defence that hindrance of access to or use of the premises was not just focused on the clergy and church staff but the congregation. He submitted that the Prime Minister’s advice on 16 March to stay at home, work from home where possible, only travel when necessary and avoid social contact was sufficient to trigger a qualifying hindrance of access or use from 16 March.

Mr Kealey QC submitted on behalf of EIO that the infectious disease carve-out had to be construed in the context in which it occurs in the wording and in particular in the context of the specified disease clause. He noted that most of the specified diseases listed were in the list of notifiable diseases scheduled to the 2010 Regulations. The coverage under that clause was not only in respect of an occurrence of a specified disease at the premises, but within a radius of 25 miles of the premises, which pointed away from this being a provision confined to an outbreak with which a local borough council would deal.

In relation to the infectious disease carve-out, he submitted that “what is not covered” in the Type 1.1 wording and the use of the word “excluding” in the type 1.2 wording did not mean that the provision was to be construed in the same way as an exemption clause exempting from liability, say for negligence. This provision in both its forms was an example of delineation of cover, to be construed by reference to ordinary principles of construction. Mr Kealey QC relied upon Impact Funding Solutions v AIG Europe per Lord Hodge JSC at [7] and Lord Toulson JSC at [35], followed and applied by Mr Peter Macdonald Eggers QC sitting as a Deputy High Court Judge in the Commercial Court in Crowden v QBE Insurance at [65] (which we quoted at [74] above).

Mr Kealey QC submitted that viewing the provisions as a whole, what they demonstrated was that EIO did not wish to provide wide infectious disease coverage, hence the carve-out, but did intend to provide some infectious disease cover limited to the list in the specified disease clause and to the circumscriptions in that clause such as the 25 mile radius limit. The provisions were to be construed against the relevant legal background, here the relevant public health legislation. It was not necessary for that legal background to be actually known by the insurer or by the average insured. It was sufficient if it was reasonably available to the parties. He relied upon Arnold v Britton at [21], Doleman v Shaw [2009] EWCA Civ 279; [2009] Bus LR 1175 at [56] per Elias LJ and Lewison on the Interpretation of Contracts (6^th ed) §4.06.

Mr Kealey QC submitted that the key statute conferring powers on the government to handle serious civil contingencies is the Civil Contingencies Act 2004 which provides a framework for action in the case of an emergency defined as “an event or situation which threatens serious damage to human welfare in a place in the United Kingdom”. The Act provides for the making of emergency regulations including in relation to public health emergencies. Separate powers in relation to public health protection are provided by the 1984 Act as amended by the Health and Social Care Act 2008. Under the 1984 Act as amended the administration of public health functions and responsibilities is largely in the hands of a “local authority” defined as district councils, county councils, London borough councils, the Common Council of the City of London, the Sub-Treasurer of the Inner Temple and the Under Treasurer of the Middle Temple. Public health protection is dealt with by Part 2A, added by the 2008 Act. Section 45C of the 1984 Act as amended enables the appropriate minister to make regulations in relation to preventing or protecting against the spread of infection or contamination in England and Wales, including the imposition of restrictions on persons and premises. The 21 and 26 March Regulations were made under that section and sections 45F and 45P.

Section 45D imposes restrictions on the power to make regulations under section 45C, including that a restriction or requirement has to be proportionate and can only be imposed by virtue of a decision taken under the regulations by “the appropriate Minister, a local authority or other person.” Several sections of the Act give enforcement powers in relation to matters such as restricting movement of infected or contaminated persons or closure of contaminated premises to a justice of the peace. Section 67 gives a right of appeal to the Crown Court against any order, determination or other decision of a Magistrates’ Court under the Act.

Amongst the key regulations made under the 1984 Act are the 2010 Regulations imposing a duty on doctors to notify the proper officer of the relevant local authority where the doctor has reasonable grounds for suspecting that a patient may have a notifiable disease, notifiable diseases being listed in Schedule 1 to the Regulations. COVID-19 was added to the list on 5 March. The Health Protection (Local Authority Powers) Regulations 2010 give the local authority powers, inter alia, to require a parent to keep a child away from school when satisfied the child is or may be infected and to require a person or group of persons to do or refrain from doing anything for the purpose of preventing, protecting against, controlling or providing a public health response to the incidence or spread of infection or contamination.

In summary, Mr Kealey QC submitted that whilst day-to-day responsibility for a number of aspects of public health protection rests on local authorities, the power to make the most intrusive and invasive orders rests not on local authorities but with magistrates’ courts. Central government has always had the overarching power to take action to protect against the incidence or spread of infection or contamination, including powers to make general provisions, contingent provisions or specific provisions in response to particular sets of circumstances, nationally and locally. An example of the latter was the Leicester Regulations. It has never been the case that the only authority competent to act in relation to public health protection is the local authority. Central government has always been an authority with competence to act in relation to local and national public health matters.

He submitted that, given this legislative framework, which the parties are to be taken to have had reasonably available to them at the time the contract was made, what the policy wording meant by “competent local authority” in both places where that phrase was used was whichever authority was competent to take the relevant action or issue the relevant order or advice in the locality, including central government, magistrates’ courts, the Crown Court on appeal from Magistrates and even potentially the High Court on appeal by way of case stated.

In relation to the specified disease clause, as already noted, a substantial number of the diseases listed are notifiable diseases on the list in Schedule 1 to the 2010 Regulations. Mr Kealey QC agreed that the list in the clause did not include SARS, no doubt because EIO did not want to insure against SARS, but there was nothing to confine the clause to local outbreaks of the diseases specified. A number of the diseases would have the capacity to spread more widely, like the plague and in that context, central government would be expected to intervene and would be the competent local authority to impose any restrictions. He submitted that here the closure of or restriction on the use of the premises was due to government order or advice as a result of the occurrence of COVID-19, an infectious disease, so the infectious disease carve-out applied and there was no cover under either of the EIO wordings.

If, contrary to that case, there was coverage in principle, we have already alluded to the EIO case that there was no hindrance of access and use before 23 March. Contrary to the FCA case, the Prime Minister’s advice on 16 March did not hinder access to or use of churches. It did not mention churches and EIO submitted that it was entirely reasonable to suppose that reasonable church goers would not have interpreted what he said as requiring them not to go or discouraging them from going to church. Furthermore, the mass gatherings which the advice told people should not take place were gatherings such as sporting events where emergency services would have been deployed, not church services. In relation to schools, whilst the FCA relied on the government announcement on 18 March that schools would close on 20 March, EIO relied upon a Department of Education press release on 18 March that schools would close on 23 March other than for children of key workers and vulnerable children. Accordingly, Mr Kealey QC submitted that the Court should go no further than making a declaration that access to or use of the relevant premises was hindered by action of government due to an emergency which could endanger human life from 23 March.

In relation to care homes which EIO also insures, it was submitted that there was no closure or restriction on use. So far as other categories of EIO insureds are concerned, the position is fact-sensitive and the Court could not make any declarations about those.

Whilst the policy wordings are not exactly a model of clarity in terms of drafting, we agree with Mr Kealey QC that the question of the construction of the infectious disease carve-out has to be approached on the basis that it is a provision delineating the scope of cover, not in any sense an exemption clause. The applicable principles are as summarised by the judge in Crowden and there is no place for the application of the principle of contra proferentem, to the extent that principle has any application in the modern law of construction of contracts.

We also agree with Mr Kealey QC that the phrase “competent local authority” must mean the same in the carve-out as it does in the specified disease clause. In the latter, given the legislative background which can legitimately be taken into account in construing the phrase, we consider it inherently unlikely that the parties intended the scope of cover provided by the clause to be limited to local outbreaks of a specified disease for which only the local district council or other local authority as defined in section 1 of the 1984 Act issues orders or advice. A number of the specified diseases are, as Mr Kealey QC pointed out, on the list of notifiable diseases under the 2010 Regulations, no doubt at least in part because of their capacity to lead to more widespread infection or contagion than in a particular locality. Many of those diseases, at least historically, have been widespread, not just the plague or diphtheria or tuberculosis but in more recent times, measles, mumps and rubella.

The point can be tested by reference to the Leicester Regulations. If one assumes for the purpose of the argument that COVID-19 was a specified disease in the list, if “competent local authority” had the narrow meaning of the local district council or other local authority as defined in section 1 of the 1984 Act for which the FCA contends, then there would not be cover in relation to the restrictions on the use of premises imposed by central government in bringing into force the Leicester Regulations. However, there would be cover if the same restrictions were imposed by Leicester City Council or if the Secretary of State asked the Council to impose the restrictions. The narrow meaning for which the FCA contends leads to an artificial and illogical result. In our judgment, Mr Kealey QC is right that the phrase “competent local authority” means whichever authority is competent to impose the relevant restrictions in the locality on the use of the premises, including central government.

Given that the phrase has that meaning in the specified disease clause, as we have said it must have the same meaning in the infectious disease carve-out. The order or advice contained in the 20 and 23 March government advice and in the 21 and 26 March Regulations was the order or advice of the competent local authority, and was as a result of an occurrence (in fact many occurrences) of an infectious disease. Accordingly, the carve-out applies and there is no cover under either EIO wording in respect of the closure of or restriction in the use of the premises.

If, contrary to that conclusion, there is coverage in principle under either EIO wording, we agree with Mr Kealey QC that access to or use of the relevant premises (whether churches or schools) was hindered by action of government due to an emergency which could endanger human life from 23 March and not before. In particular, we do not consider that the Prime Minister’s advice on 16 March can be said to have hindered access to or use of churches or schools.

Since we have concluded that EIO is entitled to deny cover on the basis of the infectious disease carve-out, it is not strictly necessary to consider causation and the counterfactual, but we do so in case this case goes further. The Parish Plus policy wording does not contain a trends clause as such but contains wording in the “Loss of Income” clause in the Basis of settlement section in these terms:

“1.

Loss of income

We will pay the difference between the income you would have received during the indemnity period if there had been no damage and the income you actually received during that period…”

Other EIO wordings, including Type 1.2 contain a trends clause in more traditional form, providing (for example) that the standard revenue or turnover is to be adjusted:

“as may be necessary to provide for the trend of the business and any other circumstances affecting the business either before or after the damage or which would have affected the business had the damage not occurred so that the adjusted figures represent as near as possible the results which would have been obtained during the relative period after the damage had the damage not occurred.”

The Parish Plus wording simply defines “Damage” as “physical loss, destruction or damage” but Type 1.2 has a more elaborate definition which says Damage means as defined under “Cover” of the Business Interruption section. The Cover section provides that EIO will indemnify the insured “if any building or other property used by you at the premises specified in the schedule for the purpose of the business is destroyed or damaged during the period of insurance by any of the insured events (destruction or damage so caused being termed damage) and the business carried on by you at the premises is in consequence interrupted or interfered with”. Although the loss of income and trends provisions in the wordings do not refer to the perils insured under the Extensions, Mr Kealey QC submitted that the basis of settlement and trends clauses are not to be confined to the material damage part of the business interruption insurance and that the word “damage” has to be manipulated so that the quantification machinery applies to the Extensions by reading it as “damage or insured peril”. This is accepted by the FCA in relation to the Type 1.2 wording in its Reply, but not accepted in relation to Type 1.1 wording. We agree with Mr Kealey QC that the wordings should all be construed so that references to “damage” are to “damage or insured peril”, given that the stem wording of the Extensions talks about the extension of the insurance provided under the business interruption section, the clear intention being that the cover under the Extensions is provided on the same basis as cover under the rest of the section.

Mr Kealey QC submitted that the loss of income and trends provisions clearly reflected the parties’ agreement that “but for” causation has to be applied in assessing recoverable loss, as they require the loss to be calculated on the counterfactual of what would have happened: “if there had been no [insured peril]” or “had the [insured peril] not occurred”.

Mr Kealey QC submitted that the essence of the insured peril under the Prevention of access-Non-Damage Extension was access prevention etc. where that has occurred by the specified reason, here by reason of action of government in specified circumstances, here due to an emergency which could endanger life. The remainder of the clause after the essence of the insured peril, access prevention etc. serves to define and restrict the type of access prevention which qualifies. The insured peril was not the emergency or the pandemic or any action of government other than the action which caused access prevention etc. to the premises. He submitted that a distinction had to be drawn between the insured peril and matters which qualify and define the insured peril. Thus, “any action of government” qualified what type of prevention or hindrance of access or use of the premises is covered and “emergency which could endanger human life” further defined and qualified what type of government action would trigger coverage.

Accordingly, Mr Kealey QC submitted that in assessing the counterfactual for the purposes of quantifying any loss, “but for” causation requires that all that is reversed out is the access to or use of the premises being prevented or hindered. In particular, the emergency endangering human life, the pandemic, is not to be reversed out. It is not an insured peril in its own right, but its only function is to qualify and define the type of government action which triggers the cover. It is to be assumed that there continued to be an emergency endangering human life from COVID-19. Mr Kealey QC submitted that if the counterfactual reversed out not just the access prevention of the qualifying type but also the emergency itself, EIO would be made insurer of all business interruption losses caused by the pandemic, which would expand the cover being provided beyond recognition.

On behalf of the FCA, Mr Edelman QC criticised this counterfactual. What the insurers were contemplating was an emergency sufficient to generate the government action since it is only if there is an emergency of sufficient seriousness to provoke the government into action that it will act, but their counterfactual took out that government action and assumed that the government did not react to the emergency contemplated by the cause. He described this variously as “cloud cuckoo land” or “salami slicing”. He submitted that the insurers’ approach was incorrect because the insured peril was the combination of events, the access or use being prevented or hindered by government action which in turn is due to the emergency. All the elements are required for cover to be triggered.

