Business Interruption Insurance Disease Extensions and the CoViD-19 Pandemic Part 1 – Context & Background

Introduction

As with most things in life it takes a major event for people to sit up, take notice, and focus on what could have been done to prevent it or mitigate its impact.

And people have short memories. Disasters fade from view and life goes on – until the next catastrophe.

For 5,000 years insurance has existed as a risk pooling mechanism to encourage trade and soften the blow when the inevitable happens. Whether it be fire, flood, pestilence or plague, societies have created structures to cope; yet we continue to come up short.

It has been 20 years since the terror attack on the USA cost insurers an estimated USD40bn. Terrorism, as an exclusion, was not in widespread use (although war, in its various guises, was). Although the 9/11 event was unprecedented (and insurers initial reaction was to look for ways to avoid or limit liability, e.g. the World Trade Centre case), the insurance market ultimately did respond, and, with government support, insurance pools are now in place to cope with any repeat.

It has been 100 years since the last major infectious disease pandemic, the Spanish Flu, spread around the world. Everyone knows about it (and other subsequent, more subdued virus spreads) yet, when CoViD-19 hitched a ride around the globe in record time early 2020, no one was really ready. Research had predicted such an event, recommendations had been made, structures had been designed, but a general apathy meant preparation did not receive the priority it warranted.

The world has been turned on its head and, while insurers provided business interruption insurance cover for such contagions, in reality, the protection offered was materially compromised by inadequate policy design, incomplete definitions and a lack of underwriting focus.

Caught by surprise, the insurance industry’s reaction was predictable – what do our policy wordings actually say and how do we minimise our losses?

In a series of articles forming a full essay, the author looks at the context for, and background to, business interruption insurance, analyses how the insurance industry has responded to demand for this type of protection, details the approach to policy wording construction using a business interruption insurance framework (including the all-important cover engagement and indemnity calculation approaches) and, as this product evolved to meet expanding demand, explores the prevention of access, public authorities and infectious disease extensions in the context of CoViD-19.

The first of these articles sets the scene with business interruption (BI) insurance put in the context of the broader insurance market, followed by backgrounding the development of BI insurance and contingent risk extensions, tracing the insurance industry’s competitive drivers and response to the demand for infectious disease cover, then, after a detailed look into the complexities of policy wording construction and calculating interruption losses, focusses in on what is an event, how to deal with business trends when adjusting BI claims, and concluding with a ready reckoner to assist with BI policy wording design, determining cover engagement and quantifying the indemnity.

Later, if time permits, the author intends to follow up with a deeper exploration of insurance portfolio management and product development drivers, how insurers manage risk and exposures, and how all this relates to their fiduciary and social responsibilities. ??

Context

Consequential Loss, Loss of Profits, Profits or Business Interruption (BI) insurance is the most flexible of indemnities commercially available as a risk transfer mechanism. It epitomises, for both insured and insurer, the principle of utmost good faith - as the black & white property loss adjustment transitions into a shades of grey estimation of how the business would have performed but for the damage.

BI insurance has been available as an adjunct to property insurance for at least 300 years (loss of rent being its first embodiment) and in general use for around 100 years - originally as fixed margin insurance then, with accounting standards evolving, as a variable cover adjusted based on business performance.

By the mid-1900s earlier concerns about gaming had fallen away. Businesses large and small now rely on BI insurance just as much as, if not more than, property insurance – as the packaged insurance products available for most sectors, evidence. The two products go hand-in-hand; and BI losses can be multiples of the associated property damage.

While this analysis focusses on what was, until recently, one of the more obscure extensions available for a BI insurance policy, i.e. interruption due to disease, readers should keep in mind the broader context, i.e. –

1.??????the development and evolution of financial products,

2.??????the fiduciary role the insurance industry plays in protecting lives and livelihoods, and

3.??????corporate conduct and social responsibility.

As society evolves, so do expectations. For some time we have been traversing a socio-political watershed between the economic neo-liberalism of the last 30 – 40 years and a new wave of competition, consumer and environmental concerns. Not that anything is particularly new or different in this latest societal transition. But clearly more evident is a growing understanding of the downside of unfettered capitalism and light-touch regulation. Access to information on, and the greater transparency delivered by, the internet,[1] is driving this generational shift – which includes earlier generations who realise their faith in corporations (conduct) and government (policy) was misplaced.

This new era comes with a heightened expectation that insurers will appropriately balance their fiduciary, environmental, social and governance responsibilities when designing products, setting rates, underwriting risks and handling claims; and that they will partner with society and government to ensure best outcomes. This ranges from access to affordable insurance with sustainable pricing all the way through to fair claim settlement and a graduated, thoughtful approach to shifts in insurance availability with an end focus on resilience and recovery, especially in relation to disasters.

