Fairfax 2024

[Parsad]

5 hours ago, gfp said:

I assume Mr. Block will be long gone by the AGM.  I think Fairfax does have a more "accurate" or "up-to-date" book value than, say, Berkshire and should trade at a lower price to book ratio than Berkshire all else being equal (they are not equal, Fairfax has higher growth, more float leverage and generally lower quality earnings vs. BRK).  Berkshire's carrying values rarely get re-marked higher by transactions.  Marmon and Pilot are two recent examples of required Fairfax-style write-ups but we will probably never see BNSF or GEICO marked up on Berkshire's balance sheet.

 

Again that comes back to Fairfax's reporting requirements as a Canadian reporter and Berkshire's reporting requirements as a U.S. reporter.  Under IFRS, you are required to mark assets at fair value or comparable value.  Berkshire hasn't adjusted See's Candies cost since acquiring it I believe. 

 

It's why Buffett says that Berkshire's intrinsic value is far higher than book value.  That analogy cannot be used with Markel or Fairfax, where there isn't massive amounts of undervalued assets on the book.  While both companies should trade higher than book...1-2 times, Berkshire probably should be trading between 2-3 times book.  Cheers!

[Parsad]

5 hours ago, TwoCitiesCapital said:

Going to be honest - the share price reaction to an a amazing earnings report is a bit disappointing

 

Makes me wonder if 1) the market had it "right" and the earnings rally was what we saw in January or 2) if we're back to the days of the getting the earnings look for free and the stock responds 2-3 days later

 

Either way - I am somewhat shocked we didn't get a pop of 5-10% from market participants who haven't been following this as closely as we have. 

 

I think markets had a pretty good idea book value would be significantly higher, thus the stock had a nice run up.  Now it will probably continue to rise moderately as we approach each quarterly report, because the markets know that there will be consistent growth in book value for several years.  The bulk of the gains from where the stock hit bottom have happened.  Now it will be based on annual growth of book and any market speculation where the multiple of book or earnings might be marked higher.  Cheers!

[ValueMaven]

5 minutes ago, Parsad said:

 

Again that comes back to Fairfax's reporting requirements as a Canadian reporter and Berkshire's reporting requirements as a U.S. reporter.  Under IFRS, you are required to mark assets at fair value or comparable value.  Berkshire hasn't adjusted See's Candies cost since acquiring it I believe. 

 

It's why Buffett says that Berkshire's intrinsic value is far higher than book value.  That analogy cannot be used with Markel or Fairfax, where there isn't massive amounts of undervalued assets on the book.  While both companies should trade higher than book...1-2 times, Berkshire probably should be trading between 2-3 times book.  Cheers!

 

BNSF is on BRK's books at purchase price of ~$45B, despite having a FV of at least $150B ... so yea, there is that

[Munger_Disciple]

5 hours ago, gfp said:

I assume Mr. Block will be long gone by the AGM.  I think Fairfax does have a more "accurate" or "up-to-date" book value than, say, Berkshire and should trade at a lower price to book ratio than Berkshire all else being equal (they are not equal, Fairfax has higher growth, more float leverage and generally lower quality earnings vs. BRK).  Berkshire's carrying values rarely get re-marked higher by transactions.  Marmon and Pilot are two recent examples of required Fairfax-style write-ups but we will probably never see BNSF or GEICO marked up on Berkshire's balance sheet.

 

+1

Book value & its growth are good proxies for the intrinsic value of FFH for this reason I think assuming FFH's marks are conservative. GAAP accounting is the main reason BRK's intrinsic value far exceeds book and likely one of the main reasons Buffett abandoned the book value metric (that and significant share repurchases above book but below intrinsic value). 

Edited February 16, 2024 by Munger_Disciple

[MMM20]

49 minutes ago, Parsad said:

 

Again that comes back to Fairfax's reporting requirements as a Canadian reporter and Berkshire's reporting requirements as a U.S. reporter.  Under IFRS, you are required to mark assets at fair value or comparable value.  Berkshire hasn't adjusted See's Candies cost since acquiring it I believe. 

 

It's why Buffett says that Berkshire's intrinsic value is far higher than book value.  That analogy cannot be used with Markel or Fairfax, where there isn't massive amounts of undervalued assets on the book.  While both companies should trade higher than book...1-2 times, Berkshire probably should be trading between 2-3 times book.  Cheers!

How does the massive float at FFH factor into your calculation of intrinsic value these days?

