Fairfax 2020

[Spekulatius]

From CNBB : ''Coronavirus will be the largest loss on record for insurers, Lloyd’s of London says''

Would Fairfax insurance businesses face such a large loss as well?

https://www.cnbc.com/2020/05/14/lloyds-of-london-coronavirus-will-be-largest-loss-on-record-for-insurers.html

Bry

Throwing that one in again to get you guys' input. Could Fairfax insurance businesses face similar losses? As large as Lloyd's?

I think you need to think about covid on a sub-by-sub basis. 

Prem assures us that the US subs (C&F) used the industry standard virus exclusions in their commercial contracts, which should protect them from business interruption claims.  It seems like about half of the US states have enacted legislation to provide immunity to nursing homes for covid claims ( https://time.com/5835228/nursing-homes-legal-immunity-coronavirus/ ) so that too should help limit litigation.  So far, I have not read about any litigation involving C&F, but there are definitely many cases already launched against other US primary insurers (KJP posted this in the P&C thread: https://www.law360.com/pennsylvania/articles/1273308 ).

Given that Northbridge didn't take a covid provision, one presumes that they used some sort of exclusionary language for business interruption.  What is more, nursing home litigation risk in Canada is less pronounced because getting a class certified is a bit more difficult, and you can usually only sue for actual economic damages in Canada -- punitive damages are very, very rare in Canada, and that would be the expensive part of the claim because the economic damage from the premature death of an 88 year-old is pretty minimal because most of them don't have a job or run a business (ie, when an senior citizen dies a year or two sooner than he should have, virtually no income is lost).

Zenith is one of the subs that causes me consternation.  Zenith has stated on its website that a covid related workers comp claim needs to demonstrate that the virus was caught in the workplace rather than in the community ( https://www.thezenith.com/wp-content/uploads/Zenith-Coronavirus-Update.pdf ).  In most cases, that's a pretty hard thing to prove, but I do wonder whether there won't be some employers/employees who argue that the existence of a covid cluster in a workplace is adequate proof of virus provenance.  In particular, nursing homes, public transit agencies and slaughter houses have all had a heavy incidence of covid, resulting in time off and sometimes death of employees.  I question whether some of that might ultimately come back to Zenith if groups of employees convince a judge or jury that the existence of an employment related cluster is adequate proof that covid was actually a workplace injury.  But, I don't recall seeing any covid provision for Zenith.

Like every other reinsurer in the world, Odyssey is a bit of a concern because it's impossible for a shareholder to have any idea of what exactly is being reinsured.  Odyssey took a $50m covid provision, so obviously there are at least a few policies that are a problem.  But, is Odyssey reinsuring a primary company that was sloppy in the wording of its business interruption insurance?  Hard to know, but it's a point of concern.

The other sub that might be a problem is Brit.  Brit does business in the UK and apparently some of the commercial policies written by UK insurers did not have a virus exclusion, so that might be the driver behind Lloyds' large estimate of indemnities.  Brit did take a covid provision in Q1, but who knows the extent of their problem?  Did they write many commercial policies with business interruption?  What reinsurance have they taken out on their commercial book?


At this point, we don't have many options other than to take Prem at his word and assume that the covid claims will not be very large.  We don't have many options other than to accept that the $84m provision in Q1 is a fair estimate of the ultimate covid liability.  But, I certainly don't blame you for being concerned about how this might evolve.


SJ

It is important to note that any provision taken for Q1 represents the status as determined on 3/31 , but most likely earlier than that to determine losses.
I think we will see much larger losses in Q2 and Q3, but they doesn’t just apply to FFH subs, it applies to all Insurers.

On 3/31, the epidemic was just ~4 weeks old and insurance losses will take time to play it.

[StubbleJumper]

...
Zenith is one of the subs that causes me consternation.  Zenith has stated on its website that a covid related workers comp claim needs to demonstrate that the virus was caught in the workplace rather than in the community ( https://www.thezenith.com/wp-content/uploads/Zenith-Coronavirus-Update.pdf ).  In most cases, that's a pretty hard thing to prove, but I do wonder whether there won't be some employers/employees who argue that the existence of a covid cluster in a workplace is adequate proof of virus provenance.  In particular, nursing homes, public transit agencies and slaughter houses have all had a heavy incidence of covid, resulting in time off and sometimes death of employees.  I question whether some of that might ultimately come back to Zenith if groups of employees convince a judge or jury that the existence of an employment related cluster is adequate proof that covid was actually a workplace injury.  But, I don't recall seeing any covid provision for Zenith.
...
At this point, we don't have many options other than to take Prem at his word and assume that the covid claims will not be very large.  We don't have many options other than to accept that the $84m provision in Q1 is a fair estimate of the ultimate covid liability.  But, I certainly don't blame you for being concerned about how this might evolve.
SJ

