Fairfax 2024

[Haryana]

2 hours ago, dartmonkey said:

Float cuts both ways.  Catastrophe losses will be magnified within FFH simply because of the asset to equity leverage.  So good times...tons of income.  Bad times...significant losses.  There is no net tangible value of float other than it is a more useful version of debt.  Leverage is leverage. 

 

I would say the leverage is investment leverage, not insurance leverage. 

 

Any pure insurer is at risk of having claims exceed premiums. But if you take Fairfax with its $30b of float , more than its market cap, and you compare it with Berkshire whose float is less than a quarter of its market cap, the leverage risk is on the investment side - at Fairfax, a negative return will be hugely negative because of the leverage, and a good return will be greatly amplified. 

 

But I don’t see a downside of this, if it’s mostly invested in treasuries, as it is at Fairfax. You can be pretty sure you’re not going to lose a lot of money on treasuries. I’d say that Fairfax might be safer with a lot more leverage on a bond portfolio than Berkshire with 20% of its assets in Apple. 

 

I see it as a simple thesis that Fairfax is better than cash because that cash is in Treasuries with better than free leverage.

 

[UK]

10 hours ago, Viking said:

2.) What is the size of IFRS 17 impact?

 

TBD - but not a concern

 

"In the fourth quarter of ’23, the net earnings of $1.3 billion included pre-tax net expense of $781 million, and the net earnings in the full year of 2023 of $4.4 billion included a pre-tax net benefit of $210 million related to IFRS-17. The pre-tax amounts are reported within two financial statement lines in the consolidated statement of earnings."

 

It seems that this item has reversed back quite a bit in q4?

[nwoodman]

Discussion on BNN regarding results and the CC. I'm not sure who the grey-haired gentleman is and he is a bit waffly for my money.  However, the upshot is a message of

 

1. They've been here before and survived. 

2. Stellar results.  

 

https://www.bnnbloomberg.ca/video/fairfax-operating-income-jumps~2868708

 

and another one

 

 

I think this coverage has been gold.

Edited February 17, 2024 by nwoodman

[UK]

9 hours ago, nwoodman said:

1. Forward Guidance

Prem Watsa

“Now as I’ve said for the last number of quarters, the most important point I can make for you is to repeat what I’ve said in the past - for the second time in our 38-year history, I can say to you, we expect - there is of course no guarantees - sustainable operating income of $4 billion, operating income consisting of $2 billion-plus from interest and dividend income, $1.2 billion from underwriting profit with normalized catastrophe losses, and $750 million from associates and non-insurance companies. This works out to over $125 per share after interest expenses, overhead and taxes. Of course, fluctuations in stock and bond prices will be on top of that, and these fluctuations only really matter over the long term.’

 

Plus: "When you put all of that together, we look at that operating income of $4 billion as a pretty conservative number."

[Hamburg Investor]

8 hours ago, Parsad said:

 

Float cuts both ways.  Catastrophe losses will be magnified within FFH simply because of the asset to equity leverage.  So good times...tons of income.  Bad times...significant losses.  There is no net tangible value of float other than it is a more useful version of debt.  Leverage is leverage.  In an outlier event, BRK will be the last to fall!  Cheers!

I'd say "a more riskless version of debt". And leverage and leverage can be very different in terms of risk.

Berkshire Hathaway was also heavily float hedged for a long time. I sometimes get the impression that many people forget that. You have to make a decision: Either Warren actually took on too much risk for a long time and got pretty lucky. Or Warren's way of using float was good in terms of risk adjusted returns. You can't say both. As investors, we need to understand how to minimize risk. I think float is a key technology to achieve high risk-adjusted returns. Of course, you can't go too far with float. I've heard a lot of value investors say that a big mistake was not buying Berkshire Hathaway in the 1970s or 1980s because the price was too high for them.