Ingenious though Mr Kealey QC’s argument is, it seems to us that it suffers from the fundamental fallacy of mischaracterising the insured peril. Contrary to his argument the insured peril is not prevention or hindrance of access to or use of the premises with the government action and the emergency simply qualifying or defining that peril. On a proper analysis of these insuring clauses, as with others which the Court is considering, the insured peril is a composite one involving three interconnected elements: (i) prevention or hindrance of access to or use of the premises (ii) by any action of government (iii) due to an emergency which could endanger human life. For cover to be triggered that composite peril must have caused the interruption of or interference with the business.

In assessing the loss suffered by the policyholder where cover is triggered two related preliminary points similar to the points we made above in relation to the RSA 3 wording apply here. First, the “loss of income” and trends provisions in the wordings are not delineating cover, but are part of the machinery for calculating the business interruption loss on the basis that there is a qualifying prevention of access. It would seem contrary to principle, unless the policy wording so requires, for the loss to be limited by one or more of the elements of the insured peril still being included in the counterfactual assessment of what the position would have been in if the insured peril had not occurred.

Second, the object of the loss of income and trends provisions in these wordings is to put the insured in the same position as it would have been in if the insured peril had not occurred. This can be seen clearly once the literal meaning of “damage” is manipulated so that it means “damage or insured peril”. Again, once the true nature of the insured peril is identified, it is clear that the counterfactual requires not only the prevention or hindrance to be stripped out but the government action and the emergency, since the insured peril comprises all three elements. “But for” causation upon which the insurers, through Mr Kealey QC, placed so much emphasis, does not dictate a different answer. The correct answer to the question it poses “but for what?” is “but for the insured peril”, which is the composite one we have identified.

That the counterfactual is one where all the elements of the insured peril are removed, not just the prevention or hindrance but the government action and the emergency and its economic and social effects, is not only the correct construction of the policy wordings, but accords with commercial and practical reality. As noted above in respect of Arch, the insurers’ approach is an artificial one which ignores the inextricable connection between the various elements of the insured peril, both as a matter of legal analysis and as a matter of practical reality, given the nature of the pandemic emergency.

Thus, in the example of the church collection which we discussed with counsel during the course of argument, if, before the insured peril was triggered on 23 March (assuming for these purposes that the infectious disease carve-out does not relieve EIO from liability), the collection had gone down to 80% of what it was in the equivalent period last year, that loss of income before the insured peril was triggered is not recoverable under the Parish Plus policy, for the same reasons as we gave in relation to the Arch wording. If we assume that between 23 March and the date in July when churches were allowed to reopen, the collection had gone down to 10% of what it was in the equivalent period last year, the 70% loss of income would in principle be recoverable under the policy. It would be no answer for insurers to argue that the real cause of the loss of income was the economic effect of the pandemic rather than the closure of the church on the advice of the government, as the prevention or hindrance of access by the government action is inextricably bound up with the emergency and its social and economic effects.

We have already set out details of the Hiscox wordings and policyholders in dealing with the hybrid clauses in these wordings. As we noted there, the wording in Hiscox 1 with most policyholders is a Professions BI wording used to insure professional service businesses including small accountancy firms and solicitors, virtually all of which, according to Hiscox, are in Category 5. All the Hiscox 1 wordings contain a nondamage denial of access (“NDDA”) clause. The Hiscox 2 group consists of twenty three wordings. The wording with the largest number of policyholders is again a Professions BI wording the vast majority of which are issued to professional services businesses in Category 5. Five of the Hiscox 2 wordings have a NDDA clause. None of the Hiscox 3 wordings contains a NDDA clause. There are four Hiscox 4 wordings. In the region of three quarters of policyholders in this group are insured under the lead wording, a Retail BI wording. One Hiscox 4 wording (but not the lead wording) contains a NDDA clause.

The most common form of NDDA clause is in these terms:

“We will insure you for your financial losses and other items specified in the schedule, resulting solely and directly from an interruption to your activities caused by:… Non-damage denial of access

an incident occurring during the period of the insurance within a one mile radius of the insured premises which results in a denial of access or hindrance in access to the insured premises, imposed by any civil or statutory authority or by order of the government or any public authority, for more than 24 consecutive hours”

Two of the Hiscox 2 wordings have a variant, a NDDA clause which states: “within the vicinity” and only refers to denial or hindrance in access being imposed by the police or other statutory authority. Thus it reads:

“An incident during the period of insurance within the vicinity of the business premises which results in a denial of or hindrance in access to the business premises imposed by the police or other statutory authority.”

The FCA contends that “incident” is not to be limited in terms of duration or geographical scale in the manner suggested by Hiscox. It should be given its ordinary, natural meaning which is an occurrence or an event. The incident can be local, regional or national and provided that its occurrence encompasses the applicable one mile radius there is cover. Mr Edelman QC gave the example of the Great Fire of London which could be described as an incident occurring on a wide scale and if access to the insured’s premises in the suburbs was hindered by authority response to the incident, there would be cover, provided that the fire encroached within one mile of the insured’s premises. The fact that the preponderance of the fire may have been outside the one mile zone would be irrelevant. Mr Edelman QC submitted that the fact that the clause required the denial of access or hindrance in access to last for more than 24 consecutive hours pointed to an incident which continued for some considerable time, not one of short temporal duration. The FCA’s pleaded case was that the pandemic was both an “emergency” for the purposes of certain other wordings and an “incident” for the purposes of the Hiscox wordings.

In relation to the alternative wording in the Hiscox 2 policies, Mr Edelman QC argued that “vicinity” should be given the same meaning as spelt out in the RSA 4 wording, an area or areas surrounding or adjacent to the insured premises in which events that occur would be reasonably expected to have an impact on the insured or its business. In the case of COVID-19 the vicinity of the insured premises could comprise the entire country.

In the alternative, the FCA argues that the requirement of “an incident” is satisfied by the occurrence of COVID-19 within the one mile radius or in the vicinity of the premises (if “vicinity” was to be given a narrow meaning), in other words upon proof

The clause is wider than others the Court has to consider since it provides cover not just when there is a denial of access but a hindrance in access. Mr Edelman QC submitted that this was not limited to access by the insured (which Hiscox accepts, since the preceding clause in a number of the Hiscox wordings, the denial of access clause, refers to “your access” but the NDDA clause does not). He submitted that access to insured premises by customers or clients was hindered by the restrictions on movement contained in Regulation 6 of the 26 March Regulations. Hiscox’s argument that as long as there was no physical or legal impediment to access, there was no hindrance, was wrong. The Regulations enacted by the government deterred or prevented customers from visiting premises other than within the permitted exceptions, which was sufficient to constitute “hindrance in access”.

In relation to the related point made by Hiscox that the stem wording made it clear that the cover was for financial losses resulting solely and directly from an interruption to the insured’s activities, not an interference with those activities, Mr Edelman QC submitted that Hiscox was wrong to contend that “interruption” meant complete cessation. He referred to the “Loss of attraction” clause in a number of the Hiscox wordings:

“insured damage in the vicinity of the insured premises or any fundraising event resulting in a shortfall in your expected income or gross profit for more than two consecutive days”

Mr Edelman QC submitted that it was clear from this cover of a shortfall income that reduction in customer footfall was covered, which would occur when there was interference with the business, rather than complete cessation.

On behalf of Hiscox, Mr Gaisman QC described the FCA’s case on the NDDA clause variously as “feeble” and “hopeless”, noting that the Hiscox Interveners were not running any case relying on the clause, notwithstanding that several of its policyholders had policies which contained the clause. Mr Gaisman QC agreed that an “incident” is to be equated with “an occurrence” and “an event” and submitted that Lord Mustill’s dictum in Axa Reinsurance v Field at 1035 that: “In ordinary speech, an event is something which happens at a particular time, at a particular place, in a particular way” was equally applicable to “an incident”. His primary submission was that, applying that meaning, the FCA could not satisfy the requirement for “an incident”. In context, what the clause was focusing on was something occurring within a one mile radius of the premises or in the vicinity of the premises, some local incident such as a burst water main or a gas leak or a traffic accident or a crime, which caused the authorities to impose or order a denial of or hindrance in access. A nationwide pandemic such as COVID-19 was not “an incident” within the meaning of the clause. It was too geographically dispersed, too prolonged, too variegated and too non-specific to qualify as “an incident”. Likewise, he submitted that the presence of someone within the one mile radius or in the vicinity of the premises who had COVID-19 could not possibly be described as “an incident”. The person might come and go without knowing he had the disease and people might not know he was infected. Such an undetectable happening could not be “an incident”.

He submitted that it was no answer for the FCA to say the pandemic was everywhere in the UK, giving the example of the room from which he was making his submissions. On the assumption that the room contained neither a sufferer from COVID-19 nor the virus itself, saying that the pandemic was everywhere in the UK did not mean that it was in the room. The pandemic is everywhere but nowhere in particular. The requirement in the clause that an incident must occur within a one mile radius is a stipulation of a specific geographical condition which must be satisfied. The word “within” connoted that the incident must be within the circle, not outside it. That requirement of proximity was inserted to ensure that only local incidents were covered.

In relation to the words in the NDDA clause “a denial of access or hindrance in access”, Mr Gaisman QC submitted that what was intended by these words was a general denial or hindrance applicable to all members of the public. In the paradigm case of a bomb scare or a gas leak, none of this caused any difficulty: all members of the public would be denied or hindered in access. It was not enough to trigger the clause that a certain class of people cannot get access, the denial or hindrance has to be to the public as a whole.

He submitted that the restrictions on movement imposed by Regulation 6 of the 26 March Regulations did not amount to a denial of or hindrance in access to particular business premises, citing the example of a man in Manchester who could only leave his home for permitted purposes. He submitted that it was a misuse of language to say that he had been denied access or hindered in access to every business he wanted to visit. He might have been denied or hindered in the use of the premises but “access” and “use” are different concepts. Inability to use insured premises is covered by the Hiscox public authorities clause (which we have considered above in the previous section of this judgment dealing with the so-called “hybrid clauses”) not by the NDDA clause.

In relation to the words in the NDDA clause: “imposed by any civil or statutory authority or by order of the government or any public authority”, Mr Gaisman QC submitted that “imposed” and “by order” connoted that which is mandatory, something which has the force of law, in this instance the 21 and 26 March Regulations and nothing else. The various pieces of government advice, including as to social distancing, however strongly worded were not mandatory and were neither imposed nor by order.

Mr Gaisman QC submitted that the further insurmountable problem which the FCA faced concerned causation. The wording of the NDDA clause made it clear by use of the words “results in” which were words of proximate causation that it was “an incident occurring… within a one mile radius of the insured premises” which must have caused the denial of or hindrance in access imposed by the government. He submitted that the cause of the imposition of the Regulations (assuming they denied access or hindered access to the insured premises) was the national pandemic emergency, but for the reasons he had already given, that was incapable of being “an incident… within a one mile radius of the insured premises”. On the FCA’s alternative case of “an incident within a one mile radius of the insured premises” being the presence of a person infected with COVID-19 within the one mile radius, even if that was capable of amounting to an incident, which it was not, it could not possibly be said to have been causative of the restrictions imposed by the government.

In our judgment, the FCA’s entire case on the NDDA clause founders on the requirement for “an incident”. We agree with Mr Gaisman QC that this word should be given the same essential meaning as “an event”: something which happens at a particular time, at a particular place, in a particular way. The further geographical restriction that the incident occurs “within a one mile radius of the insured premises” or

“within the vicinity of the insured premises” seems to us to confirm that this clause is intended to cover local incidents, of which the paradigm examples are a bomb scare or a gas leak or a traffic accident. The fact that there is only cover if the denial of or hindrance in access is for more than 24 hours does not detract from the fact that the incident is an event which happens locally. In those examples of the bomb scare or gas leak or traffic accident, it would be quite common for access to be restricted by order of the government or police or other authority for more than 24 hours.

Contrary to the FCA’s pleaded case “an incident” is not synonymous with “emergency” or “danger”. The incident could be said in any given case to cause an emergency or danger to arise, but it is not synonymous with it. Whilst, in other wordings than the Hiscox wordings we are considering, the COVID-19 pandemic can properly be described as “an emergency which is likely to endanger life” (as in the Arch and EIO wordings considered above), it is a misuse of language both generally and on the correct construction of the NDDA clause to describe the pandemic as “an incident” let alone “an incident occurring… within a one mile radius of the insured premises”. It may amount to a state of affairs but it is, as Mr Gaisman QC put it, too geographically dispersed, variegated, prolonged and non-specific to amount to an incident. We consider that Mr Gaisman QC’s analysis is correct. Saying that the pandemic is everywhere does not mean that it is present at particular insured premises and certainly does not mean that it amounts to “an incident occurring… within a one mile radius of the insured premises”. It is no answer for the FCA to say that there is an incident if someone with COVID-19 is present within the one mile radius. As Mr Gaisman QC said, that person might or might not know that he or she had COVID-19 and, in any event, it is a misnomer to describe the presence of someone in the radius with the disease as “an incident” for the purposes of the clause.

We consider that this is a narrow, localised cover intended to insure events or incidents which occur within the one mile radius. The localised nature of the cover is even more marked in the case of the alternative wording requiring the incident to occur “in the vicinity of the insured premises”. As Mr Kealey QC submitted on behalf of MSA, the vicinity of premises is an elastic concept, but it does connote neighbourhood. In a given case it might encompass a greater area than the one mile radius, in another case it might encompass a smaller area. However, other than possibly in the case of wordings which give “Vicinity” a specific defined and extended meaning (as in RSA 4), it cannot mean the entire UK simply because, even assuming that the pandemic could be described as “an incident”, it would be reasonably expected to have an impact on the insured or its business.

