During the past month there has been one thing dominating the headlines: Covid-19 (also known as Coronavirus).

As the virus continues to spread across Asia, Europe, Africa and the US, health related fears continue to mount. Supply chains have been interrupted; large events cancelled; offices closed; travel plans cancelled and the effects will continue to be felt as the days and weeks go on.

In addition to the obvious concerns as to the virus’ threat to health, there is an increasing concern as to the likely economic impact of the virus. Policyholders are scrutinising their policies to identify potential cover and insurers and reinsurers are seeking to evaluate their potential Covid-19 exposure.

The obvious classes of business facing potential claims are Travel; Life; Health; Event/Contingency; and Business Interruption. However, there is the potential for other less obvious classes of business to be impacted too, such as Errors & Omissions; Directors & Officers; Shipping/Marine; Employer’s Liability; and Public Liability.

To the extent impacted, reinsureds will inevitably be looking at how Covid-19 related losses may be aggregated in order to maximise recovery under any reinsurance. There are, of course, issues such as which policy year will respond, and whether the outwards reinsurance is ‘back to back’ with the inwards claims. However, this article looks at the scope for aggregating Covid-19 losses on the basis of some of the market’s more common aggregation language.

How did the Covid-19 outbreak start?

Covid-19 is a virus. It is a member of the coronavirus family that has never been encountered before. Like other coronaviruses, it has come from animals. Investigations as to the exact source of Covid-19 are ongoing but it is generally understood that many of those initially infected either worked or frequently shopped in a market in Wuhan, China in late 2019 which sold both live and dead animals.


The purpose of aggregation clauses in reinsurance contracts is to enable two or more separately covered losses to be treated as a single loss for deductible or other purposes when linked by a unifying factor.

Aggregation language varies from wording to wording and the potential to aggregate Covid-19 losses will require an analysis of how the unifying factor is defined in the particular reinsurance wording concerned.

Event-based clauses

Whilst there is no standard reinsurance aggregation clause, a common London market clause will provide cover up to specified limits and subject to specified deductibles for “each and every loss”. “Each and every loss” is often defined in a similar form to the below:

“…each and every loss and/or occurrence and/or catastrophe and/or disaster and/or calamity and/or series of losses and/or occurrences and/or catastrophes and/or disasters and/or calamities arising out of one event”

Some aggregation clauses have “occurrence” (rather than “event” as above) as their unifying factor. It is generally considered that the meaning of “occurrence” is synonymous with the meaning of “event” and so the principles applicable to the meaning of “event” are generally applied to the meaning of “occurrence” and vice versa.


The meaning of “event” has been considered in several well-known authorities which have established the following principles:

In the case of Caudle v Sharp it was held an “event” only arises where:

  • There is a common factor which could properly be described as an event.
  • The losses satisfy the test of causation.
  • The losses are not too remote for the purposes of the particular clause defining “each and every loss”.

An event is not a state of affairs, a cast of mind or a course of conduct or inaction.

In the Dawson's Field Arbitration, it was held that the unities of (a) cause, (b) locality, (c) time, and (d) the circumstances and purposes of the persons responsible are aids (no more) in determining whether the circumstances of the losses involve such a degree of unity as to justify their being described as "arising out of one occurrence". The same could be said to apply to event-based language.

“Arising out of”

The meaning of the phrase “arising out of one event” was considered in Scott v Copenhagen Reinsurance Company (UK) Ltd.

As well as confirming the applicability of the four unities test (cause, place, time and intention) for event-based wording, the Court also held that “arising out of” when coupled with “one event” requires more than a “weak causal connection” and that, instead, “a significant causal link” is needed.

Cause-based clauses

Cause-based language is becoming more common and can often be found in US reinsurance contracts. Cover for the aggregation of losses may be provided by reference to the same “originating cause”; “common cause”; or losses “attributable directly or indirectly to one…cause”.

Cause-based language generally allows greater scope for aggregation than event-based language as set out in Axa Reinsurance (UK) v Field:

In ordinary speech, an event is something which happens at a particular time, at a particular place, in a particular way…..A cause to my mind is something less constricted. It can be a continuing state of affairs; it can be the absence of something happening.”