Risk is higher with more float, as it is with every "more" of any leverage. We don't disagree on that one. We don't disagree on BRK being a fantastic company and being the last to fall (that's why it's my second biggest holding with 14%). Where we disagree: I think 130% of float like at FFH is ok and the chances are higher than the risks (so higher price is ok); and you say the opposite - the float risk is so high, that even higher "normal" (so the roes in the times where the business is not wiped out nearly totally) returns on equity do not justify a higher price, but the other way around. This is not black or white and everybody has to define his risk tolerance. That's fine. I don't think there are a lot of ways for getting high roes with less risk than by using (not too much) float. There's no free lunch. Without taking risk, no return. Risk comes from not knowing what you're doing. Don't loose money. Don't forget that rule. Of course. Simple rules, but still not easy to do.

There are a lot of other risks to consider when comparing Berkshire versus Fairfax. The country risk. I would consider this to be greater at Berkshire than at FFH with its rapidly growing international insurance business and its large positions in India and Greece. The management risk. I would also rate this higher at Berkshire in the long term, as management at Berkshire is likely to change much sooner than at Fairfax; and Prem has a longer track record in his position than Warren's successor (who has experience, but as CEO of Berkshire he starts with a track record of 0 years).

 

BTW: I have a small portion in Protector Forsikring, a Scandinavian investment company that is growing very fast and of course it does so with a high float compared to the assets, the management is great. Sometimes I wonder if I shouldn't have more than just 1.5% in Protector; but I don't as their track record is not long enough for me (20 years. Writing it that way, it does seem quite long now).

[MMM20]

13 hours ago, Parsad said:

 

The leverage of float is already accounted for in the portfolio income.  I don't give it any additional weight than that, since float cuts both ways. 

 

If you are valuing FFH on earnings, then float is accounted in the income/loss statement.  If you are valuing FFH on book value, float is also accounted for since it will have both a positive and negative effect on book value depending on catastrophe losses. 

 

Float is just a more useful version of debt.  There is no net tangible increase or decrease in value from float. 

 

Cheers!

I just have a hard time understanding the view that this should trade anywhere close to book value when the $30B+ of float does not show up as an asset on the balance sheet. Adding this structural ability to borrow tens of billions of dollars at 0 or better into to our estimate of intrinsic value, I can’t see how one can credibly argue that the output is anything less than 1.5x book and probably much higher. Buffett taught us this decades ago so I’m not onto something clever. What is the counter argument? That leverage - even of the highest possible quality and at their scale a nearly irreplicable (smart folks have certainly tried) economic asset not liability - introduces so much volatility that it flips to fragility and therefore merits a much lower valuation? Lower than anything else out there?

 

Edited February 17, 2024 by MMM20

[MMM20]

13 hours ago, dartmonkey said:

Float cuts both ways.  Catastrophe losses will be magnified within FFH simply because of the asset to equity leverage.  So good times...tons of income.  Bad times...significant losses.  There is no net tangible value of float other than it is a more useful version of debt.  Leverage is leverage. 

As long as you realize Mr. Buffett disagrees with you - he has written about it extensively for decades and it explains nearly all of Berkshire’s excess returns in the middle decades if you really dig into it. He has even said that he wouldn’t trade $1 of float for $1 of equity. What does that mean for intrinsic value? This is the key question for FFH investors nowadays IMHO. 
 

Edited February 17, 2024 by MMM20

[ValueMaven]

I feel like I was very lucky.  I've owned FFH in the past - but largely missed this recent runup.  I was able to add aggressively the day of, and the day after the MW short thesis, and was able to get FRFHF around $910 per share (below BV).  Honestly, this quarter was awesome, particuarly around the interest and dividend income and improved underwriting!  Look at FFH's valuation difference compared to even Arch!!  Please consider me a long-term shareholder - and this board has been very helpful in keeping even a casual observer updated.  

 

Blocks question was so weak ... damn

[ValueMaven]

dumb question - whats the difference between FFH.TO vs. FRFHF - which is OTC.  Honestly, I dont even really know how OTC works.  