It follows that there is no cover under the NDDA clause in respect of business interruption losses caused by the restrictions imposed by the government in response to the national pandemic. In those circumstances it is not strictly necessary to consider the other aspects of the clause but we do so briefly in case this matter goes further. So far as concerns the requirement that the restriction on access is: “imposed by any civil or statutory authority or by order of the government or any public authority”, as we said in relation to “imposed” in the section of the judgment dealing with Hiscox hybrid clauses, we agree with Mr Gaisman QC that both “imposed” and “by order” convey a restriction which is mandatory, not merely advisory, in other words a restriction which has the force of law. Only the restrictions imposed by the 21 and 26 March Regulations qualify. As Mr Gaisman QC pointed out, the third paragraph of the preamble to each set of Regulations refers expressly to the restrictions and requirements imposed by the Regulations:

“The Secretary of State considers that the restrictions and requirements imposed by these Regulations are proportionate to what they seek to achieve, which is a public health response to that threat.”

It follows that government advice or recommendations, whether before or after either set of Regulations came into effect, cannot have imposed or ordered a denial of access or a hindrance in access, however strongly worded the advice or recommendations were, since they did not have the force of law.

There would be much force in Mr Gaisman QC’s submission that the word “interruption” in the stem wording should not encompass “interference” as opposed to complete cessation of the business, since had it been intended that interference with the business would be covered, the wording could and would have said so, as other wordings we have to consider do. However, as we say above in respect of the Hiscox hybrid clauses, it seems to us clear from a number of the clauses which follow the stem wording, at least in the Hiscox 1 lead wording, that “interruption” is intended to mean “business interruption” generally, including disruption and interference.

One such clause is Clause 5, the loss of attraction provision which we have quoted above. Since that contemplates a shortfall in expected income, it is clearly encompassing not just the case where the insured damage in the vicinity leads to the insured premises closing because of lack of customers but the case where the premises remain open but because of the insured damage in the vicinity, there are fewer customers. Hiscox sought to address the obvious difficulty which this provision places in the way of its argument that “interruption” means total cessation by submitting that the loss of attraction provision should have been included as one of the additional covers, Clauses 17 to 19, which are not contingent on interruption at all, such as “Cancellation and abandonment”. However, that would require rewriting the wording. The loss of attraction provision is clearly included as one of the covers contingent on “interruption” in the stem wording (as it is also in the lead Hiscox 4 wording).

Mr Gaisman QC submitted that this was an extreme example of “the tail wagging the dog” but, as the Court pointed out during the course of argument, the loss of attraction provision is not an isolated example of the wording only really making sense if the word “interruption” encompasses disruption and not just complete cessation. Clauses 7 and 9 provide cover in respect of “specified customers” and “specified suppliers” in these terms:

“insured damage arising at the premises of any specified customer”

“insured damage arising at the premises of any specified supplier”

The expressions “specified customer” and “specified supplier” are defined as: “Any direct customer [supplier] of yours operating and based at the address individually stated in the Business interruption section of the schedule.”

It seems to us that these provisions contemplate that the insured can specify particular customers or suppliers of particular importance to the business and obtain cover if insured damage at that customer’s or supplier’s premises disrupts the insured’s business. The only way in which those provisions could be applicable if “interruption” meant total cessation would be if the customer or supplier were, in effect, the sole supplier or customer upon whom the business was totally dependent for supply or custom. This was in effect the construction for which Mr Gaisman QC contended. Whilst such an extreme case would be covered, it is inherently unlikely in our judgment that the intention was to limit the cover to one single specified supplier or customer.

The insuring clauses also include Unspecified customers and Unspecified suppliers provisions. It is only necessary to quote the former, as the latter is in materially similar form:

“insured damage other than loss or damage caused by flood or earth movement, arising at the premises of any of your direct customers operating and based in the European Union (including in the United Kingdom or Gibraltar) other than any specified customer”

Any suggestion that because “interruption” has to mean complete cessation and does not encompass disruption, this cover is limited to the situation where either the insured only has one unspecified customer or supplier and due to damage at its premises has to close the insured business or the insured is so dependent upon one unspecified customer or supplier that although it has other customers or suppliers it has to close the insured premises, would place an unwarranted limitation on the cover. Once again, the use of the word “any” points to these provisions covering all situations where damage at the premises of a customer or supplier disrupts the insured’s business.

Accordingly, in our judgment “interruption” in the stem wording in Hiscox 1 and Hiscox 4 is to be interpreted as not being limited to complete cessation but as including disruption to or interference with the business. The question remains as to the meaning and scope of “a denial of access or hindrance in access to the insured premises” in the NDDA clause. As we have already noted, it is only where the denial or hindrance in question is imposed by the 21 or 26 March Regulations that this provision could respond, even if the FCA were correct about the meaning of “an incident”. Although Mr Gaisman QC’s primary case was that the Regulations did not impose any denial or hindrance of access rather than use, it seems to us that this is too absolutist a position. In relation to businesses required to close by either set of Regulations, that is businesses in Categories 1, 2, 4, 6 and 7, it seems to us that there was a denial of access. Furthermore, in the case of businesses where under the Regulations people were only allowed to access the premises for limited purposes, such as to run a takeaway service in a pub or restaurant and to collect the takeaway food or drink from that pub or restaurant, whilst there was not a denial of access in those circumstances, we consider that there was a hindrance in access, since no-one was allowed to access those parts of the premises which would normally be used for in-house dining or drinking. In the case of such businesses there would have been an “interruption to [the insured’s] activities” on the basis that interruption also encompasses disruption for the reasons we have given.

So far as businesses which were either allowed to remain open (essential shops and businesses in Category 3) or about which the Regulations were silent (professional service businesses and manufacturing in Category 5), we agree with Mr Gaisman QC that it cannot be said that there was any denial of or hindrance in access to such premises imposed by or by order of the government. We also agree with him that Regulation 6 imposing restrictions on movement other than for permitted purposes did not impose any denial of or hindrance in access to insured premises, as opposed to use of such premises. In the case of people who could and did work from home (as in the case of many professional people) it seems to us a misuse of language to say that they were denied access to or hindered in their access to their offices. To the extent that they needed to access their offices to obtain files etc., they could freely have accessed the premises. At most there was a restriction on use of the offices because they could work from home, but since the Regulations were silent about businesses in Category 5, it cannot be said that any such restriction on use was imposed by or by order of the government.

The FCA relied upon the so-called “2 metre social distancing rule” to argue that there was hindrance in access, for example to essential shops and supermarkets because only so many people were allowed in the shop at any one time and other people had to queue to get in, two metres apart. However, as Mr Gaisman QC pointed out the 2 metre rule is not contained in the Regulations and is only government advice or guidance, so that it cannot be said to have been imposed by government or any relevant authority within the meaning of the NDDA clause.

Given our conclusion that there is no cover under the NDDA clause in respect of business interruption losses caused by the restrictions imposed by the government in response to the national pandemic because the FCA cannot establish in any given case that there was “an incident occurring… within a one mile radius of the insured premises”, we can deal with the issue of causation briefly. The cause of the imposition of the restrictions was the national pandemic which cannot be described as “an incident”. The position as regards causation might be different if the NDDA clause had referred to an emergency endangering human life, as in the case of the Arch and EIO wordings, but it did not and, even on the basis that Mr Edelman QC is correct about the counterfactual analysis in relation to those other wordings, that analysis cannot overcome the insurmountable difficulty the FCA faces that it cannot establish that the restrictions on access imposed were caused by “an incident”.

The position under the FCA alternative case is no better as regards causation. Even if the presence of a person with COVID-19 within the radius or in the vicinity could be said to be “an incident” which it cannot, for the reasons we have given, it simply cannot be said that any such localised incident of the disease caused the imposition by the government of the restrictions.

As noted above, MSA 1 is a Commercial Combined Policy purchased predominantly by policyholders in Categories 3 and 5. It includes as additional cover to the Business Interruption Section 6 of the policy at Clause 1 the action of competent authorities (“AOCA”) clause which provides:

“We will pay you for:

1.

Action of competent authorities

loss resulting from interruption or interference with the business following action by the police or other competent local, civil or military authority following a danger or disturbance in the vicinity of the premises where access will be prevented provided always that there will be no liability under this additional cover for loss resulting from interruption of the business during the first 24 hours of the indemnity period.

We will not pay more than £50,000 under this additional cover for a period not exceeding 12 weeks.”

As noted above, MSA 2 was issued in several forms to suit Retail (including Categories 3 and 4) Leisure (including Categories 1 and 2) and Office and Surgery (including Categories 3 and 5). It included at Clause 8 of Section A-Automatic cover, a Prevention of access-non damage clause in these terms: “8. Prevention of access – non damage

MSA 3 was specialist cover for forges which were within Category 5 businesses never required to close pursuant to any government legislation or regulations. The policy included a Prevention of access clause in these terms:

“1)

Prevention of access

Loss resulting from interruption of or interference with your business because of

a)

damage as insured by this section resulting from damage to property in the vicinity of the premises which will prevent or hinder the use of the premises or access thereto whether your property at the premises will be damaged or not; and

b)

action by a competent public authority following threat or risk of damage or injury in the vicinity of the premises which will prevent of [sic] hinder use of the premises or access to them whether your property will be damaged or not is included but excluding

(i)

the first 6 hours of any interruption or interference; or

(ii)

any interruption or interference with your business because of outbreaks of either foot & mouth disease or avian flu.”

We will consider the parties’ positions and set out our conclusions on coverage in relation to each of these wordings in turn. In relation to MSA 1, it is accepted by MSA that “action of…other competent local, civil or military authority” encompasses action by the government. MSA also accepts that “action” encompasses government advice or guidance as well as the making of Regulations. MSA’s case is that coverage under the AOCA clause was not triggered for two reasons: (i) because none of the government action prevented physical access to any insured premises in circumstances where prevention of access requires nothing less than making physical access to the insured premises physically or legally impossible; and (ii) because no danger in the vicinity of any insured’s premises caused the government advice or Regulations.

On behalf of the FCA, Mr Edelman QC submitted that equating prevention of access with physical or legal impossibility put the test far too high. It would almost never be physically or legally impossible to access any part of premises for any purpose at all. He submitted that the approach of MSA was wholly unreal. It admitted that only Category 2 businesses required to close by the Regulations suffered a prevention of access. It denied that Category 1 businesses, pubs, cafes bars and restaurants, suffered a prevention of access because they could theoretically have turned themselves into takeaways. As with other wordings, including the Arch wording which we have already considered, he submitted that prevention of access included partial prevention and prevention of access for specific purposes required by the insured business, for example for customers to make purchases. Prevention of access did not require any physical or legal constraint. It was in this context that he gave the example of the hardware store to which customers could not go save in exceptional circumstances, after the restrictions on movement were imposed by the Regulations. He submitted that this was as much a prevention of access as if police tape had been placed across the entrance door. Accordingly, the FCA’s primary case was that from the 16 March advice to work at home and only travel for essential purposes, there was a prevention of access in respect of all insured premises.

As to what constituted: “a danger…in the vicinity of the premises” the FCA case is that there was an emergency or danger from at least 3 March when the government published its action plan and quarantining was in place, alternatively 12 March when the government elevated the risk level to high following COVID-19 being designated a notifiable disease and the first reported UK death. MSA admits that presence and/or real risk of the presence of COVID-19 amounted to an emergency which could endanger human life from 12 March onwards. Mr Edelman QC submitted that from 3 March, alternatively 12 March, the risk of contracting the disease was a danger everywhere in the UK and thus in the vicinity of any UK premises, a conclusion supported by MSA’s admission that from 12 March there was an emergency likely to endanger human life.

In their skeleton argument MSA expend considerable energy in relation to MSA 1 in demonstrating that the word “following” in the first line of the AOCA clause is synonymous with “resulting from” and connotes proximate cause or something close to it. We do not consider that this is something which matters very much, given that the FCA accepts that “following” does mean something more than a temporal or sequential connection and connotes some causal connection, however loose, between in this instance the government action and the interruption of or interference with the business. In other words, if the FCA is right as to the construction of the clause, nothing will turn on the precise causative effect of “following”.

Mr Kealey QC submitted that prevention means legal impossibility of gaining physical access to the premises. Difficulty in gaining access was not prevention but hindrance. The FCA case conflated the two concepts, but they were not synonymous. Where the draftsman of MSA 1 wanted to provide cover for hindrance in access or use he did so expressly, as in Clause 7 headed “Prevention of access” which provided that: “Consequential loss as a result of damage to property near the premises which prevents or hinders the use of the premises or access to them will be deemed to be damage.”

He also submitted that the fact that the AOCA clause referred not just to interruption but interference with the business did not dilute the meaning of prevention of access. He gave the example of a business with several premises, one of which suffered an enforced prevention of access which might interfere with the carrying on of the business as a whole.

Only “action” by the government which had the force of law could prevent access to the premises, which meant only the 21 and 26 March Regulations. Government advice or guidance was just that, not mandatory, and therefore not “action by [the government] where access will be prevented”. Mr Kealey QC relied upon the article by Lord Sumption in The Times on 26 March referring to the Prime Minister’s orders on 23 March:

“…in his press conference Boris Johnson purported to place most citizens under virtual house arrest through the terms of a press conference and a statement on the government website said to have “immediate effect”. These pronouncements are no doubt valuable as “advice”, even “strong advice”. But under our constitution neither has the slightest legal effect without statutory authority.

At the time of writing (Wednesday morning), it is unclear what power the prime minister thought that he was exercising. The relevant powers of the government are contained in the Public Health (Control of Disease) Act 1984 and the Civil Contingencies Act 2004. But it is doubtful whether either authorise the prime minister’s orders, which is presumably why the Coronavirus Bill has been introduced.

The ordinary rule is that a person may not be detained or deprived of his liberty without specific statutory authority. The 1984 act contains powers to restrict movement, but they are exercised by magistrates and apply only to particular people or groups who have been infected or whom they may have infected. The Civil Contingencies Act confers a temporary power of legislation on ministers that is exercised in a national emergency, but no specific power to detain people at home.