In terms of COVID-19 by priority of potential impact, event cancellation comes first, then business interruption (dynamics developing) and then workers comp exposure (at Zenith). The evolving "presumption" definition makes sense. Typically, the presumption of a work related "disease" (infectious or not) does not apply and the worker has to "prove" that there is causality. However in certain instances (by using classes of work or criteria for exposure) the presumption threshold can be modified. Work-related exposure to COVID-19 will be specific to state or provincial jurisdiction but similar principles apply. Some areas will use a class of work method (ie healthcare workers, first responders, working in meat processing plants?) and states like California will use a criteria method. The end result (the story has evolved when compared to the Zenith link mentioned above; this was released relatively early in the game) was influenced by inputs given by players like Zenith. There may be significant pockets of costs but the typical accepted claim using the criteria will likely result in very manageable costs. Also, the typical person who will get COVID-19 will be asymptomatic or will have limited symptoms. Zenith has a long history of strong underwriting and this episode may even help them to strengthen their moat when rate increases will be authorized by regulators.
https://www.thezenith.com/wp-content/uploads/Agent-GovernorExecutiveOrder.pdf

Thanks for the thoughtful observations, Cigarbutt.  I guess my disquietude about workers comp reflects my limited knowledge about the subject.  It was an enormous cost, largely underwritten by Uncle Sam, for the 9/11 response.  There is some talk of public funding for covid workers, but this is yet unclear.  I agree that most will have extremely mild symptoms that they might not even notice, some will have a modest WC claim if they are off work for a month or two because that would likely exhaust their paid sick leave.  The ones that get expensive will be the tail cases -- those that have long-term lung issue, extreme PTSD and those that actually perish from covid.

In the US there have already been about 100 public transit workers who have died from covid (https://www.theguardian.com/world/2020/apr/20/us-bus-drivers-lack-life-saving-basic-protections-transit-worker-deaths-coronavirus).  My guess is that there will be hundreds of workers from nursing homes that will meet the same fate.  And then there will be the other less visible deaths from places like the slaughter plants that we have already mentioned, plus the retail workers who worked in essential businesses.  What kind of indemnities will be triggered?  A couple million bucks per death if the policy covers income replacement for 5, 10, or more years?  It doesn't take a lot of imagination to envision 1,000 fatalities in the US, resulting an industry level indemnity of perhaps a couple billion divided by however many underwriters are in that space.  If the government does get involved, that becomes more manageable (just like what it was for the 9/11 responders), but it seems perplexing to me that this has not already been a cause for at least a modest provision at Zenith.

Anyway, maybe this will end up being nothing-burger for Zenith, but I still find that idea quite counterintuitive.


SJ

[Cigarbutt]

...
Zenith is one of the subs that causes me consternation.  Zenith has stated on its website that a covid related workers comp claim needs to demonstrate that the virus was caught in the workplace rather than in the community ( https://www.thezenith.com/wp-content/uploads/Zenith-Coronavirus-Update.pdf ).  In most cases, that's a pretty hard thing to prove, but I do wonder whether there won't be some employers/employees who argue that the existence of a covid cluster in a workplace is adequate proof of virus provenance.  In particular, nursing homes, public transit agencies and slaughter houses have all had a heavy incidence of covid, resulting in time off and sometimes death of employees.  I question whether some of that might ultimately come back to Zenith if groups of employees convince a judge or jury that the existence of an employment related cluster is adequate proof that covid was actually a workplace injury.  But, I don't recall seeing any covid provision for Zenith.
...
At this point, we don't have many options other than to take Prem at his word and assume that the covid claims will not be very large.  We don't have many options other than to accept that the $84m provision in Q1 is a fair estimate of the ultimate covid liability.  But, I certainly don't blame you for being concerned about how this might evolve.
SJ

In terms of COVID-19 by priority of potential impact, event cancellation comes first, then business interruption (dynamics developing) and then workers comp exposure (at Zenith). The evolving "presumption" definition makes sense. Typically, the presumption of a work related "disease" (infectious or not) does not apply and the worker has to "prove" that there is causality. However in certain instances (by using classes of work or criteria for exposure) the presumption threshold can be modified. Work-related exposure to COVID-19 will be specific to state or provincial jurisdiction but similar principles apply. Some areas will use a class of work method (ie healthcare workers, first responders, working in meat processing plants?) and states like California will use a criteria method. The end result (the story has evolved when compared to the Zenith link mentioned above; this was released relatively early in the game) was influenced by inputs given by players like Zenith. There may be significant pockets of costs but the typical accepted claim using the criteria will likely result in very manageable costs. Also, the typical person who will get COVID-19 will be asymptomatic or will have limited symptoms. Zenith has a long history of strong underwriting and this episode may even help them to strengthen their moat when rate increases will be authorized by regulators.
https://www.thezenith.com/wp-content/uploads/Agent-GovernorExecutiveOrder.pdf