[gfp]

9 minutes ago, ValueMaven said:

dumb question - whats the difference between FFH.TO vs. FRFHF - which is OTC.  Honestly, I dont even really know how OTC works.  

 

There is not a ton of difference really.  FFH.TO trades in Canadian dollars and is marginable.  FRFHF trades in US dollars and is usually not marginable.

 

Sometimes in a US-domiciled IRA it is easier to buy FRFHF since you can't borrow Canadian dollars and it adds an extra step to first purchase the exact amount of Canadian currency you need to make the trade and then buy the FFH.TO.  In taxable accounts I just buy the Toronto listed shares.

[Spooky]

37 minutes ago, MMM20 said:

As long as you realize Mr. Buffett disagrees with you - he has written about it extensively for decades and it explains nearly all of Berkshire’s excess returns in the middle decades if you really dig into it

 

This is essentially correct - there is a good paper out there called Buffett's alpha from 2013, I've pasted the abstract below:

 

Berkshire Hathaway has realized a Sharpe ratio of 0.76, higher than any other stock or mutual fund with a history of more than 30 years, and Berkshire has a significant alpha to traditional risk factors. However, we find that the alpha becomes insignificant when controlling for exposures to Betting-Against-Beta and Quality-Minus-Junk factors. Further, we estimate that Buffett’s leverage is about 1.6-to-1 on average. Buffett’s returns appear to be neither luck nor magic, but, rather, reward for the use of leverage combined with a focus on cheap, safe, quality stocks. Decomposing Berkshires’ portfolio into ownership in publicly traded stocks versus wholly-owned private companies, we find that the former performs the best, suggesting that Buffett’s returns are more due to stock selection than to his effect on management. These results have broad implications for market efficiency and the implementability of academic factors.

 

[StubbleJumper]

1 hour ago, MMM20 said:

I just have a hard time understanding the view that this should trade anywhere close to book value when the $30B+ of float does not show up as an asset on the balance sheet. Adding this structural ability to borrow tens of billions of dollars at 0 or better into to our estimate of intrinsic value, I can’t see how one can credibly argue that the output is anything less than 1.5x book and probably much higher. Buffett taught us this decades ago so I’m not onto something clever. What is the counter argument? That leverage - even of the highest possible quality and at their scale a nearly irreplicable (smart folks have certainly tried) economic asset not liability - introduces so much volatility that it flips to fragility and therefore merits a much lower valuation? Lower than anything else out there?

 

 

 

First off, float is definitely valuable, there's no question about that.  That's what tends to attract value investors to the insurance industry. To correct a couple of points in your post, float does show up on the balance sheet as investments (ie, we take policy holders' premiums and invest them and that's where the float goes) and as policy liabilities.  What is more, float is not always at zero-cost or better, as there have been many years when the CR exceeded 100.  But life is definitely grand when that zero-cost or better situation prevails!

 

How much "credit" we should give to it in a company's market valuation is yet another question.  Buffett proposed that float is effectively money that you borrow from somebody else, but as long as your insurance company is a going-concern and never shrinks its book, you never need to repay that money that you borrowed.  He then went one step further and stated that if you borrow money, never need to pay it back, and can do whatever you want with it, then that's akin to equity.  Personally, I have always had a quibble with this because a regulated insurance company can't just do whatever it wants with float.  There's a reason why FFH's portfolio allocation has traditionally been two-thirds to three-quarters fixed income and that's because you always need to be able to pay indemnities.  The downside of that is that we know that long-term returns on fixed income have traditionally lagged equity returns by a considerable margin.  To me, I don't buy Buffett's argument that we ought to value a dollar of float at the same value as a dollar of unrestricted equity because I'm effectively forced to put 66 cents or 75 cents of that dollar of float into an inferior investment vehicle.