In the present national mood the prime minister’s orders will probably have strong public support and people will be inclined to comply whether they are binding or not. Yet we are entitled to wonder what kind of society we have become when an official can give orders and expect to be obeyed without any apparent legal basis, simply because it is necessary.

…

… There is a difference between law and official instructions. It is the difference between a democracy and a police state. Liberty and the rule of law are surely worth something even in the face of a pandemic.”

Because prevention of access connoted that it had to be legally and physically impossible to access the relevant insured premises, Mr Kealey QC went so far as to submit that, even with the passing of the 21 and 26 March Regulations, the AOCA clause did not bite at all, because even in the case of businesses which were required to close by the Regulations, the policyholder could still gain physical access to the premises, notwithstanding that it could not use the premises for the business. He submitted that a prevention of access cover could not be transformed into a prevention of use cover. In any event, given that MSA 1 was predominantly, although not exclusively, used for businesses in Categories 3 and 5, which were never required to close pursuant to the Regulations or otherwise, there was no prevention of access in relation to the premises of those businesses.

In relation to “a danger…in the vicinity of the premises” Mr Kealey QC accepted that the risk of a disease in the vicinity would qualify, but emphasised that although vicinity was an elastic concept, it connoted the neighbourhood of the insured premises, this being in essence a local cover. He was inclined to accept that the example we put to him in argument of a measles outbreak in a town which led the local authority to prevent access to all the schools in the town would qualify. As to whether there was in any given case a qualifying danger would depend upon the facts of the particular case. The action of the government, here the imposition of the regulations, was not in any sense caused by the danger of COVID-19 in the vicinity of any insured premises.

As we said above in our discussion of coverage under the Arch wording, the touchstone of prevention is impossibility, whereas hindrance connotes that access is rendered particularly difficult. To that extent, we agree with Mr Kealey QC that the FCA case conflates prevention with hindrance. However, we consider that Mr Kealey QC’s submission that access has to be physically and legally impossible goes too far and is unduly narrow. The “premises” is defined in the policy as “the buildings and the land…at the property address shown in your schedule occupied by you for the purpose of the business”. This demonstrates that what the AOCA clause is focusing on is prevention of access to the premises for the purpose of carrying on the business. It seems to us that the approach advocated by Mr Lockey QC on behalf of Arch that “prevention of access to the premises” connotes closure of the premises is a realistic one, equally applicable here. Accordingly, if the other elements of the cover under the clause are satisfied, closure of the premises is sufficient to amount to access being prevented, even if the policyholder could still physically access the premises for other purposes than carrying on the business, such as essential maintenance.

We also consider that, as in the case of the Arch wording, only total closure rather than partial closure will amount to prevention of access, from which it follows that any insured business which carried on part of its existing business, such as an existing takeaway service, as permitted by the Regulations, did not suffer total closure or, therefore, a prevention of access. In relation to businesses which, after the Regulations came into force, started up a takeaway or internet service from the premises which had not previously been carried out (and it is unclear whether there are any MSA 1 policyholders who fulfil that criterion) we would be inclined to reach the same conclusion as we reached in relation to the Arch wording, that in such a case there was still a prevention of access because the “business” which is defined in MSA 1 as the business specified in the schedule can no longer be carried on from the premises and the insured is carrying on a new “business”.

However, what is clear is that, in the case of businesses which were never required to close their premises (specifically businesses in Categories 3 and 5), as with the Arch policyholders in the same categories, there is no question of access to the premises having been prevented. It is no answer for the FCA to rely upon the restrictions on movement imposed by Regulation 6 of the 26 March Regulations to argue that customers’ ability to visit many premises was severely limited. At most, in the case of businesses which remained open or were not required to close, that was a hindrance in use, not a prevention of access.

The next question raised by the AOCA clause is what is meant by “action by the police or other competent local, civil or military authority”. It is immediately to be noted that unlike other wordings, such as the Arch wording, the clause does not refer to “action” and “advice”. In our judgment, Mr Kealey QC is right in his argument that in the context of this clause “action of [government] where access will be prevented” connotes steps taken by the relevant authority which have the force of law, since it is only something which has the force of law which will prevent access. There was discussion during the course of argument about police cordons and whether they had the force of law, but what emerged is that under various statutory regimes, individuals other than those permitted to go through a cordon, such as emergency workers, would be breaking the law if they went through a cordon.

As Lord Sumption pointed out, the government advice on 16 March and the days that followed, however strongly worded, did not have the force of law and, accordingly, did not amount to “action” within the meaning of this clause. It was only with the passing of the 21 March Regulations requiring businesses in Categories 1 and 2 to close and the more wide-ranging 26 March Regulations, that there was “action” by the government within the meaning of this clause. That is another reason why the FCA case that the advice of the Prime Minister on 16 March caused a prevention of access is wrong.

The final issue in relation to coverage under the AOCA clause concerns the requirement for the government action to be: “following a danger or disturbance in the vicinity of the premises”. In our judgment, Mr Kealey QC is right that these words demonstrate that the cover under this clause is a narrow, localised form of cover. The paradigm example of what it covers, as he submitted, is the bomb scare or gas leak in the vicinity or neighbourhood of the premises which causes the authorities (whether the police or the army in those examples), exercising statutory powers, to evacuate insured premises and require policyholders and their employees and customers not to access the premises. We do consider that the undefined term “vicinity” does have a local connotation of the neighbourhood of the premises. What that constitutes may depend on the nature of the danger or disturbance and the particular facts of the case, but, contrary to Mr Edelman QC’s submissions, we do not consider that the entire country can be described as in the vicinity of the insured premises, from which it follows that, on the true construction of the AOCA clause, the government action in imposing the Regulations in response to the national pandemic cannot be said to be following a danger in the vicinity, in the sense of in the neighbourhood, of the insured premises.

Even if there were a total closure of insured premises pursuant to the Regulations, there could only be cover if the insured could demonstrate that it was the risk of COVID-19 in the vicinity, in that sense of the neighbourhood, of the insured premises, as opposed to in the country as a whole, which led to the action of the government in imposing the Regulations. It is highly unlikely that that could be demonstrated in any particular case. The narrow, localised nature of this cover means that the wider issues of causation and counterfactuals, such as we have discussed in relation to the Arch and EIO wordings above and such as we discussed earlier in the judgment in relation to the so-called “disease clauses” and “hybrid clauses” do not arise.

The principal issue in relation to Clause 8, the Prevention of access-non damage clause in MSA 2, concerns the meaning of the words: “an incident within a one mile radius of your premises.” MSA contends that an incident is “a distinct and specific happening” and adopts the submissions of Hiscox in relation to its NDDA clause with which Clause 8 is in effect materially identical. On behalf of the FCA, Mr Edelman QC made the same submissions as he made in relation to the Hiscox NDDA clause. Mr Kealey QC added one point in relation to MSA 2, that the Prevention of access-non damage clause here referred internally to “interruption to your business caused by an incident” etc. and, unlike other insuring clauses such as Clause 4 loss of attraction and Clause 6 the notifiable disease clause (which we considered earlier in this judgment) was not prefaced by the words “consequential loss as a result of” or “following”. “Consequential loss” was defined in this wording as “Loss from interruption of or interference with the business”. The fact that Clause 8 did not refer to “consequential

Given that the clause is materially identical to the Hiscox NDDA clause, it is not necessary to repeat the conclusions we have already reached in relation to the NDDA clause as to why the FCA’s case founders on their inability to show “an incident” and why, even if there were cover, it is only the 21 and 26 March Regulations which imposed or ordered any denial of or hindrance in access. The only difference between the two wordings of any significance is the one to which Mr Kealey QC referred. Because the reference to “interruption” is not in stem wording covering a number of insuring clauses which point to “interruption” including “disruption”, and Clause 8 does not refer to “consequential loss” as defined, it seems to us that Mr Kealey QC is right that “interruption” in this clause must bear its strict meaning of cessation. We also consider that the FCA case on causation fails for the same reasons as we gave in relation to the Hiscox NDDA clause.

The FCA case in relation to the threat or risk of injury is the same as its case in relation to “danger” under MSA 1. Mr Edelman QC submitted that the real risk of the presence of COVID-19 was a threat or risk of injury throughout the UK, which therefore included the vicinity of the insured premises. The wording was wider than the AOCA clause in MSA 1 because it encompassed prevention or hindrance in the use of and access to the insured premises. It was common ground that use of or access to premises is hindered where it is made more difficult or is inhibited and whether the difficulty or inhibition applies to the insured and/or its employees and/or its customers. In relation to “action by a competent public authority”, Mr Edelman QC submitted that, as in the case of the AOCA clause in MSA 1, this encompassed not only the Regulations but the government advice from 16 March onwards.

Given that this clause was wider than the AOCA clause in MSA 1, Mr Kealey QC accepted also that “action [of government]” which hindered use of or access to premises could encompass not just the imposition of Regulations having the force of law, but government advice as well. So far as government advice was concerned, the Prime Minister’s advice on 16 March could not amount to prevention of access or use anywhere in the UK. It might amount to hindrance, but that would depend on the particular facts of any given case. The announcement on 18 March that schools would close from 23 March meant that, as was also accepted by EIO, there was a hindrance of use of schools from 23 March. MSA accepts that the Prime Minister’s advice on 20 March that businesses in Categories 1 and 2 should close constituted a hindrance of use but did not constitute prevention of access or use. However, forges (the specialist cover provided by MSA 3) were in Category 5 and were never required to close. Mr Kealey QC submitted that there was no prevention or hindrance of access or use to or of forges.

Mr Kealey QC accepted that COVID-19 was an “injury”. He also accepted, as with the AOCA clause in MSA 1 that the risk of a disease in the vicinity would qualify, but emphasised that although vicinity was an elastic concept, it connoted the neighbourhood of the insured premises, this being in essence a local cover. He repeated in essence the submissions he made in relation to the AOCA clause, that no threat or

The MSA 3 wording is much wider than the AOCA clause in the MSA 1 wording, since it encompasses not only prevention of access but prevention of use and hindrance of both access and use. In those circumstances, Mr Kealey QC correctly accepted that “action of a competent public authority” had a wider meaning in this wording and encompassed not only the Regulations but the advice of the government. However, the essential problem which the FCA faces in relation to this specialist forge cover is that forges are in Category 5, businesses about which the advice and the Regulations are silent and which were never required to close. In relation to such businesses it is impossible to say that the advice or the Regulations prevented access or use. At most it could be said that they hindered use of insured premises, for example by restricting movement so that members of the public could not visit retail premises except for essential purchases. However, we are doubtful as to how many members of the public ever visit forges and we were informed by Mr Kealey QC that there have been no claims by insureds under this specialist cover. Were there to be any claim, there would be a question of fact as to whether the insured could establish that the government advice or Regulations had caused a hindrance of use of the insured premises.

We consider that the other difficulty faced by the FCA in relation to this wording is the requirement that the government action is: “following threat or risk of…injury in the vicinity of the premises”. We repeat what we said about the words: “in the vicinity of the premises” in relation to the AOCA clause in MSA 1. Although “vicinity” is an elastic concept, in ordinary usage it connotes neighbourhood and, contrary to the FCA submissions, the entire UK is not in the vicinity of any insured premises. This is narrow, localised cover and, even if an insured could demonstrate that the government advice or Regulations had caused a hindrance of use of the insured premises, there could only be cover if the insured could also demonstrate that it was the risk of COVID19 in the vicinity, in that sense of the neighbourhood, of the insured premises, as opposed to the country as a whole, which led to the action of the government in giving the advice and imposing the Regulations. As with the AOCA clause, it is highly unlikely that that could be demonstrated in any particular case. Also, as with that clause, since this is narrow localised cover, the wider issues of causation and counterfactuals which arise in other contexts, do not arise.

RSA 1, the Cottagesure policy wording, does not contain a prevention of access clause. RSA 2.1 and 2.2, which do, are wordings for policies entered by Eaton Gate MGU Ltd as a Managing General Underwriter on behalf of RSA. RSA 2.1 is entitled “Restaurants, Wine Bars, Public Houses Policy” and is thus issued to such businesses in Category 1. RSA 2 is entitled “Shop Policy” and was entered by policyholders with one or more shops, therefore falling into Categories 3 and 4.

In RSA 2.1, the relevant Prevention of Access clause is one of the Extensions to the Business Interruption cover under Section 2, the main business interruption insurance

Under “What is covered”, Extension F headed “Prevention of Access-Public Emergency” provides:

“The actions or advice of a competent Public Authority due to an emergency likely to endanger life or property in the vicinity of the Premises which prevents or hinders the use or access to the Premises”

Under “What is not covered” the Extension provides:

“Any loss

a)

during the first four hours

b)

during any period other than the actual period when access to the Premises was prevented

c)

as a result of labour disputes

d)

occurring in Northern Ireland

e)

as a result of the diseases specified in Extension A (a) diseases [which does not include COVID-19]

Any amount in excess of £10,000”

The RSA 2.2 wording is very similar. The stem wording to the Extensions to Section 7 the Business Interruption cover is the same as in RSA 2.1. The relevant extension, Extension F, headed “Public Emergency”, is in identical terms to the cover provided by Extension F “What is covered” in the RSA 2.1 wording. Under “What is not covered” the first four exceptions are identical to those in the RSA 2.1 wording. The fifth exception, however, is different. It provides: “e) As a result of infectious or contagious diseases any amount in excess of £10,000”.

As a consequence of an exclusion in the Prevention of Access-Public Emergency Extension in the RSA 3 wording for any Infectious Diseases covered under the Infectious Diseases Extension, it is accepted by the FCA that the Prevention of AccessPublic Emergency Extension in the RSA wording is not applicable. We have considered the Infectious Diseases Extension above in the section of this judgment dealing with disease clauses.