Outside of direct regulation, governments also have a role to play with risk mitigation and transfer in support of an active, stable and relevant insurance sector. Governments can and do act as insurer of last resort (not just for ‘uninsurable’ risks like war and terrorism, but in managing insurance cycles), they create demand through legislation (e.g. compulsory insurance), they reduce risk through broader regulatory activity (e.g. controlling land development and construction quality), and support risk transfer alternatives (e.g. through government insurance offices, reinsurance arrangements, public schemes, risk pools and mutuals).[2]

Historically, private insurance markets have experienced periods of instability where availability contracts and they become less relevant as a result. We are, again, in one such period; with insurers reaction to the impact of CoViD-19 being the latest example of how principles that underpinned the re-regulation wave of the last 20 years have yet to be fully embedded.

Extensions that shifted BI insurance from an indemnity predicated on insured property damage to one including off-premises and non-physical contingencies like supply failure, prevention of access, or murder, suicide and disease occurrences, reflect in microcosm how these dynamics have more recently played out.

The insurance industry has been agile in creating and expanding products to cope with increasingly sophisticated demand, e.g. environmental pollution and cyber insurance. They have also been adept in managing expectations where providing cover for particular risks is beyond their means, e.g. war.

But this aptitude has been compromised and, while it is difficult to pinpoint exactly when this started (the market drivers are more readily apparent), the industry has now found itself in a real bind - through engineered dissonance in policy wordings from attempts to provide cover for some aspects but not others, especially in limiting exposure to catastrophic risk. In reaction to the drafting gymnastics involved and the unintended consequences that resulted, insurers have further backed themselves into a corner with obscure, legalistic arguments that strike at the heart of the insurance compact.

While tighter regulatory oversight has significantly improved the ability of insurers to absorb financial shocks, maintaining shareholder return and sustaining credit ratings still dominate their thinking when responding to market cycles and catastrophe events. Recent history (e.g. the Canterbury Earthquake Sequence) demonstrates that these tactical responses are unsustainable – with customers increasingly disconnected and courts/legislatures responding more pointedly – and, perversely, creating further claim inflation pressure through extended delays, legal battles and repeat claim adjustments.

The risk of irrelevancy for the private insurance market is, arguably, at its highest point since the late 1800s/early 1900s when, along with a raft of first-time insurance legislation, ?government insurance offices were established in Australia and New Zealand in response to capacity, competition and oversight needs. It is not coincidental that this activity followed an extended period of economic disruption which included questionable practices by, and a material contraction in, the private insurance market. But it was also a time when insurance demonstrated its value in catastrophes, e.g. the 1906 San Francisco Earthquake, along with a growing recognition of the need to strengthen balance sheets. That period also saw the introduction of insurance tariffs – the anti-competitive nature of which subsumed by the pressing need for underwriting discipline.

In its latest incarnation,[3] insurers’ interpretation of the BI insurance disease extension (in response to the CoViD – 19 pandemic) may well be the proverbial straw - tipping the scales in the direction of a fixed value, industry indexed, declaratory[4] trigger and parametric compensation (pre- or post-event) future for BI insurance at best; or, at worst, direct government intervention in coverage, rating (incl. risk-rating boundaries) and underwriting across a wide range of general insurance products.

After all, what is the point of fire insurance that doesn’t include bushfire, water damage insurance that doesn’t include flood, wind damage that doesn’t include cyclone, or disease insurance that doesn’t include contagious diseases?

Or, from a government’s perspective, why create demand (e.g. public liability and professional indemnity through registration and licensing requirements) if insurance markets aren’t sustainable?

Note 1: The author recognises the role governments must play in encouraging/enforcing insurance (and broader financial) market stability, not just the capacity of participants to meet their obligations but also in striking an appropriate affordability/availability balance, and notes they need to step up their partnership with insurers in light of the increasing frequency and cost of catastrophe events.

Note 2: The analysis that follows is informed by both the author’s Australasian insurance background and, with Australasian insurance law heavily influenced by English law, United Kingdom developments. Often jurisdictionally inconsistent, USA developments will be referenced only where directly relevant; noting that the standard Insurance Services Office business interruption wording requires actual physical damage to trigger cover, whether the peril be fire, hurricane or disease (making cover for the latter peril almost meaningless).

Background

Businesses can be interrupted (or interfered with) by more than just events causing damage to (incl. loss of use of) property used by the business at the premises. Property damage at suppliers/customers or to utilities and services used or relied upon by the business affects product/service delivery; while ‘non-damage’ events like murder, suicide, defective product (e.g. food poisoning), an outbreak of disease or poor sanitation can affect customer access/willingness to buy.

In fact, occurrences anywhere along the supply, manufacturing and delivery chain can cause business interruption and, in recognition, business interruption (BI) insurance cover has expanded, accordingly.