 

Edited February 16, 2024 by MMM20

[steph]

57 minutes ago, Parsad said:

 

Again that comes back to Fairfax's reporting requirements as a Canadian reporter and Berkshire's reporting requirements as a U.S. reporter.  Under IFRS, you are required to mark assets at fair value or comparable value.  Berkshire hasn't adjusted See's Candies cost since acquiring it I believe. 

 

It's why Buffett says that Berkshire's intrinsic value is far higher than book value.  That analogy cannot be used with Markel or Fairfax, where there isn't massive amounts of undervalued assets on the book.  While both companies should trade higher than book...1-2 times, Berkshire probably should be trading between 2-3 times book.  Cheers!

 

[steph]

I used to agree with this, but much less nowadays.  Reason being that with BRK you also pay 1,4 times the value of Apple and other huge listed portfolio + huge cash pile.  You don’t pay the real value of the unlisted companies, but you pay a big premium on cash and listed portfolio that has become a very big part of BRK. 

[glider3834]

4 hours ago, Haryana said:

You are adding dividends as if they were paid at the end.
They were paid annually at the beginning of each year.

 

When you account for that, the CAGR comes at ~16.5%.
Please see if you agree, time value of dividend.

 

I assumed no reinvestment of divs (eg assume you pop dividend money in a no interest account each year and leave it there and add it to your share gains over period) - which appears thats how Fairfax got their CAGR calc of 11.7% for 2017-2022 but feel free to double check.

 

You do see total return measures that include reinvested divs, so I think assuming no reinvestment like Fairfax do is more conservative.

Edited February 16, 2024 by glider3834

[villainx]

1 hour ago, Parsad said:

 

I think markets had a pretty good idea book value would be significantly higher, thus the stock had a nice run up.  Now it will probably continue to rise moderately as we approach each quarterly report, because the markets know that there will be consistent growth in book value for several years.  The bulk of the gains from where the stock hit bottom have happened.  Now it will be based on annual growth of book and any market speculation where the multiple of book or earnings might be marked higher.  Cheers!

 

It depends on time horizon, but if quality is good, I'm okay with moderate price action.  Berkshire staying in the 1.3-1.4 book range for quite a long bit of time allowed me to accumulate.   And it allows for better buyback. 

[nwoodman]

Just worked my way thru the Q4 23 conference call.  Jen Allen did a good job of addressing the MW allegations.  Lots of good information but my top four  responses were as follows:

 

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.’

 

2. Hard Market and Underwriting Discipline

 

Peter Clarke

“Sure, thanks Tom. I guess to address the Odyssey question first, in the fourth quarter Odyssey non-renewed a large residential property quota share, around $340 million of unearned premium they returned to the client, and that reduced their premium in the fourth quarter. But for us, it just shows the discipline Odyssey has and the focus on underwriting profit, and for us, that’s a great thing. They wrote that quota share for about two years. In their mind, the margins weren’t there going forward and they took the action necessary, so that was very good.

On the Brit side, we mentioned in prior quarters that they were reducing their catastrophe exposure, re-balancing it, and you continue to see that coming through the top line in the premium. A lot of the exposure they’re dropping is in the binder business, which takes a little longer to run off, and that’s why you’ve seen it come through a number of quarters.

On the pricing side, on the reinsurance side, we’re still seeing for most of our companies double-digit pricing, mainly on the property side, and then in insurance, mid-single digit price increases with the exception, as you highlighted, D&O and cyber, which had a lot of price increases over the last number of years, has been slowing down and actually reducing, so we haven’t been growing in those lines as much.”

 

3. Reserve Releases

Peter Clarke 

 

“I think our companies are still being very prudent on the hard market years - 2020, 2021, 2022, holding back from a lot of the favorable development that they’re seeing in those lines and just waiting that through to see how it ultimately plays out. We’re very focused on the effects of inflation and claims inflation in particular, so--but generally speaking, we think our reserves are in a very good position and we’re hoping going forward will benefit us.’ 

 

 

4.Associate income in particular Atlas/Poseidon

Prem Watsa

 

“By the way, our associate income, Atlas has provided the disclosure because of the new build program before they were taken private. They gave you a forecast - $300 million going to $600 million by 2025, and as of today, we still think that forecast is appropriate. When you put all of that together, we look at that operating income of $4 billion as a pretty conservative number.”

10,000m view

We are close to completing the share price regression to the mean phase.  I think the baseline from here is at least 12% CAGR.  We can argue what the upside can be but that will take care of itself.

Edited February 16, 2024 by nwoodman

[Hamburg Investor]

1 hour ago, Parsad said:

 

Again that comes back to Fairfax's reporting requirements as a Canadian reporter and Berkshire's reporting requirements as a U.S. reporter.  Under IFRS, you are required to mark assets at fair value or comparable value.  Berkshire hasn't adjusted See's Candies cost since acquiring it I believe. 

