Fairfax 3.0

[Dazel]

I am going to refrain from discussing the past (after a brief explanation) as much as I possible can on this board because I have done a lot of it over 2 decades! It creates two narratives and I will take the perspective of third or Fairfax 3.0 in this thread. Viking is the king so I will not even pretend to put the numbers work like he does but I will give a narrative and the psychology behind Fairfax 3.0. It might be the most important “thing” for most after all the shares have exploded higher.

 

(Cdn prices)

Fairfax 1.0 (narrative they can do no wrong…Warren Buffett of the North)

$3 to $600-1984-2000

superstardom for Prem Watsa and Fairfax.

 

Fairfax 2.0 

$600 to $60 2000-2003 (insurance companies are garbage Fairfax 1.0 a fraud?)

 

$60 to $400 2003-2008 (not a fraud but insurance companies are still in trouble..Big Short and long bond bets of the decade

$400 to $750 2008-2016 (insurance improving Fairfax is a bet against the market even though their shorts were decimated)

 

$750  to $350 2016-2020 (pulled short bets off the table and were left for dead in Covid, company under earning they did not reach for yield      market hated it and killed the stock)

 

Fairfax 3.0

 

$350- $2000  2021-2024 (Re anointed king of the world, cash put to work in an all in bet as well Prem going all in personally, insurance operating profits sky rocket, stock follows)

 

$2000-?          2025  -The magic of compounding at work! How did they do that?
 

 

Fairfax 1.0

Everything went right!

-hard insurance market at the beginning…they doubled their premiums in their first year

-1990 recession that hit the stock they took advantage and bought back 20% of the company from Markel. (Greatest trade of alltime?)

-the stock joined the bubble market briefly hitting $600 setting up a tough Fairfax 2.0
-High stock price fueled large takeovers that had hidden insurance tail risk that would set up hell for Fairfax 2.0 

 

Fairfax 2.0 there was great frustration with one step forward and two steps back and there are reasons for it that I will not go into detail on an explanation because they “don’t” matter now. 
 

-insurance losses of companies taken over almost sunk the company

-stocks were expensive  

-2 crashes nailed one (greatest trade of alltime 2007-2008) and got smoked in the second one (2020)

-Blackberry disaster

-short selling disaster

-zero and negative interest rates (This was the biggest obstacle for growing book value and compounding money)

-underneath they were building the foundation for Fairfax and smartly doing it with growing their “rocket fuel” insurance companies through smart acquisition and building. This will be Prem’s greatest achievement the man can build insurance companies like no one else his talent here is unmatched. 

Fairfax 3.0

 

It’s built! Time to print money and buy back every last share possible share before the world figures out the intrinsic value per share is above $3000. (USD) Why is intrinsic value this high compared to the market price? Prem and his team built a pure play insurance empire from basically scratch so the book value of these insurance companies is recorded at the cost of building them “not acquiring them”! If these companies were sold individually they would demand many multiples of their current balance sheet value.

 

Proof examples of the intrinsic value above the recorded book value at Fairfax

 

They sold Lombard that was built from scratch for almost 3X book value, $1.7b premiums=$4.6b value.

 

https://www.fairfax.ca/press-releases/fairfax-sells-shares-of-icici-lombard-2017-09-27/#:~:text=Fairfax Financial Holdings Limited (“Fairfax,of approximately US%24548 million.

 

First Capital was sold for 3X book value

 

https://www.fairfax.ca/press-releases/fairfax-and-mitsui-sumitomo-insurance-enter-into-strategic-alliance-and-sale-of-first-capital-2017-08-23/

Global pet insurance sale gain $1.2B! 6X book value…They made so much above book value they did not even announce the gain they were embarrassed!

 

https://www.fairfax.ca/press-releases/fairfax-announces-successful-completion-of-sale-of-global-pet-insurance-operations-2022-10-31/

 

Intrinsic value is likely at least 3X book value and growing=$3,000 usd per share with the “rocket fuel” of the float that adds considerably to intrinsic value. To be clear Fairfax will “Not” trade at 3X book this week or month  but from the starting point of today it will eventually get there because the earnings power and intrinsic value of the companies will take it there. Companies are not sold at 3x because of their value that day they are thought to grow into and above the purchase price paid by the acquirers.

