01 May 2020 | 

Insurance Marine News has today published below article based on an bulletin by Peter MacDonald Eggers QC and Douglas Grant of 7KBW Chambers

Whilst many of our readers may find its content on the technical side, it has potentially a huge impact on how the insurance markets evolve over the next few years.

To those of you unfamiliar with these structures, all but a very few insurers after having underwritten its insurance policies, proceeds to reinsure them.

This is done primarily for two reasons;

Firstly, to model their portfolio so that losses can be maintained within a desired retention level, say the first USD 5 million, a so-called Excess of Loss program. So whilst the exposure the insurer assumes on their individual insurance policies may fluctuate from say 10 to 20 million, the risk they retain would always be capped at the first 5 million. This then means that the retained portfolio of an insurer will consist of a harmonised series of 5 mill exposures, and the risk the insurer ultimately carries will be the frequency of losses; will they be above or below his model premium (and even here there are separate reinsurance solutions that can moderate any excess impact)?

Secondly, the reinsurance is there to remove any catastrophe losses beyond the financial capability of the insurers balance sheet (or a comfort level established by the insurers management or board). With this in mind the underwriting team will have guidelines and limits that prevents them from writing risks where the exposure is beyond the combined sum of their retention and their reinsurance (say 5 million retention plus 15 million reinsurance = 20 million underwriting capacity). This is a fairly straightforward calculation, but then it gets more complicated as risks can accumulate. So, assume the underwriter has written two risks (vessels) at USD 20 million each against his underwriting capacity of 5+15 million, and it turns out those precise vessels end up having a collision where both vessels are lost. This is where the effect of a cumulative event which is discussed in the bulleting below begins to become clear; the insurer now ends up with 40 million dollar of losses and, whilst they come from two vessels, the cause of the loss is one event; the collision. So, the net result of the insurer would be this; 2x20million = 40 million less 5 million retention = 35 million less reinsurance 15 million = uninsured excess amount 20 million. Add to this the original 5 million and the insurers will have retained 25 million, so there goes his harmonised series of containable losses.. The basic structure of the world’s most used marine insurance conditions additionally means that the policy can pay up to three times the actual policy limit (Collision Liability, Salvage Sue and Labour) and the insurers’ problem is exasperated. What was meant to be contained is not without further reinsurance not contained.

So in short this second reason why insurers buys reinsurance is to take off catastrophe losses where accumulation or “event” exposures would otherwise unduly impact on their underwriting result, or indeed take the company down. The problem with catastrophes and event accumulations is it stretches the human imagination. Many of you will have read “The Black Swan: The Impact of the Highly Improbable” the 2007 book by author and former options trader Nassim Nicholas Taleb, and you may also be aware of the basic idea of “Ellsbergs Paradox”: people overwhelmingly prefer taking on risk in situations where they know specific odds rather than an alternative risk scenario in which the odds are completely ambiguous  (“The Devil that you know”).

The Black Swan essentially says that anything that can happen will happen. Ellsberg essentially says, but since we can’t know what this will be, we will focus on what we have seen in the past has happened.

Apply this on marine insurance: you insure against damages to your ship on your Hull and Machinery insurance (risk is to a degree controllable and most claims modest in size), but you do not buy Trade Disruption Insurance which geopolitical events away from your ship (no control and, although few and far between, have the potential of ruining your company). Edge has for many years been in the market with TDI solutions and generally pitched them as catastrophe insurances. The sophisticated organisations that buy them increasingly are looking to make them more primary in nature – that is with lower self-retentions. All the way up to the declaration of Covid-19 as a pandemic TDI policies contained quarantine cover, and it was not at all expensive. The trick here as everywhere is to buy when the risk is not known. This requires a bit of abstract thinking, reading “The Black Swan: The Impact of the Highly Improbable” is quite revealing in this regard.

So, we can work out that two or several ships can collide, we can work out that if a range of jack-up platforms in the Mexican Gulf are all in the path of a force 5 hurricane can all be lost in the same event. But what about other events such as the closing of the Suez Canal, or a pandemic like Covid-19?