 

It's why Buffett says that Berkshire's intrinsic value is far higher than book value.  That analogy cannot be used with Markel or Fairfax, where there isn't massive amounts of undervalued assets on the book.  While both companies should trade higher than book...1-2 times, Berkshire probably should be trading between 2-3 times book.  Cheers!

Isn‘t that thinking about price to book ratios a bit „static“? In the end it‘s all about roe and not equity alone. And I see FFH better positioned fir a high roe:

 

- BRK is more like a conglomerate with an additional insurance arm. I haven‘t the exact numbers, but BRK has around 35 per cent leverage on equity through float; while FFH has around 130 per cent (and MKL is somewhere in between).

- Assuming both FFH and BRK get 4% on float and both get 10% on equity, than BRKs return on equity would be 11.5% and FFHs would be 15.2%. If it’s 6%, than BRKs roe is 12.0%, FFHs 17.8%. Even if BRKs „real“ equity would be bigger, it‘s returns on that equity will be lower for sure, so FFH will grow stronger

- BRK is 25 times bigger than FFH. It‘s pretty hard getting high roes for BRK over the next 1 or 2 decades on its equity; not so for MKL/FFH. 
- Just look at 2023 and the next 4 years: It‘s hard to argue for an roe way under 20%. That‘s a double on equity every 3.6 years (with 18% it’s every 4 years). I don‘t know anybody seeing BRK doubling every 3.6 or 4.0 years; that could happen to FFH though (I am not predicting that, but there is a chance).

- And today it’s way more normal times than it‘s been with such low interest the years before (if interest is zero, than it‘s way easier for BRK to not loose against MKL and FFH, as float doesn‘t give any returns, but FFH is levered most.

- MKL and FFH both grew stronger than BRK since 1986; I don’t see any reason, why that should change ultimately.

- Munger once said something like (from memory): In the end kver very long timeframes your CAGR return will

be relatively close to roe; regardless if you pay a high or low price. 
- That’s why I think, I‘d happily pay way higher pb ratios for FFH than for BRK. Let‘s assume an roe of 15% (Prems minimum goal…). What do I like to pay for that? Maybe a pe ratio of 15 to 25, let’s say 20. 15% roe on equity gives earnings of 0.15 at a pe ratio of 20 gives a pb ratio of around 3 to us. And BRK? Let’s say roe of 12% is doable. Than a lower pe ratio seems reasonable - say 15. Than you get a pb ratio of 1.8. 

[SafetyinNumbers]

2 hours ago, gfp said:

 

While a dividend is a dividend, I don't think it will have much of an effect here since Eurobank is receiving equity method accounting.  We are already recognizing our full share of Eurobank's profits as profit so a Eurobank dividend won't likely show up in the official 'interest and dividend income.' 

I think you are correct it won’t impact EPS but I think any dividend would reduce carrying value and go into dividend income. Lately, i don’t think the market is sophisticated at all but some investors might value dividend income higher than associates income because the former is more certain and the latter more volatile. 
 

On a separate note, I think the restating of 2022 earnings for IFRS 17 will make FFH screen better once the financial statements are filed and all of the data stores refreshed. Book value grew by $125 instead of $27 in 2022. The ROE over the last three years is almost impossible to believe for a company of this size. 
 

There is nothing to suggest that the next three years can’t look similar although that’s not my base case. I do think the odds of a 20% CAGR in BV including dividends to the end of 26 is higher than a 10% CAGR.

 

I think that it good enough to beat the market over the next three years  but most institutions are playing a different game. Now that game does include quants who might like the restated numbers more because of the smoothing. IFRS 17 is better for investors because of how investors invest. Looking backwards.
 

Who can blame them. It has worked and is working but FFH will get a lot prettier in a few weeks once the computers can see the numbers. It will look prettier to investors too especially on the returns table in the Annual Report.
 

[Dynamic]

The $125 estimate of normalized earnings would improve imply 12.2% normalized earnings yield at today's close of 1022 USD, which seems attractive, and with 4 years of approx $2 bn in interest and dividends pretty much assured, it seems likely to me that without big cat losses or a disastrously softening insurance market that we could well substantially beat the normalized figure over the next few years and perhaps can build the normalized run rate to something even higher. So 12% ish, is probably a fairly conservative lower bound expectation and there's a good prospect of something in the high teens to maybe lie 20s at least for the next 4 years or so.