There is further English authority on the meaning of cause-based language, albeit in the context of insurance (as opposed to reinsurance) policies, including Countrywide Assured Group Plc v Marshall. In this case the Court held that “original cause” language allows the aggregation of losses where there is a unifying factor in the history of the claim. Event or occurrence based wording describes ‘what’ has happened, whereas cause based wording describes ‘why’ it has happened.

Covid-19 and Aggregation

So, is Covid-19 a single “event” or “cause”?

There is no direct authority on the aggregation of reinsurance claims related to a single communicable disease/virus, so decisions as to aggregation will need to be undertaken with reference to the specific words of the reinsurance policies in issue; the nature of the underlying claims that the reinsured is seeking to aggregate; and some of the authorities on the meaning of “event” and “cause” as discussed above.

Event-based language

Whilst a conclusion on the exact source of the Covid-19 outbreak is yet to be reached, on current information, it appears that Covid-19 originated from the Wuhan market. Is that enough to aggregate all Covid-19 related losses into one event, though?

The “event” could be deemed to be either the consumption or sale of infected produce, or the negligence in the handling of the meat at the Wuhan market (the “Primary Infection”). However, it will be argued that the time, and even place, unities are not satisfied if it is found that (a) various people were infected by consuming meat at the Wuhan market; (b) that different types of meat, fish, or birds were infected; (c) that they were supplied by different suppliers; and/or (d) they were purchased/consumed on different days or even different times of day. Further, it may be that each supplier selling infected meat or each person infected by the consumption of infected meat is considered a separate event. If event-based language is interpreted in that way, the scope for aggregating losses “arising from” that individual’s consumption will be difficult to trace, and linking losses of say, disrupted travel plans in the UK to the supply of infected meat in Wuhan, is likely to face hurdles in satisfying the “significant causal link” test.

Aggregation of Primary Infection losses with losses stemming from the infection of individuals who come into contact with those already infected (“Secondary Infection”) may run into difficulties in satisfying the unity of “cause”. However, there may be scope for aggregating Secondary Infection losses together where the unities of time and place are satisfied. Whilst Secondary Infections in Wuhan could be deemed one event (where people are infected at a similar time), it will be more difficult to aggregate Secondary Infections in Wuhan with Secondary Infections in, say, Italy. There may even be difficulties in aggregating Secondary Infections in Wuhan, as the virus continued to spread within the City over several months (making unity of time a challenge to satisfy). Where the Secondary Infection has spread beyond Wuhan and across the world, aggregation on the basis of “event” language gets even more difficult (depending on the specific facts/wording at play) and again, the “significant causal link” test is likely to present a challenge in terms of linking Secondary Infections in one country to, for example, disrupted travel plans in another.

Attempts may be made to aggregate losses arising out of the arrival in Europe or even individual countries of the virus but again the unities of time and place will be a challenge. The unity of cause may also be an issue depending on the types of loss. The arrival of Covid-19 in the UK has a far stronger causal link to loss of life in the UK due to infection (and any life assurance claims) than it would to, say, any decision by a sport’s governing body to cancel their sporting events on account of Covid-19 fears. Most non health/life related losses will more likely be deemed to be caused by the responses taken to try to contain the outbreak as opposed to the outbreak itself.

Each decision, by each governing body to cancel their events may well be capable of aggregation as one event but whether they can be aggregated with decisions of other governing bodies or even with employers closing their offices on account of the same Covid-19 fears may be more of a challenge given the problems of time and place. That analysis could change if the cancellations were as a result of a government mandate banning public gatherings and/or the closure of offices (the mandate could be deemed to be one event) but cancellations as a result of different governments making similar mandates based on Covid-19 are less likely to be capable of aggregation under event-based language.

Cause-based language

Aggregation of Covid-19 related losses under event-based aggregation language is inevitably going to be more difficult than cause-based language. We can certainly see it being less problematic to aggregate Covid-19 losses on the basis of “cause-based” wordings. It is likely to be easier to prove that Covid-19 was a unifying factor across all Covid-19 related claims in one class of business e.g. travel insurance claims or even (but perhaps more challenging) across several classes of business.


Whilst investigations regarding the origin of Covid-19 remain ongoing (and focus is quite rightly on public safety) it is still too early to determine the impact of Covid-19 on the insurance market.

In the event that there is significant impact at insurance level, insurers will be looking at ways to maximise their reinsurance recovery and there is potential for this issue to grip the reinsurance market in a way we have not seen for some time.