Thanks for the thoughtful observations, Cigarbutt.  I guess my disquietude about workers comp reflects my limited knowledge about the subject.  It was an enormous cost, largely underwritten by Uncle Sam, for the 9/11 response.  There is some talk of public funding for covid workers, but this is yet unclear.  I agree that most will have extremely mild symptoms that they might not even notice, some will have a modest WC claim if they are off work for a month or two because that would likely exhaust their paid sick leave.  The ones that get expensive will be the tail cases -- those that have long-term lung issue, extreme PTSD and those that actually perish from covid.
In the US there have already been about 100 public transit workers who have died from covid (https://www.theguardian.com/world/2020/apr/20/us-bus-drivers-lack-life-saving-basic-protections-transit-worker-deaths-coronavirus).  My guess is that there will be hundreds of workers from nursing homes that will meet the same fate.  And then there will be the other less visible deaths from places like the slaughter plants that we have already mentioned, plus the retail workers who worked in essential businesses.  What kind of indemnities will be triggered?  A couple million bucks per death if the policy covers income replacement for 5, 10, or more years?  It doesn't take a lot of imagination to envision 1,000 fatalities in the US, resulting an industry level indemnity of perhaps a couple billion divided by however many underwriters are in that space.  If the government does get involved, that becomes more manageable (just like what it was for the 9/11 responders), but it seems perplexing to me that this has not already been a cause for at least a modest provision at Zenith.
Anyway, maybe this will end up being nothing-burger for Zenith, but I still find that idea quite counterintuitive.
SJ

The post's aim was to moderate the existential worry but it's early and developing and there will be costs to Zenith (currently about #17 in market size). If you look at Markel (currently about #23 in market size), their Q1 report appeared quite conservative overall (versus future losses) for COVID-19 exposure but they reported and commented that the workers comp threat was forming and was not counted in the reported reserve numbers. The following gives an idea of exposure range for the industry:
https://www.insurancejournal.com/news/national/2020/05/05/567599.htm
The factors that mitigate the threat are 1-the degree of "expansiveness" of states for inclusion will be proportional to various state-sponsored stop-loss schemes and 2-use of quota-share and excess of loss reinsurance (of course this is a two-way street).
One thing to watch for is that workers claims tend to rise with unemployment (interesting topic on its own) and that may end up more significant than the disease itself.

[StubbleJumper]

Thanks Cigarbutt.  I hadn't seen the article from Insurance Journal, so that is interesting in particular.  A couple of the more elaborate industry level loss estimates gives a bound for Zenith.  It looks like perhaps 16 loss points, before reinsurance and government funding, might be the reasonable estimate, with a bound of perhaps 50 loss points.  So, for an outfit like Zenith that writes $750m of premium, that would be maybe ~$120m before reinsurance and government funding, but possibly as much as $375m .  As you said, it's probably not an existential question, but it's curious that no provision was taken in the first quarter.


SJ

[bearprowler6]

Allied World's Reinsurance book of business is also worth watching as Covid claims are identified. Noteworthy here is the facultative coverage offered by AWRe related to workers compensation.

Also, as mentioned much earlier in this thread, Fairfax's overall exposure to 3rd party reinsurers as quantified through its Reinsurance Recoverables also deserves special attention as the extent of Covid claims unfolds over the next several quarters.

[Bryggen]

Another Fool article on Fairfax. Simplistic and positive view.

https://ca.finance.yahoo.com/news/canada-warren-buffett-why-now-160053674.html

I may be seeking confirmation bias, but interesting.

[patterson]

Another Fool article on Fairfax. Simplistic and positive view.

https://ca.finance.yahoo.com/news/canada-warren-buffett-why-now-160053674.html

I may be seeking confirmation bias, but interesting.

Bryggen,

I agree with the article's conclusion that it's a good time to buy Fairfax, but keep in mind that the Motley Fool has published various iterations of the same article for years now about "Canada's Warren Buffett" and this is just the latest one.

[Bryggen]

Tracking Prem Watsa's Fairfax Financial Holdings Portfolio - Q1 2020 Update
https://seekingalpha.com/article/4349793

[Cevian]

Where's Exxon? I could have sworn he said Exxon on the call, even mentioned what the dividend was yielding at the time (I think 10%).

[John Hjorth]

Where's Exxon? I could have sworn he said Exxon on the call, even mentioned what the dividend was yielding at the time (I think 10%).

Personally, I hear you, Cevian,

Also, personally, I'm [also] puzzled by this, & need to go back to listen to the conference call.