 

I've only had a position in FFH (and now FRFHF too) for about half a year since becoming convicted I needed a starter position but I'm glad to have taken the chance to add to my position at about $910 USD last week and make this a high conviction part of my concentrated portfolio.

[Viking]

16 minutes ago, Dynamic said:

The $125 estimate of normalized earnings would improve imply 12.2% normalized earnings yield at today's close of 1022 USD, which seems attractive, and with 4 years of approx $2 bn in interest and dividends pretty much assured, it seems likely to me that without big cat losses or a disastrously softening insurance market that we could well substantially beat the normalized figure over the next few years and perhaps can build the normalized run rate to something even higher. So 12% ish, is probably a fairly conservative lower bound expectation and there's a good prospect of something in the high teens to maybe lie 20s at least for the next 4 years or so.

 

I've only had a position in FFH (and now FRFHF too) for about half a year since becoming convicted I needed a starter position but I'm glad to have taken the chance to add to my position at about $910 USD last week and make this a high conviction part of my concentrated portfolio.

The interesting thing to me is a few short years ago lots of people invested in Fairfax more for the investment gains not operating income. Now it has flipped and the focus is on operating income (which is not a bad thing).
 

The funny thing is the investments/insurance holdings have never been better positioned. At the same time, we are seeing record earnings getting reinvested each year into new income streams. This suggests to me that investment gains will be very robust moving forward.

Edited February 16, 2024 by Viking

[Santayana]

14 minutes ago, Viking said:

At the same time, we are seeing record earnings getting reinvested each year into new income streams.

This is the secret sauce.  The market is starting to recognize the earnings power that's there, but doesn't understand the compounding that can happen from here.

[StubbleJumper]

53 minutes ago, SafetyinNumbers said:

There is nothing to suggest that the next three years can’t look similar although that’s not my base case. I do think the odds of a 20% CAGR in BV including dividends to the end of 26 is higher than a 10% CAGR.

 

 

 

This is an interesting question.  The past 3 or 4 years have been fascinating because even with BV growing at 15% or 20%, FFH has been capital constrained.  If your insurance subs had an ROE of 15%, they pretty much needed to retain the totality of that income if they wanted to grow their book by 15%+ the next year (which they did, over and over!).  Every dollar that the insurance subs retained enabled them to write $1.70 or $2 of new premium, which earned 7% from underwriting profit last year plus roughly 4-5% from interest (Treasury bond rates), meaning that the capital retained in the insurance subs could basically earn an ROE of 20%-ish.  In short, over the past few years, the insurance subs were able to basically suck up pretty much as much new capital as FFH could earn, and it was effectively reinvested in growing the book very profitably at a rapid pace.

 

Okay, so what about 2023?  The insurance book grew 4.8% in 2023, which is far lower than the return on equity for the year.  Unless we see FFH suddenly put the pedal to the metal on underwriting, the insurance subs won't be sucking up nearly as much capital in 2024 to be invested in growing the book at the delicious ~20% ROE.  So, now maybe FFH will have to take a step or two down on the hierarchy of capital uses in 2024 by dividending more money to the holdco to be used for something else.   But, whatever that "something else" is, it will probably have an ROE lower than 20%.

 

Let's just hope that the underwriting accelerates in 2024. 

 

 

SJ

Edited February 17, 2024 by StubbleJumper

[SafetyinNumbers]

10 minutes ago, Santayana said:

This is the secret sauce.  The market is starting to recognize the earnings power that's there, but doesn't understand the compounding that can happen from here.

Not just at the Fairfax level but also at big holdings like Atlas and Eurobank.

[SafetyinNumbers]

14 minutes ago, StubbleJumper said:

 

 

This is an interesting question.  The past 3 or 4 years have been fascinating because even with BV growing at 15% or 20%, FFH has been capital constrained.  If your insurance subs had an ROE of 15%, they pretty much needed to retain the totality of that income if they wanted to grow their book by 15%+ the next year (which they did, over and over!).  Every dollar that the insurance subs retained enabled them two write $1.70 or $2 of new premium, which earned 7% from underwriting profit last year plus roughly 4-5% from interest (Treasury bond rates), meaning that the capital retained in the insurance subs could basically earn an ROE of 20%-ish.  In short, over the past few years, the insurance subs were able to basically suck up pretty much as much new capital as FFH could earn, and it was effectively reinvested in growing the book very profitably at a rapid pace.