 

Over the past 3 or 4 years, underwriting conditions for FFH have been fabulous.  The cost of float has been negative (ie, CR<100) and, more recently, sovereign debt rates have soared, resulting in a financing differential that has been out of this world.  This makes float look like an astoundingly good deal!  But, it hasn't always been the case.  Every year in his annual letter, Prem publishes an excerpt of a table that depicts FFH's cost of float, prevailing government bond rates, and the resulting financing differential.  The average financing differential over most lengthy time periods has been about 4.5%.  That's good, but nothing even close to what we've seen during 2023, which by my eye-ball estimate was a whopping 10.3%!  I believe that was FFH's most favourable financing differential since 1991.

 

If you want to value FFH, think about the company's long-term return on equity, the ability to reinvest at that long-term ROE, and select a P/BV that is appropriate.  If you assess the long-term ROE high enough and you believe that FFH can reasonably reinvest at that ROE, you can easily make the argument that the company is worth 1.5x BV.  But, when doing that exercise, it is important to remain mindful of the financing differential table and the growth in BV table that Prem publishes every year in his annual letter.

 

 

SJ

[Ghost]

21 hours ago, SharperDingaan said:

 

Keep in mind that no buying doesn't mean no money waiting on the side-lines to come in. If MW round 2 doesn't happen soon it's not going to, so give them some time  ; it's a small community, and the laughter is echoing. Their last success was a long time ago, and if this is the best that they can do now ...... well, the walls are closing. 

 

Somebody lend them a few shares, to buy back cheaper!

 

SD!

 

 

 

 

This situation that FFH is currently in (attack by short sellers) reminds me of the Gamestop saga, except of course FFH today is actually undervalued. 

 

As much as I love this board, perhaps we should all start posting on Reddit to help our friends at FFH and really stick it to MW.

 

 

[Hektor]

I guess Berkshire mostly writes short tail insurance, while Fairfax tends to write more long tail insurance. Is this assumption true? If true, how, if at all, does this impact the quality of the float at either companies?

[gfp]

1 hour ago, Hektor said:

I guess Berkshire mostly writes short tail insurance, while Fairfax tends to write more long tail insurance. Is this assumption true? If true, how, if at all, does this impact the quality of the float at either companies?

 

Berkshire writes some very large, very long tail policies.  They are a specialist in messy long tail stuff like the Lloyd's deal and the more recent AIG deal.  They also write some shorter stuff like 1 season Cat and a lot of what GEICO writes - but I wouldn't say that Berkshire's business is shorter duration than Fairfax's on average.

 

An example:

https://www.reuters.com/article/idUSKBN1541TN/

 

$10 Billion premium paid to Berkshire up front (float).  First claims payments by Berkshire didn't start to go out for many years later.

 

A description from the WSJ at the time:

"The agreement with Berkshire’s National Indemnity Co. requires AIG to pay the first $25 billion of claims as they come due. It is expected to be at least several years before Berkshire would begin tapping the roughly $10 billion for its portion of responsibility. The Berkshire unit will pay 80% of net losses and related loss-adjustment expenses if more than the $25 billion is needed for policyholders. Berkshire’s exposure is capped at $20 billion."

Edited February 17, 2024 by gfp

[Matthew Lembo]

1 hour ago, StubbleJumper said:

 

 

First off, float is definitely valuable, there's no question about that.  That's what tends to attract value investors to the insurance industry. To correct a couple of points in your post, float does show up on the balance sheet as investments (ie, we take policy holders' premiums and invest them and that's where the float goes) and as policy liabilities.  What is more, float is not always at zero-cost or better, as there have been many years when the CR exceeded 100.  But life is definitely grand when that zero-cost or better situation prevails!