 

 

“Rocket Fuel/Float” and operating earnings rising to the mean

 

The knock on Fairfax has always been that their insurance operating earnings lagged their peers and they relied on investments to heavily. Not anymore! Their operating earnings are now approaching $5b per year and analysts are waking up to running the numbers with disbelief. They are so large that it is tough to grasp the possibility of how high they are headed. Berkshire had this “aha” moment many years ago before the stock went vertical and Fairfax 1.0 experienced it in the late 1990’s. 
 

Berkshire is the yardstick in the investing world but very few (especially these days) understand that Buffett and Munger used math to goose  returns. Buffett came up with the term “rocket fuel for insurance float” and Prem and his team started Fairfax with the idea they could copy his model. This is not new to any of us old timers on this board but to the newcomers “float” and the cost of it is the math behind Fairfax 3.0 and the secret to Berkshire’s success. I am going to compare Fairfax 3.0 to Berkshire in a 2003 snapshot to show the valuation of each comparatively without bias at their respective 38 year mark to show how compounding works in the two companies and what is possible for Fairfax. 
 

These numbers will “Not” be perfect and there will be errors as I do not have the time nor the interest in being precise! This will be more art than science. What’s possible in the future is up to Prem and his team. Mr. Buffett is unequivocally the best longterm investor of alltime the Fairfax team will not be able come close to his numbers but I hope to prove they don’t have too. I strongly suggest you read Buffett’s Berkshire 2003 chairman letter and all of them for that matter especially for the times we are in.

 

https://www.berkshirehathaway.com/letters/2003ltr.pdf

 

Edited Sunday at 04:04 PM by Dazel
Time constraint

[LC]

1 hour ago, Dazel said:

$2000-?          2025  -to be continued all the time I have for now…

 

 

Oh c'mon! Don't leave us hanging, buddy!

[Dazel]

Below is What the insurance numbers can look like compounded vs other less than average companies.

This is from page 4 of Berkshire’s 2003 annual report. Mr. Buffett explains the difference between book value and earnings between Berkshire textiles and the change in earnings beginning with insurance company National Indemnity purchase. Fairfax is all National Indemnity and insurance this is in the business that Buffett covets. 

This annual report is from the 38th year of Buffett ownership and Fairfax is in their 38th year. You will notice that Berkshire’s insurance operating earnings were almost the same as Fairfax at $5.42b!

 

Edited Sunday at 04:00 PM by Dazel

[Dazel]

 

[Hoodlum]

46 minutes ago, Dazel said:

 

It is interesting how it was commented that the P&C industry regularly operates at a substantial underwriting loss.   We have seen this change with Fairfax but also with much of the industry over the past decade.  Is this something that is sustainable or will we eventually go back to the historical trend. 
 

The expectation of Substantial Underwriting Losses is what really stood out to me. 
 

 

Edited Sunday at 04:32 PM by Hoodlum

[LC]

3 hours ago, Hoodlum said:

It is interesting how it was commented that the P&C industry regularly operates at a substantial underwriting loss.   We have seen this change with Fairfax but also with much of the industry over the past decade.  Is this something that is sustainable or will we eventually go back to the historical trend. 

 

Most economic theory and business experience would suggest not to expect long term underwriting profitability for the industry as a whole. 

 

Large scale insurance is a commodity business, pricing is transparent, and there are many sophisticated participants.

 

Maybe a continued string of catastrophes (Los Angeles for example) will help sustain pricing in the near term. And maybe some thesis is that climate change will create an upward shift in catastrophe frequency and/or severity. Not sure that is really knowable, however. 

 

The other long thesis is regulatory capture, which could limit entrants and protect pricing. I don't know much about that as it's state-level, I take folks like Buffett at their word when they talk about insurance industry dynamics.

Edited Sunday at 08:22 PM by LC

[Viking]

5 hours ago, LC said:

 

Most economic theory and business experience would suggest not to expect long term underwriting profitability for the industry as a whole. 

 

Large scale insurance is a commodity business, pricing is transparent, and there are many sophisticated participants.