We have already referred to RSA 4 in the context of the disease clauses. As we said there, the structure of RSA 4 is significantly different from that of the other lead policies in the test case. The “Specified Causes” under Clause 2.3 include at (xii):

“In the event of interruption or interference to the Insured’s Business as a result of:…

…

xii.

Prevention of Access – Non Damage during the Period of Insurance where such interruption or interference is for more than eight (8) consecutive hours…

within the Territorial Limits, the Insurer agrees to pay the Insured the resulting Business Interruption Loss.”

The term “Prevention of Access-Non Damage” is defined in Definition 87:

“i.

the discovery of a bomb or similar suspect device or the threat, hoax or deceptive information of a bomb or similar suspect device … in the Vicinity of the Insured Locations; ii. the actions or advice of the police, other law enforcement agency… governmental authority or agency in the Vicinity of the Insured Locations; … and/or

iii.

the unlawful occupation of … other property in the Vicinity of the Insured Locations by any individuals …

which prevents or hinders the use of or access to Insured Locations during the Period of Insurance.”

We repeat the definition of “Vicinity” in Definition 120:

“…an area surrounding or adjacent to an Insured Location in which events that occur within such area would be reasonably expected to have an impact on an Insured or the Insured’s Business”

The term “Business Interruption Loss” is defined to include, so far as relevant: “the amount by which the Turnover during the Indemnity Period falls short of the Standard Turnover”. “Standard Turnover” is defined in Definition 107:

“…the Turnover during that equivalent period before the date of any Covered Event which corresponds with the Indemnity Period to which adjustments have been made to take into account the trend of the Insured’s Business and for variations in or other circumstances affecting the Insured’s Business either before or after the Covered Event or which would have affected the Insured’s Business had the Covered Event not occurred so that the figures thus adjusted will represent as nearly as may be reasonably practicable the results which but for the Covered Event would have been obtained during the Indemnity Period.”

The term “Covered Event” is defined as the events described in Insuring Clause 2.1, 2.2, 2.3 or 2.4 or any applicable Extension and thus includes both the insured perils with which we are concerned in this judgment, Notifiable Disease etc. and Prevention of Access –Non Damage.

RSA does not dispute that actions or advice of the government on matters of public health would be “actions or advice of a competent Public Authority” or that the COVID-19 pandemic was a general public health “emergency”. It recognises that the insuring clause talks of prevention or hindrance of both use of and access to the premises. However, it relies upon exclusion (b) in “What is not covered”: “Any loss…during any period other than the actual period when access to the Premises was prevented” which it says delineates the scope of cover so that the sole relevant question is whether the assumed losses were incurred during a period when access to the premises was prevented. Mr Edelman QC submitted that this was in effect giving with one hand and taking away with the other. If it had been intended to limit the cover provided to prevention of access, the insuring clause would have said so. It is clear that in exclusion b) “access to the Premises was prevented” was shorthand for “use of or access to the Premises was prevented or hindered”, the intention of the exclusion being to make clear that there was no cover for the after-effects of any relevant prevention or hindrance.

Apart from its case on that exclusion, in relation to prevention of access, RSA adopts the rather extreme position of MSA that prevention requires nothing less than making physical access to the insured premises physically or legally impossible. However, it accepts that the provisions in the 21 and 26 March Regulations which required premises to close did amount to a prevention of use (and a fortiori a hindrance of use) of the insured premises. However, RSA does not accept that either the advice about social distancing, self-isolation, working at home and staying at home from 16 March or the restrictions on movement in Regulation 6 of the 26 March Regulations (to all of which it refers compendiously as “Social Distancing Measures”) prevented access to insured premises.

The principal matter in dispute between the parties in relation to this wording is the meaning of “emergency…in the vicinity of the Premises”. As in relation to the MSA 1 AOCA clause which refers to “a danger or disturbance in the vicinity of the Premises” Mr Edelman QC submitted that the emergency was the pandemic, as RSA accepted and, since the pandemic was everywhere in the UK, occurring in all areas, it was “in the vicinity of the Premises” for the purposes of this provision. He submitted that the provision does not require that the “emergency” is only in the vicinity. RSA’s interpretation, that the insurance is intended to cover where there is a local “emergency”, did involve inserting the word “only”.

Mr Turner QC on behalf of RSA submitted that the broad purpose of this clause is to provide an indemnity against the effect on access to and use of insured premises of restrictions imposed by the emergency services. By their very nature emergencies which affect access to or use of the insured premises are likely to be in the vicinity of the premises.

In relation to the RSA 2.2 wording, Mr Turner QC also relied upon exclusion e). He submitted that the placing of the words “any amount in excess of £10,000” within the exclusion was an obvious mistake in the drafting. As the RSA 2.1 wording demonstrated in the equivalent “What is not covered”, the reference to “any amount in excess of £10,000” was intended to be a separate financial inner limit to the amount of the indemnity available under this provision, such as was found in a number of the other “What is not covered” provisions within these Extensions. A reasonable person would have understood that the inner limit within the Public Emergency Extension was intended to be free-standing. On this basis, exclusion e) was a complete exclusion in respect of infectious or contagious diseases, of which COVID-19 was one.

Mr Edelman QC submitted that it was not obvious that there was any error in exclusion e) nor was moving the words any amount in excess of £10,000 to a new line with a capital “A” an obvious correction. There was no basis for reading the wording in the manner for which Mr Turner QC contended. One could well imagine why diseases would be singled out for a sub-limit of £10,000.

We will deal at the outset with RSA’s case on the application of exclusion b) in the two wordings and with the correct construction of exclusion e) in the RSA 2.2 wording. We agree with Mr Edelman QC that RSA’s case on exclusion b) involves the absurd proposition that cover for “hindrance” and for “prevention or hindrance of use” is given with one hand and taken away with the other. If it had been intended, as Mr Turner QC submitted, to delimit cover to prevention of access, this provision could and would have said so in “What is covered”, not engaged in the tortuous process which RSA’s case involves. Although exclusion b) is not as well-worded as it could be, its intention is clear. It is excluding cover for the after-effects of the relevant prevention or hindrance, not limiting the cover it has just bestowed.

So far as exclusion e) is concerned, we agree with Mr Edelman QC that there is no obvious error in the positioning of the words “any amount in excess of £10,000” within the exclusion and that insurers might very well have good commercial reasons for inserting an inner limit in respect of disease cover. Whilst the provision as it stands could be better worded, we do not agree with Mr Turner QC’s submission that it does not read grammatically. In any event, that is no reason for the change proposed, which rewrites the clause and, conveniently for RSA, creates a complete exclusion where previously there was only a financial limit. In our judgment, were it not for the order of the wording in “What is not covered” in the RSA 2.1 wording, this point would be totally unarguable. The fact that different wording is expressed in different terms is not a reason for construing this provision to mean something different from what it clearly means on its face, that in respect of diseases there is an inner limit of £10,000. Even if RSA were able to establish that in positioning the words in the exclusion, they had made a mistake, the mistake was entirely theirs, to which insured policyholders cannot be said to have contributed, so there is no scope for the application of the doctrine of rectification.

The RSA 2.1 and 2.2 wording is, like MSA 3, wider than some of the other wordings before the Court, since it encompasses not only prevention of access but prevention of use and hindrance of both access and use. We have already concluded in relation to other wordings that prevention of access does not require physical or legal impossibility. Prevention would be demonstrated by the requirement to close and, even if that were not prevention of access, it would clearly be prevention of use. Because these wordings also cover hindrance of access and use, the much-discussed example of the pub or restaurant which maintained an existing take-away service would still be suffering a hindrance of access or, at the very least of use, since it could not use the restaurant space for in-house eating or drinking and its customers could not access that

It also seems to us that what RSA describes as the Social Distancing Measures and, in particular, Regulation 6 of the 26 March Regulations restricting movement could be said, in a given case, to have hindered use of insured premises, in the sense that it either prohibited a potential customer from visiting non-essential retail premises at all or only permitted that customer to do so for the purposes of essential purchases. Whether or not there was such hindrance would depend upon the particular facts of the case.

As in the case of the MSA 1 AOCA clause, we consider that the principal difficulty which the FCA faces in relation to this wording is the requirement that the “emergency” (“danger” in the MSA 1 AOCA clause) is “in the vicinity of the Premises”. We consider that Mr Turner QC is right that these words are a clear indicator that this is a narrow, localised form of cover, the paradigm example of which would be the police cordoning off an area in which the insured premises are located because of intelligence that materials for bomb-making were located in that area, the example Mr Turner QC gave. It does not seem to us that the fact that, unlike the MSA 1 AOCA clause, this clause also encompasses hindrance and use, in any sense requires an expansion of the geographical limitation on its scope. As we have already said, however elastic the concept of “vicinity” it connotes neighbourhood, the area surrounding the premises. The FCA submitted in relation to the insurers’ submissions on the meaning of “in the vicinity” generally that they involved reading in the word “immediate” before “vicinity” but we do not accept that criticism, as it seems to us that the undefined term

“vicinity” in itself connotes an immediacy of location. What “in the vicinity” means in any particular case may depend upon the nature of the “emergency” and the facts of the case. However, contrary to Mr Edelman QC’s submissions, and at least in the absence of a special definition of the term, as in RSA 4, we do not consider that the entire country can be described as in the vicinity of the insured premises on the wording of this policy. It follows that the government action and advice in response to the national pandemic cannot be said to be due to an emergency in the vicinity, in the sense of in the neighbourhood, of the insured premises.

There could only be cover under this wording if the insured could also demonstrate that it was an emergency by reason of COVID-19 in the vicinity, in that sense of the neighbourhood, of the insured premises, as opposed to the country as a whole, which led to the actions or advice of the government. As with the AOCA clause and the MSA 3 wording, it is highly unlikely that that could be demonstrated in any particular case. Also, as with that clause, since this is narrow localised cover, the wider issues of causation and counterfactuals which arise in other contexts, do not arise.

Because the RSA 4 wording, like RSA 2.1 and RSA 2.2, encompasses prevention and hindrance of use and access to insured premises (referred to as “locations” in this wording), the parties essentially repeated the same submissions as to the meaning of those words and as to whether government actions or advice prevented or hindered use or access, so that it is not necessary to repeat those submissions.

On behalf of the FCA, Mr Edelman QC submitted that the requirement in the Prevention of Access-Non Damage definition that the actions or advice of the government were in the Vicinity of the Insured Locations was easily satisfied here. The government actions and advice took place nationally and without distinction between localities so that they took place throughout the UK including within the Vicinity of the premises.

Mr Turner QC contended on behalf of RSA that national actions or advice were not covered, as this would render the Vicinity requirement redundant. The clause only applied where the actions or advice are specific to the Vicinity of the premises even if they also impact a wider area. In relation to an argument by Mr Edey QC on behalf of the HIGA Interveners that RSA’s argument gave no effect to the words in the Vicinity definition: “in which events that occur within such area would be reasonably expected to have an impact”, Mr Turner QC submitted that RSA’s construction did give meaning to the words. He gave the example of an insured location on an industrial estate with a single route of access. Plainly anything which happened on that single route of access would be likely to have an impact on the insured premises.

It is important in considering this Prevention of Access-Non Damage provision to identify first what is the insured peril. In our judgment, this involves three interlinked elements: (i) “interruption or interference to the Insured’s Business as a result of (ii) the actions or advice of…governmental authority…in the Vicinity of the Insured Locations (iii) which prevents or hinders the use of or access to Insured Locations during the Period of Insurance.” In other words, as in the case of many of the other wordings we are considering, this is a composite peril. However, what is strikingly different about this wording is that it does not require that the actions or advice of the government are “following” or “due to” an “emergency” or “danger” in the vicinity of or within a particular radius of the insured premises. All that it requires in entirely general terms is that the actions or advice of the government are “in the Vicinity of the Insured Locations.” Even without the extended definition of “Vicinity” in the RSA wording, we consider that the actions or advice of the government, taken nationally and affecting all insured businesses will inevitably be in the vicinity of the insured premises if they lead to prevention or hindrance of use or access of the insured premises. It is no answer for RSA to say that the actions or advice also affected a substantial number of other insured premises elsewhere. Nowhere in this wording is there any limitation of qualifying actions or advice of government to such actions or advice specific to the insured premises or their vicinity.

Accordingly, we consider that in principle there will be cover available under this wording in a number of scenarios. Where insured premises closed in response either to government advice from 20 March onwards or to the Regulations, there will be a prevention of access and/or use from the moment of closure. What RSA describes as the Social Distancing Measures, beginning with the Prime Minister’s advice on 16 March, could also be said to amount to hindrance of use of insured premises. We repeat the points we made on the RSA 2.1 and 2.2 wording above.

We have set out above the relevant trends provision in the RSA 4 wording in the definition of Standard Turnover. Because of the definition of Covered Event, the quantification machinery in this wording does not require any manipulation to make it applicable to non-damage business interruption.

Mr Turner QC submitted in relation to the Prevention of Access- Non Damage clause that the correct counterfactual was one where you stripped out the actions and advice in so far as they impacted on the insured premises but everything else remained the same, including the Regulations and Social Distancing Measures, which meant that any losses would have been suffered in any event. On behalf of the FCA, Mr Edelman QC submitted that this was ridiculous and you could not engage in cherry-picking or salami slicing of that kind. All relevant aspects of the insured peril, including the nationally imposed government actions and advice had to be stripped out.

The two important preliminary points we made in relation to the trends provision in the RSA 3 wording apply with equal force here and bear repeating. First, the definition of Standard Turnover with its trends provision is not part of the delineation of cover, but part of the quantification machinery for calculating the business interruption loss on the basis that there is a qualifying prevention of access. Where the policyholder has therefore prima facie established a loss caused by an insured peril, it would seem contrary to principle, unless the policy wording so requires, for that loss to be limited by the inclusion of any part of the insured peril in the assessment of what the position would have been if the insured peril had not occurred. Second, subject to the particular wording providing for something different, the object of the quantification machinery (including any trends clause or provision) in the policy wording is to put the insured in the same position as it would have been in if the insured peril had not occurred.