 

How much "credit" we should give to it in a company's market valuation is yet another question.  Buffett proposed that float is effectively money that you borrow from somebody else, but as long as your insurance company is a going-concern and never shrinks its book, you never need to repay that money that you borrowed.  He then went one step further and stated that if you borrow money, never need to pay it back, and can do whatever you want with it, then that's akin to equity.  Personally, I have always had a quibble with this because a regulated insurance company can't just do whatever it wants with float.  There's a reason why FFH's portfolio allocation has traditionally been two-thirds to three-quarters fixed income and that's because you always need to be able to pay indemnities.  The downside of that is that we know that long-term returns on fixed income have traditionally lagged equity returns by a considerable margin.  To me, I don't buy Buffett's argument that we ought to value a dollar of float at the same value as a dollar of unrestricted equity because I'm effectively forced to put 66 cents or 75 cents of that dollar of float into an inferior investment vehicle.

 

Over the past 3 or 4 years, underwriting conditions for FFH have been fabulous.  The cost of float has been negative (ie, CR<100) and, more recently, sovereign debt rates have soared, resulting in a financing differential that has been out of this world.  This makes float look like an astoundingly good deal!  But, it hasn't always been the case.  Every year in his annual letter, Prem publishes an excerpt of a table that depicts FFH's cost of float, prevailing government bond rates, and the resulting financing differential.  The average financing differential over most lengthy time periods has been about 4.5%.  That's good, but nothing even close to what we've seen during 2023, which by my eye-ball estimate was a whopping 10.3%!  I believe that was FFH's most favourable financing differential since 1991.

 

If you want to value FFH, think about the company's long-term return on equity, the ability to reinvest at that long-term ROE, and select a P/BV that is appropriate.  If you assess the long-term ROE high enough and you believe that FFH can reasonably reinvest at that ROE, you can easily make the argument that the company is worth 1.5x BV.  But, when doing that exercise, it is important to remain mindful of the financing differential table and the growth in BV table that Prem publishes every year in his annual letter.

 

 

SJ

I think argument that Buffett is making is that if one were to borrow money (float) with a zero cost or negative cost with an infinite or very long term ie 50-75 years than the npv of that liability today is can get pretty close to zero, depending on discount rate used.. Yes we are restricted in how we can invest that liability but that liability should not be valued at par.
 

Agree tho that it all comes down to expected Roe of the business over time irrespective of the inputs to get there. 

[Viking]

On the valuation front, i continue to think that Fairfax today is still crazy cheap. 1.08 x BV and a ‘normalized’ trailing PE of around 6.4

 

Over the past 3 years what was the big miss from investors? Investors grossly underestimated the rapidly improving earnings power of the company. 
 

What is the big miss from investors today? It looks to me like investors are now grossly underestimating the power of compounding over time.
 

Compounding is probably the most important concept in investing. The fact it is getting zero attention when people discuss Fairfax tells me volumes about investors mindsets. You also see it in analysts forward estimates (pretty much all have earnings falling at Fairfax in the coming years). 
 

And i love it.

—————

Most investors missed investing in Berkshire Hathaway 30 or 40 years ago. They knew what Berkshire Hathaway was earning. And they knew Warren Buffett was an above average capital allocator. 
 

What was probably the biggest single reason they didn’t invest (or sold their shares too soon)? They didn’t understand the power of compounding. 

Edited February 17, 2024 by Viking

[Cigarbutt]

30 minutes ago, gfp said:

... but I wouldn't say that Berkshire's business is shorter duration than Fairfax's on average.

An example...

On November 1st of 2023, there was a discussion in these pages (FFH Q3 results) with some others and 'we' (in a collegial way?) had defined the Spekulatius quick measure (SQM) of reserve duration for insurers (basically using 'insurance contract liabilities' net of reinsurance divided by net premiums earned) and it looks like this basic measure (ratio of 2.1 for FFH) needed to be multiplied by 1.8 to arrive at 3.8 years (let's call this the adjusted Spek's measure (aSQM)) if i listened well during the last call ("due to the reserves have a longer duration of 3.8 years compared to Fairfax’s very short duration on the fixed income portfolio of 1.6 years", courtesy of J. Allen).

For BRK, using similar numbers (careful IFRS vs US GAAP), the SQM comes to 2.2, suggesting an overall reserve duration of 3.9-4.0 years. In the case of BRK, they have some kind of bimodal distribution of both very short and very very long tails.