 

Maybe a continued string of catastrophes (Los Angeles for example) will help sustain pricing in the near term. And maybe some thesis is that climate change will create an upward shift in catastrophe frequency and/or severity. Not sure that is really knowable, however. 

 

The other long thesis is regulatory capture, which could limit entrants and protect pricing. I don't know much about that as it's state-level, I take folks like Buffett at their word when they talk about insurance industry dynamics.

 

@LC , you bring up something that is both interesting and often stated/discussed. "Most economic theory and business experience would suggest not to expect long term underwriting profitability for the industry as a whole." 

 

This might be a great example of how theory often diverges from what actually happens in the real world. 

 

I think the real question is what is actionable for an investor from this insight? 

 

There are two general ways to go in our thought exercise:

1.) Over time, the combined ratio of all P/C insurance companies will revert to 100.

2.) Over time, the CR of all P/C insurance companies will average 100, but with the following split:

- the CR of low quality P/C insurance companies will average a number above 100

- the CR of high quality P/C insurance companies will average a number under 100. 

 

In the second example, low quality insurance companies will likely see higher top-line growth in soft markets (net premiums written). And high quality insurance companies will see lower top-line growth in soft markets. And vice versa.

 

My point is I think there is a broad assumption that 'revert to 100' means all P/C insurance companies will see their CR go to 100 (and likely higher). I agree that we could see CR tick higher for high quality P/C insurance companies I am not convinced it has to go to 100 (or higher).

 

The other important variable are reserving practices - and reserve releases. High quality P/C insurance companies have likely been building reserves in the 5-year hard market. This could buffer their reported CR in a soft market.

 

And, of course, we will get a 'historically' bad year for catastrophes at some point in the coming years (we just don't know what year). So there will be volatility from year to year (sometimes a great deal). I am talking about a smoothed CR (perhaps over a 3 to 5 year period).

Edited yesterday at 01:35 AM by Viking

[Dazel]

Hoodlum,

 

Buffett’s particular problem with almost assured losses was “Asbestos” the entire industry got smoked on this and it was the biggest problem for Fairfax at the start of Fairfax 2.0. CAT insurance is different as pricing matters most. Ie cyclical…more storms/fires/weather the higher the price and vise versa.

 

[Dazel]

[Dazel]

The Math

 

Berkshire

Buffett has set up Berkshire so that all else being equal whatever he returns on His investment portfolio roughly doubles his return on equity. I have not seen the number but I would bet he averages around 11% (Same as the S&P) but with the “rocket fuel” that take his long term average compounding ROE to 22%. When Berkshire was smaller his returns were much larger but scale has taken his long term average to the market multiple…or a little above it. 

****Dont start yelling at me! Berkshire never has their entire investment portfolio in equities that’s why Buffett’s return “on his entire investment portfolio” is the same as the S&P. Ie he has $325b in cash right now. 

Fairfax 3.0

The long term return is around 8.7% and their insurance leverage is the same so this would take the return to 17.5%. These numbers turn out to be very accurate over 38 years. Fairfax also trounced the 8.7% in the early days hence the stock sky rocketing in Fairfax 1.0

 

All of the incredible amount of work to run these companies to bring them to a “all things being equal” is there to set up the simple math above. Fairfax 2.0 struggled to keep this equilibrium because interest rates went to zero and below in Europe and a general distaste for the bull market we are still in. Fairfax 3.0 could theoretically buy the 30 year bond to maintain a 5% return and secure a 10% ROE with the insurance companies performing well. To return their long term goal of 15%ROE they need to do 7.5% return on the investment portfolio. 

[SharperDingaan]

17 hours ago, Viking said:

 

@LC , you bring up something that is both interesting and often stated/discussed. "Most economic theory and business experience would suggest not to expect long term underwriting profitability for the industry as a whole." 

 

Economic theory just offers a model; it doesn't recognise time value, and it doesn't have the provable certainty of empirical science. If it takes two years for the global industry to do a weighted average CR of 100, one then needs to discount that at some discount rate. If your CAGR was just 5%, you will be 'off' by at least 10% (1.05^2), and the global industry would look it is incurring a structural 10%+ loss every year.   

 

NA may be up, Asia flat, and Europe down .... but overall ? it's a 10% loss simply because of the need to recognise time value.

 

SD