Here, as we have said, the insured peril is interruption or interference to the Insured’s Business as a result of the actions or advice of…governmental authority…in the Vicinity of the Insured Locations which prevents or hinders the use of or access to Insured Locations during the Period of Insurance. The approach which RSA advocates involves a totally unrealistic and artificial counterfactual which assumes that the part of the government actions or advice which relates to the insured premises and their vicinity is stripped out but the nationwide actions and advice remain somehow the same. The fallacy in this approach is that the actions and advice in so far as they affect the insured premises and their vicinity are inseparable from the nationwide actions and advice. They are one and the same. The only way of establishing what the insured’s business would have achieved if the Covered Event (here the Prevention of Access-Non Damage peril insured) is to strip out all the prevention and hindrance and all the actions and advice, in other words to assume that there had been no COVID-19.

There are two Zurich policies for consideration. Zurich 1 is a Combined All Risks policy sold mainly, via brokers and other authorised intermediaries, to mid-market companies with a turnover of £5 million or more. Zurich 2 is a Commercial Combined Manufacturing policy. The policyholders under both forms of policy fall into all seven categories with which we are concerned, but with a heavy leaning towards Category 5, service businesses and manufacturers, which accounted for 84% of all the Zurich policyholders.

Each wording contains, as an extension to the standard Business Interruption cover, an Action of Competent Authorities or AOCA clause. The language of each is for present purposes materially identical. The extension is in the Schedule to the policy in the case of Zurich 1 but in the body of the wording in Zurich 2, a distinction which makes no difference to the construction of the clause. We take the relevant provisions primarily as they appear in Zurich 1. The Extensions begin with this “stem wording”:

“EXTENSIONS

Section B1

The Business Interruption cover is subject to the extensions shown below:

Any loss as insured by this Section resulting from interruption of or interference with the Business in consequence of accidental loss destruction or damage at the under-noted situations or to property as under-noted shall be deemed to be an Incident …”

The AOCA clause then provides (using Zurich 2 by way of illustration for its maximum indemnity period):

“Action of competent authorities

Action by the police or other competent local, civil or military authority following a danger or disturbance in the vicinity of the premises whereby access thereto will be prevented provided there will be no liability under this section of this extension for loss resulting from interruption of the business during the first 3 hours of the indemnity period.

The maximum indemnity period is 12 months.”

It will be immediately apparent that this AOCA clause is very similar to the AOCA clause in the MSA 1 wording which we considered above.

Zurich alone of the insurers whose wordings contain a similar provision originally denied that the government was a “competent civil authority” but that point was abandoned by Mr Craig Orr QC on Zurich’s behalf at the outset of his oral submissions. At the outset of those submissions, Mr Orr QC also pointed out that a point taken by the FCA that the Maximum Indemnity Period in Zurich 2 was 12 months somehow demonstrated that the relevant danger could last 12 months was a bad point. As Mr Orr QC pointed out, the Indemnity Period was the period after the occurrence of the relevant incident during which the results of the business are affected, not the period during which the danger or disturbance continues. This point was, wisely, not pursued by the FCA orally.

On behalf of the FCA, Ms Mulcahy QC submitted that “action” by the government could encompass not just legally binding regulations but advice or guidance. The

In relation to “prevention of access” she noted that Zurich, like MSA, take the extreme position that the phrase means physical obstruction or impossibility or alternatively, a complete cessation of the business, so that even with the passing of the 21 and 26 March Regulations, the AOCA clause did not come into effect. She submitted that this was a misreading of the Regulations and their effect. The US authorities on which Zurich relied were on different wordings and related to different situations.

In relation to “danger or disturbance in the vicinity of the premises” Ms Mulcahy QC noted that Zurich went so far as to say that “danger” could not encompass disease because there was a specific disease clause, as to which she submitted that there was no presumption against overlap and that if “danger” is given its ordinary and natural meaning, a disease is plainly a danger. In oral submissions, this point was refined by Mr Orr QC to say that “danger” does not encompass a national infectious disease pandemic.

As for Zurich’s point that “vicinity” does not mean “locality” but “immediate locality”, she submitted that this was to read the word “immediate” into the clause, which was impermissible. The Zurich wording used the phrase “immediate vicinity” elsewhere, as in Condition 1 applicable to Section G in Zurich 1, which relates to the Use of Heat and refers to clearing the area “in the immediate vicinity of the work” of loose material (the same provision appearing in Condition 2 applicable to Section K in Zurich 2). She submitted that the definition of “Vicinity” in the RSA 4 wording is a sensible, workable definition. Since the danger of the disease was everywhere in the UK, the whole country was in the vicinity of the insured premises, because the danger everywhere else could reasonably be expected to impact on the insured premises. The suggestion by Zurich that there has to be a specific local danger under this clause was equally misconceived. The fact that a danger was within the vicinity did not prevent it also being elsewhere outside the vicinity.

The principal submissions on behalf of Zurich by Mr Orr QC echoed submissions addressed by Mr Kealey QC in relation to the MSA 1 AOCA clause. First he submitted that the Zurich wordings drew a clear distinction between prevention and hindrance and access and use as demonstrated by the Extension POA 4 Prevention of Access in Zurich 1 which began: “Property in the vicinity of the Premises, loss or destruction of or damage to which shall prevent or hinder the use of the Premises or access thereto…” Prevention of access in this clause should therefore be given a narrow meaning, that access to the premises was physically obstructed or rendered impossible. This construction of “prevention of access” was consistent with US decisions on business interruption insurance where access is “prohibited” or “denied”. For example, in the decision of the US District Court (New York) in Abner, Herrman & Brock, Inc. v Great Northern Insurance Co., 308 F. Supp. 2d 331 (2004), the policy provided cover where there was a prohibition of access by a civil authority. There was cover for the period from 11 to 14 September 2001 when access to the insured premises was prohibited by the New York civil authorities in response to the 9/11 attacks, but no cover thereafter when vehicular access was restricted but pedestrian access was permitted.

Second, Mr Orr QC submitted that the phrase “in the vicinity of the Premises” or “within the vicinity” is used not just in the AOCA clause but in the Prevention of Access Extension quoted above and in the Loss of Attraction extension in Zurich 2 and should be given the same meaning wherever it occurs, connoting that the relevant event occurs close to or nearby the insured premises, in close specific proximity to the premises. This was especially so of the Prevention of Access Extension since, unless the damaged property is near or close to the insured’s premises, the damage to that property would be unlikely to prevent or hinder access to the insured’s premises.

Mr Orr QC’s third point was that these extensions to the Business Interruption cover provide pockets of cover, each limited by its own specific requirements, so that there is no blanket coverage for example in respect of danger or notifiable diseases. That was particularly apparent from the notifiable diseases extension. The same circumscription of cover was apparent in the AOCA extension: (i) that the geographical location of the danger or disturbance must be “in the vicinity”; (ii) that the danger or disturbance in the vicinity must cause the action by the relevant authority, i.e. the one must follow the other and (iii) the type of public authority action which triggered cover must prevent access to the premises. He submitted that these limitations were fundamental but were ignored by the FCA in their arguments. He noted that the causal link between the vicinity and the government action was an important distinction between this wording and the Arch “Government or Local Authority Action” clause.

Like Mr Kealey QC before him, Mr Orr QC submitted that the paradigm case contemplated by the AOCA clause is a bomb scare, a brawl or a serious traffic accident, in response to which the police or other relevant authority takes action which prevents access to the insured premises, that is shuts off access for all purposes because access to the premises is unsafe or needs to be kept clear for the emergency services or police investigations. As Riley on Business Interruption Insurance 10^th edition at §10.34 explains, these types of AOCA extensions arose out of terrorist activity in the UK in the 1980s and 1990s which involved devices that did not explode, not just those that did, so that traditional business interruption cover contingent on property damage did not respond.

Mr Orr QC submitted that since the clear objective intent of the AOCA clause was to provide cover in respect of prevention of access as a result of action by the relevant authority in response to such a local incident, that was a very long way from a pandemic and government response nationally to it. If one were drafting an extension to apply to government measures in response to a pandemic, it would certainly not look like the AOCA extension and no reasonable reader could think that this extension covered the pandemic or the wholly unprecedented measures the government took in response.

We have already noted above Mr Orr QC’s submission that “a danger” does not encompass a national infectious disease pandemic, but rather a transient incident posing a risk of danger such as a bomb threat or fire. What is meant by “a danger” is given further colour by the words “or disturbance” which reinforce that the clause is contemplating an incident specific to the locality of the premises rather than a continuing countrywide state of affairs.

He submitted that the clause was focusing on the local danger or disturbance in the vicinity and its connection with the relevant civil authority action. It was not focusing on the connection between the civil authority action and any danger or disturbance outside the vicinity. The meaning of “in the vicinity” was best encapsulated in the term “in the immediate locality”. The FCA’s argument that this could not be so because of the use of the words “in the immediate vicinity” in the condition about hot work was a false point. That condition is to do with welding and other hot work and “in the immediate vicinity” means effectively the small area where the hot work is being carried out. The phrase was being used in a very different context and did not assist.

Mr Orr QC submitted that the use of “following” in the AOCA clause meant that the relevant civil authority action (a) must come later in time than the danger or disturbance in the vicinity of the premises and (b) must have resulted from that danger or disturbance. In other words, he submitted that the causal connection was akin to proximate causation.

As we said above in our discussion of coverage under the Arch wording, the touchstone of prevention is impossibility, whereas hindrance connotes that access is rendered particularly difficult. To that extent, we agree with insurers that the FCA case conflates prevention with hindrance. However, we consider that Mr Orr QC’s submission, like that of Mr Kealey QC before him, that access has to be physically and legally impossible, goes too far and is unduly narrow. The stem wording for the Extensions in

We also consider that, as in the case of the Arch wording and the MSA 1 AOCA clause, only total closure rather than partial closure will amount to prevention of access and we repeat in that regard what we said above in the context of the MSA 1 AOCA clause about issues such as the continuation of takeaway services.

However, as in the case of the other insurers who insure policyholders in Category 5, we do not consider that there can be said to have been a prevention of access to their premises which were never required to close and about which the 21 and 26 March Regulations were silent. It is no answer for the FCA to rely upon Regulation 6 of the 26 March Regulations restricting movement. At most that was a hindrance of use of Category 5 premises which this clause does not cover.

The next question raised by the Zurich AOCA clause is what is meant by “Action by the Police or other competent Local, Civil or Military Authority”. Whilst we see the force of the submissions made by Ms Mulcahy QC as to the dictionary meaning of “action”, the word has to be read in context. In this clause, it is “action…whereby access [to the premises] shall be prevented”. As in the case of the MSA 1 AOCA clause, it seems to us that this connotes steps taken by the relevant authority which have the force of law, since it is only something which has the force of law which can prevent access. Furthermore, if it had been intended to encompass advice in this clause, the parties could and would have said so expressly, as they did in the disease clause. It follows that the only “action” of the government which would qualify under this clause is the imposition of the 21 and 26 March Regulations and any subsequent Regulations or legislation with the force of law.

The final issue in relation to coverage under the Zurich AOCA clause concerns the requirement for the government action to be: “following a danger or disturbance in the vicinity of the Premises”. It is first necessary to consider the meaning of the word “following”. We do not consider that Mr Orr QC is right that it is akin to proximate causation. Rather, as we said in relation to the MSA 1 AOCA clause, as the FCA accepts, it connotes some causal connection, however loose, between in this instance the government action and the interruption of or interference with the business. Given that there has to be such a causal connection, we do not consider that anything turns on the precise meaning of “following”.

In relation to “danger…in the vicinity of the Premises” we agree with Mr Orr QC that this clearly indicates that this is narrow localised cover intended to cover dangers occurring in the locality of the insured’s premises, of which the paradigm example is a bomb scare. We also agree that the FCA’s criticism that Zurich is reading in the word “immediate” before “vicinity” is misplaced. As we have said elsewhere, the undefined term “vicinity” in itself connotes an immediacy of location. Furthermore, the reference to “in the immediate vicinity” in the conditions about welding and hot work are in an entirely different context not concerned with the vicinity of the insured premises at all.

We also agree that the word “danger” gains some colour from its juxtaposition with

“disturbance”. The paradigm example of a “disturbance” in this context would be an affray or brawl. In our judgment “vicinity” although not defined, does have a local connotation of the neighbourhood of the premises. What that constitutes may depend on the nature of the danger or disturbance and the particular facts of the case, but, contrary to the FCA’s submissions, we do not consider that the entire country can be described as in the “vicinity” of the insured premises. As Mr Orr QC correctly submitted, the overall phrase “a danger or disturbance in the vicinity of the Premises” contemplates an incident specific to the locality of the premises rather than a continuing countrywide state of affairs.

Accordingly, on the true construction of the Zurich AOCA clause, the government action in imposing the Regulations in response to the national pandemic cannot be said to be following a danger in the vicinity, in the sense of in the neighbourhood, of the insured premises.

Even if there were a total closure of insured premises pursuant to the Regulations, there could only be cover if the insured could demonstrate that it was the risk of COVID-19 in the vicinity, in that sense of the neighbourhood, of the insured premises, as opposed to in the country as a whole, which led to the action of the government in imposing the Regulations. It is highly unlikely that that could be demonstrated in any particular case. The narrow, localised nature of this cover means that the wider issues of causation and counterfactuals, such as we have discussed in relation to the Arch, EIO and RSA 4 wordings above and such as we discussed earlier in relation to the so-called “disease clauses” do not arise.