At any rate, concerning the quality of float, longer tail lines are both a blessing and a curse as the time value of money can magnify both the favorable and unfavorable reserve development.

For FFH, if the past is indicative of the future:

[Cigarbutt]

2 hours ago, StubbleJumper said:

If you assess the long-term ROE high enough and you believe that FFH can reasonably reinvest at that ROE, you can easily make the argument that the company is worth 1.5x BV.  But, when doing that exercise, it is important to remain mindful of the financing differential table and the growth in BV table that Prem publishes every year in his annual letter.

SJ

Then, why such a big discount to the 1.5x BV numbers these days?

The graph needs some recent update..

The reason i'm asking is because, evidently, the float liability needs to be discounted and, presently, given the potential for profitable growth in float, maybe the required discount on the float liabilities suggest a higher P/B than 1.0.

i'm asking also because an example of the junk presentation by recently attacking shorts concerns the Riverstone-Barbados entity, a transaction involving two sophisticated parties exchanging an entity in run-off (where the insurance liabilities will clearly need to be paid out and maybe more than written in the books) for a P/B of about 1.0.

Isn't FFH now worth more than Riverstone-Barbados then, from a P/B value point of view?

[StubbleJumper]

34 minutes ago, Cigarbutt said:

Then, why such a big discount to the 1.5x BV numbers these days?

The graph needs some recent update..

 

To be worth 1.5x BV, effectively, you need to argue that the long-term ROE on a going-forward basis will be high enough.  We know that the first dozen years of FFH's existence were characterised by an absolutely ridiculous run of consistently high growth in BV (see the table on page 20 of Prem's annual letter from last year).  But following that start-up phase that began in 1986 and ended a dozen or so years later, there has been a period of about 20 years with much less convincing growth in BV.   And then there's been the past 3 or 4 years. 

 

It's really interesting to take that table that Prem publishes every year and monkey around with it a bit.  Erase 1986 and 1987 on the argument that you can't "start up" twice and then recalculate compound annual growth.  Break the series into the growth phase from 1986-99 and the lacklustre phase from 2000-2019 or 2020, and calculate the compound growth rates for each of those phases.  When you do this exercise, you might view the 17.7% historical average growth in BV on that table in a somewhat different light and you should certainly question whether that series can be replicated going forward.

 

That then leaves the really difficult task of guessing what all of that means for the future.  My take is that the economics of the insurance industry will not allow FFH to routinely achieve its bogey of 15% growth in BV on a going forward basis.  I'd be happy if they could grow BV by 12%, but suspect that it may be more like 10% over the long term.  That's not a knock on FFH, it's just a tough industry.  If they actually do routinely achieve that 15% goal, I'll be the happiest guy around.

 

But, turning to valuation, if you pay 1.5x BV for something that grows its book by 15%, you are basically getting a 10% earnings yield with relatively rapid growth.  But, if you paid 1.5x BV and FFH grows BV by 10%, you are getting a 6.7% earnings yield and decent growth...and I'd say you'd have slightly overpaid, but not terribly so if the earnings are rock-solid, regular and predictable (in other words all of the things that the insurance industry is not!).  At this point it's trading at a shade over 1x BV, which means that the market finally figures that FFH is worth more alive than dead (ie, if it trades at <1x BV, in theory it's worth more dead than alive, so just break it up, sell off the pieces or run them off, and give the proceeds to the shareholders!).  I'd say that it's definitely worth 1.2x but I guess we'll see over the coming months whether the market agrees with that.  Maybe it'll rattle off 8 or 10 years of fabulous returns that will make me think it's worth 1.5x?  Time will tell....

 

 

SJ

[TwoCitiesCapital]

15 hours ago, Haryana said:

 

I see it as a simple thesis that Fairfax is better than cash because that cash is in Treasuries with better than free leverage.

 

 

Just be careful applying this in practice. 