This is where the bulletin quoted below gets very interesting. Take our example with the colliding vessels and multiply this with one hundred (or more) for Covid-19, that is the theoretical number of Covid-19 exposed policies written by an underwriter, and apply the analogy of the colliding vessels. The good news for a direct insurer is if this is considered “one event” as considered below, then the direct insurer will pay only one retention (5 million) not 100×5 million and losses in excess of these 5 million will be collected from reinsurers. So far so good. But then we are back to abstract thinking; did the insurer foresee that an event with this level of cumulation could occur? Let’s do the numbers and for the sake of the calculation use the median exposure from the examples above (10-20 million) which is 15 million. 100×15 million = 1.5 billion of losses. If this is a specialist marine hull insurer he is likely to have bought reinsurance to the tune of USD 100 million which means he has USD 1.4 billion of un-reinsured losses coming back at him. He will have burnt through the ceiling of his reinsurance program and must collect from his equity.

Whilst the numbers are purely illustrative the effects are clear to see. This means:

  1. An “event” declaration of Covid-19 lets direct insurers initially off the hook, but
  2. Significantly damages high level and catastrophe reinsurers, who as writers of event cancellations (Olympics etc) are already exposed for very large losses. These are the companies that supply reinsurance capital to the world’s marine insurance markets, they will want their money back which again means H&M rates will not fall anytime soon, further and worse
  3. Any direct insurer who has an unknown accumulation, or a not very vivid imagination, may find himself burning through his reinsurance and taking hits far beyond anything envisage.


It is quite possible that Covid-19 exposure to the marine market is small, but to anyone who writes or find himself significantly exposed this is one to watch. No doubt this debate will find its way into the legal system before the financial fallout of Covid-19 in the insurance market has been finally summed up.




“Aggregation in insurance and reinsurance: Covid-19
Peter MacDonald Eggers QC and Douglas Grant of 7KBW Chambers have written on the matters that could arise relating to aggregation and Covid-19. 7KBW noted that, in a globalized economy and in a globalized world, the Covid-19 pandemic was unprecedented in its impact on commercial operations and on currency, equity, bond and commodity markets.

The types of insurance policies under which claims might be made include Business Interruption, Event Cancellation, Political Risks, Trade Credit, Marine and Aviation, and the full range of Liability policies. Sitting atop all of these products are Reinsurance Policies.

The writers noted that, given the scale and depth of the Covid-19 pandemic and its global effect, one of the most pressing insurance and reinsurance issues was that of aggregation.

“To what extent can separate losses and claims be treated as one (aggregated) for the purposes of applying policy limits and/or deductibles?”

Because aggregation may stand to benefit either the (re)insured or the (re)insurer, where the same aggregation language applies to both limits and deductibles in the one policy, the writers said that aggregation clauses “are not to be approached with a predisposition towards either a broad or narrow interpretation”.[1]

In order to consider the opportunities for aggregation, it is necessary to identify Covid-19 and analyze the governmental and commercial responses to it. Covid-19 is a coronavirus disease. It is caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). SARS-CoV-2 belongs to the Coronavirus family, which is one of seven human pathogens and one of three causing a severe clinical syndrome, the other two being SARS-CoV and (MERS)-CoV. SARS-CoV-2 shares a substantial proportion of its genome with SARS-CoV.

In the midst of WHO’s designation of COVID-19 as a PHEIC on January 30th 2020 and the introduction of WHO guidelines, including for public health measures and infection control, certain national governments and agencies began to take steps to contain or delay the spread of the SARS-CoV-2 virus.

In early March 2020, the UK government introduced the Health Protection (Notification) (Amendment) Regulations 2020 (SI 237/2020), amending the Health Protection (Notification) Regulations 2010, which had the effect of classifying COVID-19 as a notifiable disease, e. imposing an obligation on registered medical practitioners to notify the relevant authorities of any suspicion that a patient has the disease. There were similar amendments in Scotland, Wales and Northern Ireland.

In mid-March 2020 the UK government gave advice on social distancing. This had an immediate impact on a number of commercial operations within the United Kingdom. In late March 2020 the UK government required a large number of businesses to close premises and imposed restrictions on individuals’ movement. Other national governments and agencies and, within a federation, state or provincial governments imposed similar restrictions at different times.

