Insider In Full: Coronavirus: The biggest insured loss in history?
After a rout that started in February and ran for most of March, financial markets have stabilised thanks to government stimulus and signs of flattening curves in many major economies...
But the easing of fears for (re)insurers on the asset side of the balance sheet has been cancelled out by a sharp deterioration in the industry's private assumptions about the scale of underwriting losses.
Following a week of conversations with senior industry figures, we would stress four central points where the picture has shifted:
- Quantum: Talk of coronavirus as the biggest insured loss event in history is now relatively common, with a number of sources pointing to an $80bn-$100bn total, and few now believing that it will generate claims below the $40bn that came from the World Trade Center.
- SME cat: The significant ramp-up in loss expectations primarily reflects the realisation that there is a more widespread issue than initially thought around the breadth of BI cover on SME policies, making a highly asymmetrical loss by country and by insurer increasingly likely.
- Treaty T&Cs: Reinsurers are getting ready to draw the battle lines to resist these property claims, creating an escalating risk of an asymmetrical loss between insurers and reinsurers.
- Retro trapping: Despite the brewing fight on wordings between insurers and reinsurers, which is likely to take years to resolve, there is enough fear around losses on the reinsurer side that some carriers intend to aggressively trap collateralised retro (roughly 60 percent of the market).
Even 10-14 days ago the sector was still talking about a medium-sized US hurricane as a reasonable proxy for the event, but the idea that this is an Irma-sized ($33bn) event has been dropped entirely.
"The mood change is stark," a senior broking source said.
We have described it as a fool's game in the past to try to estimate the insured losses of storms that are days away from landfall. And coronavirus is a much more complex event – it operates across multiple lines of business and countries, is poorly modelled, and interacts with myriad different wordings and regulatory and judicial approaches – only multiplying the folly.
All of our sources were also working from different scenarios for the duration and scale of the disease, as well as the depth and length of the recession. And they were also building the loss differently from its component parts (event cancellation, property, credit, mortgage, liability lines etc).
However, directionally it was clear that Covid-19 is developing into a loss to rival in scale anything we have seen before, and one which will exceed everything we have seen before in complexity.
Universally sources were more bearish than they had been, and the loose ranges they offered were higher. Furthermore, the tendency to describe this event as bigger than Katrina ($80bn) was widespread.
"It is obvious it is the worst [insured loss] ever – even in the best-case scenario," one reinsurance source said.
An important corrective to these perspectives is that at a generalised industry level there is an incentive for the bad news to be exaggerated – particularly given that right now the sector has a target on its back – but the assessment does seem to be shared by senior reinsurance broking sources.
And it is further understood that the private messaging at senior management meetings in some cases is the same.
Mounting loss fears reflect increased awareness of the breadth of the wording in SME property and commercial package policies, with more coverage afforded to clients for denial of access resulting from government action than had previously been understood.
These wordings do not explicitly provide pandemic cover in most cases, but underwriters and brokers believe that they may have effectively extended cover to clients.
The industry's focus on its largest market, the US, may have slowed the realisation around these exposures. Countries flagged to this publication as issues include Canada, the UK, Germany, the Nordic countries, Switzerland, Malaysia and South Africa, with some examples of coverage in France.
Sizeable losses are now emerging for some insurers operating in these markets, according to reinsurers.
What emerges from the legislative and judicial efforts in the US to tag insurers with claims remains the greatest wildcard in the ultimate loss quantum, with the expectation still prevalent across the sector that there will be no blanket overturning of exclusionary wording.
The outlines of the coming struggle between cedants and reinsurers are already starting to become clear, as early messaging from the secondary market emphasises their intention to strictly hold cedants to policy wordings.
Reinsurers are understood to be looking carefully at hours clause wordings, with some set to argue that coronavirus represents multiple events (even if there is no yo-yo lockdown), creating either multiple retained losses on cat treaties or scope for frequency at the bottom end.
Sources also flagged pending issues around determining both the date of loss and the date that the loss ends. This is likely to be particularly challenging when there are multiple jurisdictions in the same portfolio that ran lockdowns for different dates, as would be the case with global, pan-European and US cat treaties.
If this approach to the interpretation of hours clauses – which were not designed to refer to pandemics – is ultimately upheld, then the property loss for reinsurers would skew to quota share and property per risk treaties.
Other wording disputes could include the definition of "all natural perils" – language which is common in retro contracts – with a lack of clarity on whether this includes pandemic.
Sources have also stressed that reinsurers are using language that implies that they will not blindly follow the fortunes of their clients. If they do not believe that the underlying wording extends BI cover in a pandemic, then multiple sources told this publication that they would consider any payment to be ex gratia and consequently subject to commercial imperatives.
Already there is an expectation that disputes will be numerous, highly consequential given the sums involved and drawn out over many years.
If insurers with soft property wordings do find themselves with large sideways losses that their reinsurance structures will not effectively respond to, they could face solvency issues.
To date the impact of coronavirus on the ILS market has focused on a presumed rush for liquidity and a short-term inflow of new capital, but fears of a major reinsured property loss will bring the underwriting loss to the market.
With most retro renewing at 1 January, there is plenty of time for most reinsurers to get a better handle on likely losses – including the implications of the T&C disputes – but there will be some early test cases in the way of 1 June renewals.
As such, retro will take a long time to play out. But there are the ingredients here of a nightmare event for ILS, with a potentially huge but also highly uncertain loss quantum that could take many years to fully determine. Such an event could – in the worst-case scenario – test the already strained capital release mechanisms of the sector to destruction.
Insurance Insider delivers global wholesale, specialty, and (re)insurance intelligence that enables you to act first. Redeem your complimentary 14-day trial for more premium content from Insurance Insider.