 

Fairfax owned a ton of cash/short duration bonds in 2020. The stock still was down over 50+% from the peak despite that. 

 

This was also demonstrated in 2008 where the stock of Fairfax was falling even while it's CDS were printing money. 

 

In a panic, it'll trade like a stock regardless of how well it's underlying assets are doing.

[Cigarbutt]

29 minutes ago, StubbleJumper said:

 

To be worth 1.5x BV ... Time will tell....

SJ

Your post is helpful. But there's room between a p/b of 1.0 and 1.5 or more.

The likelihood of seeing intrinsic value (can use book value as proxy) is somewhat elevated over the next few years but it's less clear if this will be recognized in the marketplace with a higher p/b?

There are reversible factors for this but there are also irreversible factors, such as their unusual ability to perform atypical transactions such as the monetization of the runoff sub mentioned before. This approach (is great in a way but) leaves them vulnerable to less enthusiasm from the market, to reasonable criticism and, unfortunately, to junk/mud attacks.

The sale of Riverstone-Barbados was a great way to monetize an asset in times of need but included aspects which had costs from a risk management and quality of earnings point of view. The CVR tied to the deal effectively was equivalent to an adverse reserve development cover (liability side of the transaction) sold by FFH as part of the deal and the guarantee on the asset side was effectively an off-balance sheet arrangement reported through a derivative-type of disclosure which corresponded to, effectively, FFH buying a total return swap on the underlying common stocks held in exchange for some kind of cash flow streams, while FFH continued to "guarantee" the value of the underlying securities to the buyer. Now this aspect of the transaction has become much smaller and manageable but it was quite large at the time of the sale..

Anyways, Fairfax is Fairfax and will continue to be, i guess, to a large degree and potentially in a lumpy way..

[Viking]

1 hour ago, StubbleJumper said:

 

To be worth 1.5x BV, effectively, you need to argue that the long-term ROE on a going-forward basis will be high enough.  We know that the first dozen years of FFH's existence were characterised by an absolutely ridiculous run of consistently high growth in BV (see the table on page 20 of Prem's annual letter from last year).  But following that start-up phase that began in 1986 and ended a dozen or so years later, there has been a period of about 20 years with much less convincing growth in BV.   And then there's been the past 3 or 4 years. 

 

It's really interesting to take that table that Prem publishes every year and monkey around with it a bit.  Erase 1986 and 1987 on the argument that you can't "start up" twice and then recalculate compound annual growth.  Break the series into the growth phase from 1986-99 and the lacklustre phase from 2000-2019 or 2020, and calculate the compound growth rates for each of those phases.  When you do this exercise, you might view the 17.7% historical average growth in BV on that table in a somewhat different light and you should certainly question whether that series can be replicated going forward.

 

That then leaves the really difficult task of guessing what all of that means for the future.  My take is that the economics of the insurance industry will not allow FFH to routinely achieve its bogey of 15% growth in BV on a going forward basis.  I'd be happy if they could grow BV by 12%, but suspect that it may be more like 10% over the long term.  That's not a knock on FFH, it's just a tough industry.  If they actually do routinely achieve that 15% goal, I'll be the happiest guy around.

 

But, turning to valuation, if you pay 1.5x BV for something that grows its book by 15%, you are basically getting a 10% earnings yield with relatively rapid growth.  But, if you paid 1.5x BV and FFH grows BV by 10%, you are getting a 6.7% earnings yield and decent growth...and I'd say you'd have slightly overpaid, but not terribly so if the earnings are rock-solid, regular and predictable (in other words all of the things that the insurance industry is not!).  At this point it's trading at a shade over 1x BV, which means that the market finally figures that FFH is worth more alive than dead (ie, if it trades at <1x BV, in theory it's worth more dead than alive, so just break it up, sell off the pieces or run them off, and give the proceeds to the shareholders!).  I'd say that it's definitely worth 1.2x but I guess we'll see over the coming months whether the market agrees with that.  Maybe it'll rattle off 8 or 10 years of fabulous returns that will make me think it's worth 1.5x?  Time will tell....