Irrespective of the type of event, for the BI insurance indemnity to engage there are 4 key tests[5]:

While these tests are expressed in relation to direct damage to property at the premises, they apply, equally, to events causing damage in the vicinity of the premises[6] (with or without damage to property used by or relied upon by the insured) and to the actions of authorities exercising their responsibility (to avoid or limit injury or damage, or control movement) that prevent, restrict or hinder access to businesses (by owners, employees or customers) – where those BI insurance extensions are included.

Three such extensions (in order of recent evolution, expanding from damage into non-damage events, and from their actuality to the apprehension of them happening) have come into focus due to the CoViD-19 pandemic. Starting with their historical context as background, other parts of this essay will analyse how they have become more complex to interpret due to misalignment with the policy wording, drafting inconsistencies and coverage overlap.

The Prevention of Access (PoA) extension was originally designed to respond to business interruption caused by access to the insured premises being prevented, restricted or hindered by property damage in the vicinity arising from both local (e.g. fire or explosion) or wide-area (e.g. windstorm or earthquake) events. It mattered not how access (or egress) was affected, e.g. debris on the street, unwillingness to enter due to actual or perceived danger, or through the action of public authorities; and it did not require the insured’s premises/property to be damaged by the same event.[7]

Government-related actions can be compulsory or advisory, physical or non-physical. The Acts of Civil Authorities or Public Authorities extension expanded the PoA damage events to include closure of or denial of access to insured premises due to loss or damage or for additional nominated reasons (such as murder, suicide or defective sanitation) and, over time and especially for bespoke policy wordings, to any act of a government body or authority for whatever reason. It did not matter whether the action was before (in anticipation of) or after (in reaction to) the event. Like the shift from nominated perils to “all risks” property damage cover, any act by authorities leading to damage/inability to use property and the related business interruption was covered unless otherwise excluded. Note: In its earlier forms, this extension would specifically[8] exclude acts arising from infectious animal and human disease – such cover being the domain of a Disease extension.

The Murder, Suicide and Infectious Disease extension had similar origins and was designed to further extend BI cover (where access issues were involved) to cater for certain instances where customers could not (were denied access by forced closure) or would not (made a voluntary choice) visit the business. Like physical damage by insured perils, these contingencies needed a strong element of unpredictability/uncertainty for insurers to be comfortable offering cover.

Instances of murder, suicide, food poisoning and defective sanitation usually apply to, and only have an impact on, specific premises; with surrounding businesses affected only temporarily, e.g. access/egress restrictions while the event is investigated.

However, a greater degree of difficulty is involved in determining the source of an infectious disease, so limiting it to an outbreak at the insured premises was impractical – hence the use of an, albeit arbitrary, location radius.

Similarly, any access or movement restrictions imposed by authorities to prevent/limit the spread of disease could involve an area much larger than just the immediate neighbourhood and impact disproportionately more businesses.[9]

More often than not these events did not cause property damage. To overcome the limitation imposed by the Damage trigger, these ‘non-damage’ contingencies were deemed by the extension to be ‘damage’, or the policy definition of Damage was extended to include the inability to use the property at the premises, i.e. loss of use, because of such events. Deeming is simpler and was the first approach used; the latter ‘damage’ definition expansion left things open to interpretation and has complicated the CoViD-19 response for some insurers as a result.

The wider the cover footprint - through supplier, customer, public utilities, prevention of access and infectious disease extensions - the greater the exposure aggregation. While the physical location of nominated suppliers and customers were readily aggregable, utility networks and supply chains[10] were less so (other than single points of failure such as major power stations or individual pipelines). Insurers were practiced at calculating natural perils exposures for property damage and business interruption. Catastrophe modelling had advanced due to the increasing number and severity of such events and, while never perfect, modellers continually tested assumptions and applied error margins.

Projecting the potential cost of contagious[11] diseases was not so advanced. Epidemics/pandemics occur infrequently and are usually more widespread than natural peril catastrophes. Because of this, data was incomplete and impacts less predictable, especially as occurrences often skip generations and happen within different economic/social paradigms.

Note: Consideration of insurer aggregation exposures, lateral and vertical, will be discussed in more detail in a future section of this analysis.

The first contemporary pandemic was the early 1980s acquired immune deficiency syndrome [human immunodeficiency virus (HIV)]. The insurance industry accepted the early claims under their life, accident & health products; responding to future risk with more granular underwriting and targeted exclusions in personal accident/income protection policies. While HIV and its acquired immune-deficiency syndrome (AIDS) has caused over 75 million infections/30 million deaths, its spread was gradual and, not being highly contagious, caused only limited business disruption. Reaction to the 1996 avian flu [H5N1 virus] outbreak was similarly muted as acquiring it required close contact with birds and human to human transmission was difficult.