Aggregating language in policies
The writers noted that there were various means by which insurance and reinsurance policies permit aggregation. Aggregation can be achieved by requiring the parties to apply a single limit or deductible to a number of separate losses or claims where such losses or claims arise out of a single triggering “event”, “occurrence”, or “originating cause”. Other aggregating language is found, but these are the most common.

Event or Occurrence
The most common type of aggregation clause is one that groups losses or claims by reference to a single “event” or “occurrence”. These terms are usually interchangeable, but the context may dictate that they be given different meanings.[2]

An “event” or “occurrence” is something that happens “at a particular time, at a particular place and in a particular way”.[3] Thus, in Caudle v Sharp,[4] the Court of Appeal observed that while the Second World War, the Hundred Years’ War or the Ice Age might in common parlance be described as an “event”, they cannot be treated as such for the purpose of an aggregation clause in the insurance or reinsurance context, because an “event” requires “some causative element” and a “lack of remoteness” depending on the circumstances of the particular case.

Evans LJ said that there are three requirements of a relevant “event”:

  • there was a common factor which can properly be described as an event,
  • that event satisfied the test of causation giving rise to the loss or claim, and
  • the event was not too remote for the purposes of the aggregation clause.


The “event” or “occurrence” in question need not be the peril insured against. Nor need it be the proximate cause of the loss; rather, “the causative link has to be a significant rather than a weak one”.[5]
The writers said that, in determining whether losses arise from a single event, the court would often apply the concept of “unities”, first introduced by the then Michael Kerr QC in the Dawson’s Field Award (March 29th 1972) and adopted by Rix J in Kuwait Airways Corp v Kuwait Insurance Co SAK (No. 1).[6] The unities are those of cause, locality, time and the intention or motive of any human agents.

The writers said that, while the unities were a useful guide, they should not be treated as an inflexible test. The Court would approach the matter from the perspective of the insured and scrutinise the circumstances “to see whether they involve such a degree of unity as to justify their being described as, or arising out of, one occurrence”. This involves an “exercise of judgment” to be conducted “globally and intuitively”.[7]

In IF P&C Insurance Ltd v Silversea Cruises Ltd,[8] an operator of luxury cruises was insured against loss of income resulting from government warnings regarding terrorism. This cover was subject to a “per occurrence” deductible of US$250,000. Following the 9/11 attacks, the US government issued a series of warnings to their citizens against travel abroad. The question arose how these claims should be aggregated. Tomlinson J observed that it would be “absurd” to treat individual government warnings as a separate occurrence, in part because it would be impossible to discern the causal effect of each warning on bookings. The judge equated “occurrence” with “event” and said: “Where there are multiple warnings arising out of a single defining event, at any rate one of the magnitude of 11 September, it seems to me to accord with common sense and what the parties’ intention must have been to regard those warnings … as a single occurrence, since they all arise out of the same set of circumstances, both actual and threatened.”

Originating cause
The insurance or reinsurance policy might instead aggregate losses arising out of the same “originating cause”. The writers noted that this would have a far wider reach, as Lord Mustill explained in AXA Reinsurance (UK) plc v Field: [9] “A cause to my mind is something altogether less constricted. It can be a continuing state of affairs; it can be the absence of something happening. Equally, the word “originating” was in my view consciously chosen to open up the widest possible search for a unifying factor in the history of losses which it is sought to aggregate. To my mind the one expression [originating cause] has a much wider connotation than the other .”

In Standard Life Assurance Ltd v ACE European Group,[10] the policy used the words “originating cause or source”. The Court said that “the words ‘or source’, as an explicit alternative to ‘cause’, can only have been included to emphasise yet further that the doctrine of proximate cause should not apply and that losses should be traced back to wherever a common origin can reasonably be found.”

Can losses or claims arising from COVID-19 be aggregated?
Very often, when considering issues of aggregation, the relevant event or originating cause can be quite localised. Indeed, on occasion the Court has been reluctant to extend the reach of an aggregating event too far.[11] Nevertheless, large-scale events can be an aggregating cause, such as a hurricane or an earthquake or the 9/11 attacks.[12]

What is the aggregating event or cause when it comes to claims emerging from the COVID-19 pandemic? Can all losses or claims arising from one global pandemic be treated as one loss or claim? If so, it would be a first.