Causation

The insurers prepared an 82 page joint skeleton on the issues of causation to which they contend this case gives rise and Mr Kealey QC addressed us on behalf of all the insurers on those issues for most of Day 4 of the hearing. With no disrespect to the diligence, erudition and ingenuity of insurers’ counsel, we do not propose to set out a lengthy exegesis of the relevant principles. As will be apparent from what we have said in our analysis of the various wordings, it seems to us that the issues of causation in this case resolve themselves as part of the process of construction of those wordings.

Since, as we have said, the insurers place heavy reliance on the decision of Hamblen J (as he then was) in Orient Express which they contend supports their case in relation to the counterfactual and on “but for” causation, we do need to analyse that decision. In Orient Express the insured OEH owned a hotel in the French Quarter of New Orleans. It made a claim for material damage and business interruption losses after the hotel suffered significant damage as a consequence of Hurricanes Katrina and Rita. It was closed for some months and OEH suffered significant business interruption losses. The insurance was in effect an all risks policy, the insuring clause of which provided:

“In consideration of the Insured... paying the premium... the Insurers... agree... to indemnify the Insured:

(a)

under the Material Damage and Machinery Breakdown Sections against direct physical loss destruction or damage except as excluded herein to Property as defined herein such loss destruction or damage being hereafter termed Damage

(b)

under the Business Interruption Section against loss due to interruption or interference with the Business directly arising from Damage and as otherwise more specifically detailed herein.”

There was also an insuring clause at the beginning of the Business Interruption section which provided:

“If any property owned used or otherwise the responsibility of the Insured for the purpose of or in the course of the Business suffers Damage as defined or there occurs an event or circumstances as described elsewhere in this Section of the Policy and the Business be in consequence thereof interrupted or interfered with the Insurers will pay to the Insured the amount of the loss resulting from such Interruption in accordance with the provisions contained therein”.

By extension, the policy also provided cover against Prevention of Access (“POA”) and Loss of Attraction (“LOA”) in these terms:

“This policy is extended to include reduction in Revenue incurred by the Insured:

(a)

arising out of Property in the vicinity of any location owned occupied or operated by the Insured suffering Damage or being closed (in whole or part) or deemed unusable by a competent authority and which shall consequently prevent or hinder the use of the location concerned or access thereto whether Property Insured shall be damaged or not;...

This Policy extends to indemnify the Insured in respect of a reduction in Revenue resulting directly from loss destruction or damage to property or land in the vicinity of any premises owned and/or managed by the Insured and insured under this Policy.”

The policy also contained a trends clause in the following terms:

“In respect of definitions under 3, 4, 5 and 6 above for Gross Revenue and Standard Revenue adjustments shall be made as may be necessary to provide for the trend of the Business and for variations in or special circumstances affecting the Business either before or after the Damage or which would have affected the Business had the Damage not occurred so that the figures thus adjusted shall represent as nearly as may be reasonably practicable the results which but for the Damage would have been obtained during the relative period after the Damage.”

The insurers provided an indemnity to OEH under the POA and LOA extensions but they were subject to much lower limits than the main business interruption provision under which the insurers refused to indemnify OEH, essentially on the grounds that OEH could only recover in respect of loss which it could show would not have arisen had the damage to the hotel not occurred. Since the hurricanes had devastated the whole of the surrounding area of New Orleans, OEH would have suffered the same business interruption loss even if the hotel had remained undamaged, since no one would have visited the hotel given the devastation in the vicinity.

The dispute was referred to arbitration before a distinguished panel consisting of Sir

OEH obtained permission to appeal under section 69 of the Arbitration Act 1996 in respect of two questions of law, framed as follows:

“(1)

Whether on its true construction, the policy provides cover in respect of loss which was concurrently caused by: (i) physical damage to the property; and (ii) damage to or consequent loss of attraction of the surrounding area;

(2)

Whether on the true construction of the policy, the same event(s) which cause the damage to the insured property which gives rise to the business interruption loss are also capable of being or giving rise to ‘special circumstances’ for the purposes of allowing an adjustment of the same business interruption loss within the scope of the ‘Trends clause’”.

The appeal was heard by Hamblen J. In relation to the first question of law, he noted at [20] of his judgment, the insurers’ argument (which he appears to have accepted) that the question was moot, because the tribunal had not excluded recovery of losses concurrently caused by damage to the hotel and damage to the vicinity but only losses which would have been suffered in any event but for damage to the hotel. He said that at the hearing it became apparent that the critical issue of law was the appropriateness of applying the “but for” causation test in the case. At [21] he said that OEH accepted that the normal rule for determining causation in fact was the “but for” test, generally a necessary but not a sufficient condition. He then set out the applicable principles in relation to causation in fact as summarised in Clerk & Lindsell on Torts and McGregor on Damages. The end of the citation from McGregor referred to exceptions to the “but for” test in the interests of fairness and reasonableness in the context of negligence claims:

“The typical situation where an extension of liability may prove necessary in the interests of fairness and reasonableness, with a consequent departure from the “but for” test, is where two or more acts or events or agencies are involved and the wronged claimant is unable to prove which act, event or agency has caused the harm.”

At [24], the judge recorded that OEH submitted that this was such a case requiring the relaxation of the standard, referring to the judgment of Lord Nicholls in Kuwait Airways Corporation v Iraqi Airways Co (Nos. 4 and 5) [2002] 2 AC 883 where at [73]-[74] he said:

“…Even the sophisticated variants of the “but for” test cannot be expected to set out a formula whose mechanical application will provide infallible threshold guidance on causal connection for every tort in every circumstance. In particular, the “but for” test can be over-exclusionary.

74.

This may occur where more than one wrongdoer is involved. The classic example is where two persons independently search for the source of a gas leak with the aid of lighted candles. According to the simple “but for” test, neither would be liable for damage caused by the resultant explosion. In this type of case, involving multiple wrongdoers, the court may treat wrongful conduct as having sufficient causal connection with the loss for the purpose of attracting responsibility even though the simple “but for” test is not satisfied. In so deciding the court is primarily making a value judgment on responsibility. In making this judgment the court will have regard to the purpose sought to be achieved by the relevant tort, as applied to the particular circumstances.”

At [25] Hamblen J recorded that OEH contended that the same approach should be applied in contract, contending that this was “a case of two concurrent independent causes in relation to which the application of the “but for” test would lead to the untenable conclusion that neither of the causes caused the business interruption loss.” He then set out a further passage from Clerk & Lindsell relied upon by OEH dealing with “Successive Sufficient Causes” and a passage from Hart & Honoré on Causation in the Law which posits the case of two causes being present each sufficient to bring about the same harm, giving amongst other examples that of two men firing guns at their victim simultaneously both of which lodge in his brain.

At [29] the judge said:

“Although OEH cannot point to any insurance or indeed contract case in which it has been held to be inappropriate to apply the "but for" test, it relies on the generally accepted principle that where there are two proximate causes of a loss an insured can recover on the basis that it is sufficient that one of the causes was a peril insured, provided that the other cause is not excluded – see J J Lloyd Instruments Ltd v Northern Star Insurance Co Ltd (The Miss Jay Jay) [1987] 1 Lloyd’s Rep 32. Whilst to date this has been a principle applied in respect of concurrent interdependent causes, OEH submits that it should equally be applied to concurrent independent clauses [sic].”

He then referred to OEH’s argument that some support for that approach was to be found in IF P&C Insurance Ltd v Silversea Cruises Ltd (“The Silver Cloud”) [2003]

Hamblen J then set out those difficulties, first of which was that it was agreed under this policy, by the trends clause, that a “but for” approach to causation should be adopted to the assessment of loss of revenue. Second was that whether fairness and reasonableness required that the “but for” test should not be applied was very much a matter of fact for the arbitration tribunal not for the court on an appeal limited to questions of law. At [37] he noted that the issue had not been addressed by the tribunal which had made no findings in relation to it. Third, he said at [38] that he was not satisfied that fairness and reasonableness did require that the “but for” test was not applied. He gave various examples of alternative scenarios such as an undamaged hotel in an undamaged city but concluded that since that would measure gross operating profit which OEH would have made if the hurricanes had not struck at all, that would compensate OEH for all business interruption losses even if not caused by the damage and thus not recoverable under the main insuring clause.

He continued at [39]:

“Further, it is not the case that the application of the “but for” test means that there can be no recovery under either the main Insuring Clause or the POA or LOA. If, for the purpose of resisting the claim under the main Insuring Clause, Generali asserts that the loss has not been caused by the Damage to the Hotel because it would in any event have resulted from the damage to the vicinity or its consequences, it has to accept the causal effect of that damage for the POA or LOA, as indeed it has done. It cannot have it both ways. The “but for” test does not therefore have the consequence that there is no cause and no recoverable loss, but rather a different (albeit, on the facts, more limited) recoverable loss.”

He thus concluded that the tribunal had not erred in law in adopting the “but for” approach to causation and answered the first question of law in the affirmative. In relation to the second question of law, the judge set out in detail the various arguments advanced by Mr Schaff QC (who had not appeared for OEH at the arbitration) in relation to the trends clause. It is not necessary for present purposes to repeat that detail

The judge dealt with OEH’s fifth argument at [51]:

“Fifthly, OEH submits that Generali’s argument has the remarkable result that the more widespread the impact of a natural peril, the less cover is afforded by the business interruption policy for the consequences of damage to the insured property. So, if the tsunami or hurricane or fire only affects the insured property, this will give rise to a full, unadjusted recovery for all business interruption loss caused by the damage to the insured property; yet, should the tsunami, hurricane or fire cause additional devastation to a whole swathe of properties in the same area, the claimant is somehow to be worse off because it is to be treated as if its insured property had (hypothetically) escaped damage but as if its business would still have suffered a loss of income due to the damage (or resulting loss of attraction) in the wider area. However, under this policy the amount recoverable under the main insuring clause will always depend on the extent to which the business interruption losses claimed are caused by damage. That is what the main insuring clause requires as a matter of causation. It has nothing to do with whatever other devastation the hurricanes may have caused elsewhere than at the insured property.”

In relation to the sixth argument, the judge said this at [52]:

“Sixthly, OEH submits that Generali’s approach subverts first principles in that it involves seeking to strip out from the claim for business interruption loss caused by insured damage, not merely the concurrent consequences of extraneous circumstances but the concurrent consequences of the very peril that caused the damage which was a proximate cause of the business interruption loss in the first place. However, the relevant insured peril is the damage; not the cause of that damage.”

He elaborated that reasoning at [57]:

“I agree with the tribunal that the clause is concerned only with the damage, not with the causes of the damage. What is covered are business interruption losses caused by damage, not business interruption losses caused by damage or “other damage which resulted from the same cause”. Nowhere in the Trends clause does it state that “variations or special circumstances affecting the Business either before or after the Damage or which would have affected the Business had the Damage not occurred” has to be something completely unconnected with the damage in the sense that it had an independent cause to the cause of the damage. The assumption required to be made under the Trends clause is “had the

Damage not occurred”; not “had the Damage and whatever event caused the Damage not occurred””.

He then concluded at [60]:

“The scheme of the policy is that business interruption losses caused by damage to insured property are recoverable under the main insuring clause (as is consistent with the Trends clause). Other losses not caused by damage (i.e. physical damage to the hotel) but caused by damage to the city or lack of demand are recoverable under the Loss of Attraction and Prevention of Access extensions. That is what OEH paid premium for under the Policy and that is what the tribunal held that OEH is entitled to recover.”

We consider that there are several problems with the reasoning in Orient Express. First and foremost, as we see it, there was a misidentification of the insured peril. It was an

The “proximate” cause is not the cause nearest in time to the loss but the “efficient” or “dominant cause”: see the torpedo case, Leyland Shipping Ltd v Norwich Union Fire Insurance Society Limited [1918] AC 350. Clarke, The Law of Insurance Contracts at §25-3 in the section dealing with The Proximate Cause notes in an evidently critical way that Orient Express applied: “the but for” usually the test for “factual causation” in the law of tort to a business interruption claim … without reference to established insurance precedent such as the Leyland case.”

If the question had been asked what was the proximate cause of the loss claimed, it seems to us that the correct answer would not have been business interruption arising from Damage in the abstract but business interruption arising from Damage caused by the hurricanes as a covered fortuity. If the policy wording had contained express references to the fortuities for which cover was provided and had identified Storm as one of those fortuities, it seems to us unarguable that the insured peril would be anything other than the business interruption arising from Damage as a consequence of the hurricanes which constitute Storm and we cannot see that it should make any difference in principle to the correct assessment of the insured peril that the policy was an all risks one which did not need to list covered fortuities.

Second and allied to the first point is that on the insurers’ analysis as accepted by the tribunal and the judge, the more serious the fortuity, the less cover the policy provides for the consequences of damage to the insured property, which was OEH’s fifth argument. As OEH said, if the hurricanes had only damaged the hotel, there would have been full recovery for the business interruption loss suffered as a consequence of the damage, but because the overall devastation caused by the hurricanes was to the whole city and surrounding area, OEH was on the insurers’ analysis worse off because it was to be treated as if, hypothetically, the hotel had escaped damage, but its business would still have suffered the same loss because of the devastation wrought by the hurricanes everywhere else. Given that the hurricanes will inevitably have caused widespread damage to the area and not be confined to the insured’s property, the effect of the insurers’ analysis was that the business interruption cover was rendered illusory. We do

On the basis that the hurricanes were an integral part of the insured peril, we consider that when it came to the construction of the trends clause, the judge should have concluded that the words: “had the Damage not occurred” meant that the counterfactual was one where both the damage to the hotel and the hurricanes and their effect generally were to be stripped out.