Not so in 2003 when the much more contagious severe acute respiratory syndrome [SARS-CoV virus] epidemic struck, causing insurers to pause and reset their business interruption disease extensions.[12] The 2009 swine flu [H1N1 virus[13]] and the 2013 Middle East respiratory syndrome (MERS) outbreaks reinforced that need – although, as will be seen, the response was more cosmetic than structural.

As the term ‘infectious disease’ is broad, contagions unpredictable, and sources difficult to pinpoint, the impact of an outbreak is not easily measurable. Also, there is no knowing what ‘new’ disease is just around the corner.[14] So insurers realised they couldn’t just carry on with disease extensions without the ability to more accurately calculate and control exposures.

Insureds were also in a quandary. Insurers, having offered cover for infectious diseases (which businesses usually had little chance of anticipating or managing, especially contagious varieties), it was entirely logical that, if a business could not reduce, control or eliminate the risk, then the only option left for them was to transfer it (to those who better understood it and, presumably, had the tools to predict and manage the exposure).

The reality was that, as mentioned, insurers did not have those tools because they did not have the data. Even where they did have reasonable amounts of information, competitive pressures meant resources remained heavily weighted towards insurance classes and customer segments generating the greatest volumes – exacerbating the shortage of skills available to portfolio manage business interruption (and other specialised insurance products).

Note: In Australia, a broad reset of the regulatory regime for business[15] also played a part - with New Zealand enacting similar legislation, as well as obtaining leverage off their larger financial services players having Australian parents.

It is against this backdrop that a harder, more reactionary response by insurers (to an increasing frequency of events adversely impacting margins and threatening balance sheet stability) has come about. Given this pattern of behaviour has been evident for some time, and governments of all persuasions have been slow to react (in underpinning insurance availability and affordability), the way the insurance industry responded to the CoViD-19 pandemic was entirely predictable.

Stay tuned for Part 2 where the author talks about the insurance industry’s approach to dealing with increasing pandemic risk and how the erosion in policy drafting skills compromised that response.

[1] Including the all-pervasive influence of social media.

[2] New Zealand’s Earthquake Commission and Accident Compensation Corporation, The USA’s National Flood Insurance Program and various country terrorism pools (incl. Australian Reinsurance Pool Corporation) are examples, along with government underwritten/regulated workers compensation and motor injury schemes.

[3] Other instances in recent history include the convoluted approach to flood insurance in Australia, insurers narrow interpretation of policy coverage with Canterbury Earthquake Sequence claims, e.g. with auto-reinstatement of sums insured and repair standards, and the almost pathological treatment of income protection claimants (incl. reverse underwriting), especially where mental health issues were present.

[4] e.g. that an act of terrorism, a natural catastrophe (wide-area damage), or epidemic has met a pre-set threshold.

[5] Which are important to remember when constructing policy wordings and extensions – the logic for which will be further explored as the absence of drafting discipline has played a big part in the contagious disease pickle insurers find themselves in.

[6] Including in the vicinity of supplier and customer premises, and public utilities, where these extensions are active.

[7] Like most insurance policy extensions, Prevention of Access arose from an identified gap in BI coverage where there was no damage to the insured’s premises/property (or damage was light compared to surrounding property) but there was a loss of the ability to use it due to the surrounding damage.

[8] Along with the general exclusions such as war, nuclear contamination and terrorism. Note: The absence of such specific catastrophe-style exclusions in more recent versions of this extension has added to the CoViD-19 insurer response complexity, especially with the overlap in cover it creates.

[9] The Prevention of Access extension (due to property damage in the vicinity of the insured premises) started life without a specific area (radius) limitation, even though it was recognised that insured perils can cause damage not just to the insured premises but to property locally (e.g. fire and explosion) or regionally (e.g. earthquake, hurricane and flood). Similarly, Public Authority and Disease extensions originally had no such limitation. Sum insured ($ or %) limits for insurers to control exposures only applied to supplier and customer extensions.

[10] The global impact of the 2011 Bangkok floods being a case in point.

[11] Note: Not all infectious diseases are contagious, hence the use of the latter term to describe diseases with epidemic/pandemic potential (see elsewhere in this analysis for more detail on the rationale, especially when it comes to policy wording drafting).

[12] In 2006 the Australian Prudential Regulation Authority (APRA) required all financial institutions to stress test their businesses with pandemics in mind. The focus was on business/financial system continuity and known exposures but, like the subsequent practice guide, CPG 233, did not extend to embrace the ‘unintended’ consequences of insurer policy wordings.

[13] The same virus that caused the Spanish Flu pandemic.

[14] In particular, as SARS had demonstrated, the transmission of disease, especially viruses, from animals to humans, rapid mutation, and increased transmissibility risk due to industrial scale farming and greater community connectivity, meant vaccines were constantly playing catch-up.

[15] The Australian Prudential Regulation Authority (APRA) Act 1998, Australian Securities and Investments Commission (ASIC) Act 2001 and the Corporations Act 2001.