Nevertheless, it might be said that it is a clear inter-connected occurrence – an infectious human pathogen with a global reach – and it is clearly identifiable as a cause. On the other hand, it might equally be said that it is not the pandemic which is the cause but the governmental responses to the pandemic by closing business premises and restricting the movement of people. In IF P & C Insurance Ltd v Silversea Cruises Ltd[2004] Lloyd’s Rep IR 217, the Court could not distinguish between the 9/11 attacks and the US governmental response. However, the writers said that this might be too simplistic. The issue might depend on the impact of the UK governmental action upon each type of business or industry and on the location of each commercial operation. Further, the effect of each national government’s action (not just that of the UK government) might have to be distinguished.

In order to determine the prospects of aggregation, it would be necessary to consider all relevant factors, such as

  1. the type of loss, damage or liability against which the insurance or reinsurance is affording protection,
  2. the losses suffered by the insured,
  3. the relevant industry or profession,
  4. the nature, scale and location(s) of the business or network being insured,
  5. the proximate and remote causes of the loss in question (the pandemic, the relevant governmental action, and the insured’s response to the pandemic and governmental action), and, of course,
  6. the policy language.


The writers concluded that aggregation could have a very substantial impact on the sums recoverable by the insured or reinsured”. While the case law provides guidance on aggregation clauses, parties will need to consider with care the above factors, in this novel and unfamiliar territory.”

[1] AIG Europe Ltd v Woodman [2017] 1 WLR 1168, at [14].
[2] Midland Mainline Ltd v Eagle Star Insurance Co Ltd [2004] Lloyd’s Rep IR 22, at [75]; rev’d on other grounds [2004] 2 Lloyd’s Rep 604.
[3] AXA Reinsurance (UK) plc v Field [1996] 1 WLR 1026, 1035. In Midland Mainline Ltd v Eagle Star Insurance Co Ltd [2004] Lloyd’s Rep IR 22, at [97], the Court said that a decision or a plan cannot constitute an event or occurrence, as opposed to its promulgation and implementation.
[4] [1995] LRLR 433, 438.
[5] Caudle v Sharp [1995] LRLR 433, 439; Scott v Copenhagen Reinsurance Co (UK) Ltd [2003] Lloyd’s Rep IR 696, [63]. See also Simmonds v Gammell [2016] 2 Lloyd’s Rep 631.
[6] [1996] 1 Lloyd’s Rep 664, 686.
[7] Scott v Copenhagen Reinsurance Co (UK) Ltd [2003] Lloyd’s Rep IR 696, at [81].
[8] [2004] Lloyd’s Rep IR 217, at [66]; [2004] Lloyd’s Rep IR 696.
[9] [1996] 1 WLR 1026, 1035. See also American Centennial Insurance Co v INSCO Ltd [1996] LRLR 407; Municipal Mutual Insurance Ltd v Sea Insurance Co Ltd [1998] Lloyd’s Rep IR 421. By contrast, without the word “originating”, the Court might interpret the “cause” behind the loss or claim as a proximate cause, as the New Zealand High Court recently did in Moore v IAG New Zealand Ltd [2020] Lloyd’s Rep IR 167, at [52]-[58].
[10] [2012] Lloyd’s Rep IR 655, at [259]; [2013] Lloyd’s Rep IR 415.
[11] See, e.g., Mann v Lexington Insurance Co [2001] 1 Lloyd’s Rep 1 (a riot over two days in Jakarta was not one occurrence).
[12] Caudle v Sharp [1995] LRLR 433, 438-439; IF P & C Insurance Ltd v Silversea Cruises Ltd [2004] Lloyd’s Rep IR 217; Moore v IAG New Zealand Ltd [2020] Lloyd’s Rep IR 167.”

SOURCES:        Peter MacDonald Eggers QC and Douglas Grant of 7KBW Chambers

Insurance Marine News

Illustration from 1 May 2019

Return to News