The decision in Orient Express has been the subject of academic criticism. We have already referred to Clarke. Professor Merkin in Colinvaux’s Law of Insurance (12^th ed) at §24-107 describes it as a “curious outcome that, the greater the damage to the vicinity and thus of the risk of depopulation, the less prospect there is of any recovery by the assured”. Riley on Business Interruption Insurance (10^th ed) at §15-21 states that “there must be doubt over whether it is actually a satisfactory outcome for either insurers or policyholders”. The editor makes the point that when Main Street in Cockermouth, Cumbria, flooded in 2009, insurers did not seek to argue that none of the businesses could recover much because, but for the flooding of their business the rest of the street would have been closed and effectively a building site for approaching six months, anyway. However, on the analysis accepted by the judge in Orient Express, any given insured business in Cockermouth would not have recovered for the business interruption loss it suffered because insurers could have argued that it would have suffered the same loss because even if its business was not flooded, the rest of the street was closed. In our view, the consequence which flows from the Orient Express decision, that the worse the fortuity which befalls the insured and the vicinity of the insured’s premises, the less the insurance responds, cannot have been intended.

It follows that, if we had thought that the decision in Orient Express somehow dictated the consequences in terms of cover and the counterfactual analysis for which the insurers contend in the present case, we would have reached the conclusion that it was wrongly decided and declined to follow it. However, we do not need to go that far because even if Orient Express was correctly decided, it is clearly distinguishable from the present case in relation to those wordings which we have concluded provide cover in principle. Orient Express was simply not concerned with the type of insured perils involved here, including in particular the composite or compound perils which feature in many of the policies which we have had to consider, and nothing in the judge’s analysis has any impact on the correct construction of the wordings we have been considering.

Thus, in relation to prevention of access clauses, the insured peril is, for example, a composite one involving three interconnected elements: (i) prevention or hindrance of access to or use of the premises (ii) by any action of government (iii) due to an emergency which could endanger human life and for cover to be triggered that composite insured peril must have caused the interruption or interference with the business. As we said in relation to the arguments advanced by Mr Kealey QC in relation

In relation to the ‘hybrid’ clauses, we have set out, in some detail, in the context of the Hiscox wordings involving a ‘public authorities’ clause, the way in which we consider that the counterfactual must be applied. The composite peril involves (i) inability to use the insured premises (ii) due to restrictions imposed by a public authority (iii) following, here, the occurrence of a human infectious or contagious disease. To the extent that any insured can show that there was a relevant restriction and an inability to use the premises, in assessing what loss the insured can recover, each of these interconnected elements should be removed from the counterfactual.

Similarly, in relation to the disease clauses where we have concluded that there is cover in principle, we have done so because we consider that on the correct construction of those wordings, they insure the effects of COVID-19 both within the particular radius and outside it, the whole of the disease both inside and outside the relevant area has to be stripped out in the counterfactual. One of the fundamental fallacies in the insurers’ approach is to treat the occurrence of COVID-19 within the relevant radius or

“Vicinity” of the insured premises as completely separate from its occurrence elsewhere in the country as a whole. As we have said in our analysis of several of the disease clauses, the proximate cause of the business interruption is the notifiable disease of which the individual outbreaks form indivisible parts, in other words the disease in the UK is one indivisible cause.

The alternative analysis, although we regard it as less satisfactory, is that each of the individual occurrences was a separate but effective cause, so that they were all effective because the authorities acted on a national level, on the basis of the information about all the occurrences of COVID-19. As we have said, it is artificial to say that only some of those occurrences of COVID-19 which had occurred by any given date were effective causes of the action taken at that date; and still more artificial to say that because the action was taken in response to all the cases, it could not be regarded as taken in response to any particular case.

On this analysis, it is not necessary to consider cases in which there are two effective causes of the loss, one of which is covered and the other which is not, such as JJ Lloyd

A fortiori the principle established by cases such as Wayne Tank and Pump Co Ltd v Employers Liability Assurance Association [1974] QB 57 that where there are two concurrent proximate causes, one insured and one excluded, the policy does not respond, is not relevant since on our analysis however many proximate causes there are they are all insured. It necessarily follows that the submissions of insurers about interdependent concurrent causes and independent concurrent causes are of no relevance, given our conclusions as to the correct construction of the policy wordings where we have concluded cover is available in principle.

Prevalence

Certain policies with which we are concerned require, at least on the FCA’s case, proof of the occurrence or manifestation of COVID-19 within a specified geographical area which includes the insured premises. Issues therefore arise as to what cases of the virus there were in any given area at any given time.

The possibility of the parties calling expert evidence as to the actual prevalence of the COVID-19 disease across the UK was considered at the first CMC on 16 June 2020 and the second CMC on 26 June 2020. This course was strongly opposed by insurers, partly on the basis that they had not understood that there would be factual determination of such matters at the scheduled hearing and that there was not time for expert evidence to be prepared and heard. At the first CMC, Butcher J ruled that there should not be expert evidence going to the issue of actual prevalence at the hearing fixed for July but said that that decision would not preclude argument as to whether a type of proof could in principle be sufficient to satisfy whatever burden was on the insured, nor arguments based on assumptions as to actual prevalence.

At the second CMC, the Court contemplated that it might be desirable to have a further hearing, mooted for September, to decide questions of fact after hearing expert evidence. However, it was decided that the issues for determination at the July hearing would be limited to the following matters (The Financial Conduct Authority v Arch Insurance Limited and Ors [2020] EWHC 1724 (Comm), Ruling 4):

“1.

… first of all, arguments as to the type of proof which could be sufficient to discharge a burden, and that is essentially what Mr Gaisman calls methodology; and, secondly, on the assumption that this is the best evidence available, and by “this” I mean the matters pleaded by the FCA, would it be sufficient as a matter of principle to discharge the burden of proof placed upon particular insureds? …”

Subsequently, the FCA confirmed that it would proceed on the basis described above. In compliance with these rulings, the List of Issues agreed between the parties framed the prevalence issue as follows:

“As per the ruling at the Second CMC, save where now agreed between the parties:

1.

The type(s) of proof which could be sufficient to discharge the burden of proof upon insureds; and

2.

On the assumption that the matters pleaded by the FCA represent the best evidence available, whether it is sufficient as a matter of principle to discharge the burden of proof.”

The hearing proceeded accordingly, and no expert evidence was heard on the question of prevalence. The Court is therefore not in a position to make any findings of fact about the actual prevalence of the disease at particular dates or in particular locations.

The FCA submitted that there are four types of evidence which are sufficient in principle to discharge the burden of proof on an insured as to the prevalence of COVID19 in what it refers to as a “relevant policy area”. These categories of evidence are described below, together with the relevant concessions made by the insurers.

First, and least contentiously, there may be specific evidence of a case or cases in a particular location within the relevant policy area. The example given by the FCA is “widespread reports that the care home down the road from the policyholder’s premises has been the location of an outbreak of COVID-19”. The insurers accepted in principle that an insured might be able to prove a case of COVID-19 at a particular location by specific evidence in a particular case, but submitted that it depends on an assessment of the relevant evidence in each case.

Second, there are publicly available data published by NHS England which record the number of individuals who died in hospitals in England after having tested positive for COVID-19 (“NHS Deaths Data”). The NHS Deaths Data have been published on a daily basis and include data from 1 March 2020 onwards, showing both daily and cumulative totals, aggregated at the level of NHS Hospital Trusts. The FCA accepted that the NHS Deaths Data do not show the numbers of deaths caused by COVID-19 because, for example, an individual might test positive for COVID-19, but then recover and die in hospital from a different cause. The FCA submitted that this possibility does not affect the utility of the NHS Deaths Data in showing the presence of COVID-19 in a particular area. The FCA submitted that if a person contracted COVID-19 and then subsequently died from another cause, that makes no difference to the fact that COVID19 was present in the relevant NHS Hospital Trust at a time prior to the death. In other words, identifying the cause of individual deaths is not necessary for the purposes of proving prevalence.

As noted above, the NHS Deaths Data are aggregated at the level of NHS Hospital Trusts, each of which may contain more than one hospital. There may be situations where one of the Trust hospitals is within the relevant policy area whereas others are outside it. The parties were agreed that if there was at least one death in an NHS Hospital Trust, where there was only one hospital in that NHS Hospital Trust, and that hospital was within the relevant policy area, this would show the death of a person or persons who previously tested positive for COVID-19 within the relevant policy area. Similarly, if there was at least one death in a particular NHS Hospital Trust and all the hospitals in that NHS Hospital Trust were within the relevant policy area, this would show the death of a person or persons who had previously tested positive for COVID19 within the relevant policy area.

Third, there are publicly available data published by the Office of National Statistics (“ONS”) as to deaths which occurred each week in England and Wales. These statistics include deaths where the death involves COVID-19 on the basis that the death certificate mentions COVID-19 (“ONS Deaths Data”). The ONS Deaths Data are published on a weekly basis and are aggregated at the level of local authority or health board. Unlike the NHS Deaths Data, there is no dispute that the ONS Deaths Data indicate deaths where the cause of death involved COVID-19.

The insurers accept that an insured can prove the presence of at least one case of COVID-19 within a relevant policy area in a particular week if the ONS Deaths Data showed at least one death during that week within the local authority or health board, and that local authority or health board was entirely within the relevant policy area. There was disagreement, however, as to situations where the relevant policy area did not neatly correspond with the local authority or health board. In such cases, the FCA submitted an “averaging methodology” could be used to extrapolate the number of deaths in the relevant policy area. We discuss the use of an averaging methodology further below.

The insurers did not accept that the ONS Deaths Data could, on its own, show the presence of COVID-19 within the relevant policy area on any particular day of that week. But nonetheless the parties agreed that in principle if the ONS Deaths Data showed at least one death in a particular week in a relevant policy area, then it could be inferred that there was at least one case of COVID-19 in the period “immediately prior” to that week. The insurers submitted that the length of that period was not something the Court could decide without expert evidence.

Fourth, and perhaps most obviously, there are publicly available data as to the numbers of lab-confirmed positive tests of COVID-19 across the UK (“reported cases”). Daily and cumulative totals of reported cases have been published by the UK Government and are available online. The reported cases for each day originate from the testing of specimens taken in the previous few days. The parties agreed that data from around 5 days before can be considered complete. The reported cases can be aggregated across nations, regions, Upper Tier Local Authorities (“UTLAs”) and Lower Tier Local Authorities (“LTLAs”). The allocation of a case to an area is done by reference to the home postcode of the individual who tested positive, established by ONS geographical area codes.

The insurers conceded that the reported cases on a particular date could demonstrate the presence of COVID-19 in a particular nation, region, UTLA or LTLA on that date. The relevance of cumulative totals was in dispute, however, because, as the FCA acknowledged, cumulative totals make no allowance for those who have recovered. The parties did agree, however, that a person will be infectious for a period of time, estimated to be 7-12 days in moderate cases and up to 14 days on average in severe cases. The FCA therefore submitted that an insured should not be limited to reliance on daily reported cases alone.

The most granular breakdown of daily and cumulative reported cases is at the level of LTLAs, of which there are 317 in England. A relevant policy area may overlap with LTLAs in different ways. The clearest case is where an LTLA is entirely within a relevant policy area. The parties agreed that an insured can prove the presence of at least one case of COVID-19 within a relevant policy area on a particular date if, on that date, the daily reported cases for the relevant LTLA are at least one and the LTLA is entirely within the relevant policy area. Whether the insured premises are within the boundary of the LTLA is neither here nor there.

In other cases, the relevant policy area might cover more than one LTLA. This is to be expected in the case of relevant policy areas with a 25 mile radius. The parties agreed that in such cases, an insured can prove the presence of at least one case of COVID-19 within the relevant policy area on a particular date if the daily reported cases for any one of the LTLAs in the relevant policy area is at least one, so long as that LTLA is entirely within the relevant policy area. Again, whether the insured premises are within the boundary of the LTLA is neither here nor there.

In some cases, the insured may not be able to prove the required occurrence of deaths or cases in a relevant policy area based on the types of evidence identified above. To address these gaps the FCA relied on certain methodologies which it submitted could be used by insureds, in conjunction with the available types of evidence identified above, to prove the prevalence of the disease in a relevant policy area. The first is the use of averages to determine prevalence in a relevant policy area, when the LTLA (or other area for which data are available) is not entirely within that relevant policy area. The second is attempting to establish an “undercounting ratio” to extrapolate the extent to which actual cases of COVID-19 exceeded the reported figures in particular areas and across the UK and using an “uplift” to accommodate for this.

The parties did not reach agreement in relation to situations where the relevant policy area is partly within and partly outside a particular LTLA, or where the relevant policy area is entirely within an LTLA. The FCA submitted that in such cases the insured should in principle be able to use an “averaging methodology” to establish the number of cases in a relevant policy area. The FCA put forward the same methodology in respect of ONS Deaths Data.

The averaging methodology initially proposed by the FCA was to calculate the average number of cases or deaths per square mile in a region, and to use that average to show that there was at least one case in a relevant policy area within that region. This proposal was then qualified by reference to population weighting. The weighted averaging methodology calculates the ratio between: (1) the population of the relevant policy area; and (2) the population of the relevant region for which data are available. The FCA submitted that the populations of different areas could be calculated by reference to publicly available data, including at the level of postcodes and higher level census data. The ratio could then be used to extrapolate the number of deaths or cases in the relevant policy area.

For example, if there were 100 cases in Area A, which has a population of 50,000, and the relevant policy area, Area B, has a population of 5,000, then the weighted averaging methodology would extrapolate that Area B had 10 cases (assuming in this example that Area B was entirely within Area A). The FCA gave the following illustration:

“If the policyholder wished to take (say) 16 March 2020 as the relevant date for making a claim: in Cornwall as at 16 March 2020, there was a total of 14 cumulative Reported Cases …. The population of Cornwall based on the 2011 census was 532,273; and the population of a 25-mile Relevant Policy Area from the above postcode is approximately (based on public tools as to postcode populations) 350,000. That would amount to approximately 9-10 cases in the Relevant Policy Area.”