 

SJ

Great conversation @StubbleJumper and @Cigarbutt . I think trying to model ROE more than 5 years out is largely a fools errand (i.e. the next 10 or 15 years). However, i think ROE can be modelled over the next 3 years and perhaps over the next 5 years with some accuracy. This is true for any company, including Fairfax. 

 

My guess is Fairfax has an excellent opportunity to deliver 15% ROE (on average) for each of the next 5 years. Why?

1.) As Buffett teaches us with the Aesop fable, the size, timing and certainty of cash flows are the three key variables. Fairfax will be earnings an enormous amount over the next 5 years ($20 billion?) - and we now have a high degree of certainty on this.  

2.) Fairfax is on a 6 year ‘hot streak’ when it comes to capital allocation. Financial markets have experienced wicked volatility over the past 4 years - active management matters a lot in this environment. Please re-read 1.) and add the effects of compounding. 

3.) It looks to me like Fairfax has been upgrading the quality of its insurance businesses. This suggest underwriting profit is likely to surprise to the upside moving forward (like it has in 2022 and 2023). 
4.) Fairfax has been aggressively upgrading the quality of their massive $17 billion equity portfolio for the past 6 years and the benefits of this are just starting to hit earnings. Historical numbers don’t help with this bucket - they misinform an investor.
5.) Fairfax’s $43 billion fixed income portfolio is perfectly positioned for the next few years and the benefits of this are flowing into earnings.

6.) Fairfax has been incubating holdings on its balance sheet for years (and decades). These will be monetized at the appropriate time and the company will book significant one time gains. Lumpy. But lumpy doesn’t mean they don’t exist. 

None of the 6 points above are being reflected in the current multiple. I think that’s crazy. As an investor, i love it.

If you want to look out more than 5 years you are really making a call on management and their capital allocation skills. Are they good? If so, how good?

 

Today most people on this board are now saying “yes, earnings look good for the next 3 or 4 years BUT it can’t possibly continue over the long term.”
 

With the stock trading so cheaply today it suggests Fairfax’s ROE will be poor in 2024, 2025, 2026 and every year after. IMHO, that is a massive disconnect from reality.

Edited February 17, 2024 by Viking

[MMM20]

5 hours ago, StubbleJumper said:

To correct a couple of points in your post, float does show up on the balance sheet as investments (ie, we take policy holders' premiums and invest them and that's where the float goes) and as policy liabilities.  What is more, float is not always at zero-cost or better, as there have been many years when the CR exceeded 100.  But life is definitely grand when that zero-cost or better situation prevails!

I disagree with your corrections. 
 

1) Yes of course the assets show up with an offsetting liability. But that’s missing something: the structural excess delta between the return on those investments and the cost of that financing. FFH doesn’t have to borrow at 6-8% because they can borrow at 0% by underwriting to breakeven. And that shows up in the income statement in higher earnings but not fully and properly on the balance sheet. That piece of the whole thing is an intangible asset that the accountants understate and our job as analysts to fix that. Right? I make a conservative assumption for that delta and capitalize it. Isn’t that the accurate way to handle it?
 

2) I’m talking about the ability to underwrite to breakeven or slightly better in the long run through cycles and volatility. The year to year volatility shouldn’t factor into the assessment of intrinsic value given sufficient discipline and quality on the insurance side - and they’ve proven that at this point IMHO. 

 

Does that make sense? What am I still missing?

 

Edited February 17, 2024 by MMM20

[valueventures]

Apologies if this has already been discussed in detail, but does anyone have updated thoughts around FFH's possible inclusion in the S&P/TSX 60 this year? Seems like the most mechanical means of facilitating more institutional involvement in the stock and driving up P/B.

 

I believe FFH already qualifies for the index based on its size, so curious to hear what is holding it back. Thanks!