 

Okay, so what about 2023?  The insurance book grew 4.8% in 2023, which is far lower than the return on equity for the year.  Unless we see FFH suddenly put the pedal to the metal on underwriting, the insurance subs won't be sucking up nearly as much capital in 2024 to be invested in growing the book at the delicious ~20% ROE.  So, now maybe FFH will have to take a step or two down on the hierarchy of capital uses in 2024 by dividending more money to the holdco to be used for something else.   But, whatever that "something else" is, it will probably have an ROE lower than 20%.

 

Let's just hope that the underwriting accelerates in 2024. 

 

 

SJ

Good points. I was focused more on returns on the investment portfolio and growth in associates income can push ROE closer to 20% than 10%.

[Parsad]

3 hours ago, MMM20 said:

How does the massive float at FFH factor into your calculation of intrinsic value these days?

 

 

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!

[Parsad]

2 hours ago, steph said:

I used to agree with this, but much less nowadays.  Reason being that with BRK you also pay 1,4 times the value of Apple and other huge listed portfolio + huge cash pile.  You don’t pay the real value of the unlisted companies, but you pay a big premium on cash and listed portfolio that has become a very big part of BRK. 

 

Yes, you are partially correct.  Huge investments like AAPL do have the opposite effect.  

 

That being said, a significant amount of BRK's value will never be valued correctly, thus BRK will almost always be valued significantly higher than book value.

 

That may continue to change under Abel, Todd and Ted if they invest more in public securities long-term.

 

Cheers!

[Parsad]

1 hour ago, Hamburg Investor said:

Isn‘t that thinking about price to book ratios a bit „static“? In the end it‘s all about roe and not equity alone. And I see FFH better positioned fir a high roe:

 

- BRK is more like a conglomerate with an additional insurance arm. I haven‘t the exact numbers, but BRK has around 35 per cent leverage on equity through float; while FFH has around 130 per cent (and MKL is somewhere in between).

- Assuming both FFH and BRK get 4% on float and both get 10% on equity, than BRKs return on equity would be 11.5% and FFHs would be 15.2%. If it’s 6%, than BRKs roe is 12.0%, FFHs 17.8%. Even if BRKs „real“ equity would be bigger, it‘s returns on that equity will be lower for sure, so FFH will grow stronger

- BRK is 25 times bigger than FFH. It‘s pretty hard getting high roes for BRK over the next 1 or 2 decades on its equity; not so for MKL/FFH. 
- Just look at 2023 and the next 4 years: It‘s hard to argue for an roe way under 20%. That‘s a double on equity every 3.6 years (with 18% it’s every 4 years). I don‘t know anybody seeing BRK doubling every 3.6 or 4.0 years; that could happen to FFH though (I am not predicting that, but there is a chance).

- And today it’s way more normal times than it‘s been with such low interest the years before (if interest is zero, than it‘s way easier for BRK to not loose against MKL and FFH, as float doesn‘t give any returns, but FFH is levered most.

- MKL and FFH both grew stronger than BRK since 1986; I don’t see any reason, why that should change ultimately.

- Munger once said something like (from memory): In the end kver very long timeframes your CAGR return will

be relatively close to roe; regardless if you pay a high or low price. 
- That’s why I think, I‘d happily pay way higher pb ratios for FFH than for BRK. Let‘s assume an roe of 15% (Prems minimum goal…). What do I like to pay for that? Maybe a pe ratio of 15 to 25, let’s say 20. 15% roe on equity gives earnings of 0.15 at a pe ratio of 20 gives a pb ratio of around 3 to us. And BRK? Let’s say roe of 12% is doable. Than a lower pe ratio seems reasonable - say 15. Than you get a pb ratio of 1.8. 

 

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!

[glider3834]

Looking at income from associates & non-insurance subs, its worth thinking about Grivalia Hospitality & BIAL which are significant investments. IMHO we haven't seen their normalised earnings/cash generation power reflected in Fairfax's results yet, because both have just completed major capex investment projects in 2023 & as they ramp up they are still operating below capacity in terms of passenger traffic (BIAL) and in terms of occupancy (Grivalia). Grivalia only opened the doors on its biggest resort this November https://www.linkedin.com/pulse/what-does-prem-watsa-gain-from-greek-tourism-thetotalbusiness

 

Also Fairfax has increased its ownership share in both businesses over the last 12 mths -so I am hoping in AR to see some clues as to how these might contribute to results in the coming year/s.

[gfp]

18 minutes ago, Parsad said:

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

 

I'll agree to disagree on that one! 

[dartmonkey]

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. 

[Viking]

It can be useful to look into the past. 13 short months ago i put together my estimate for 2022YE. It was dominated by investment losses (bonds and equities). Thank god for the pet insurance sale.

 

The more interesting forecast on the chart below is the estimate for 2023. At the time it looked crazy high. And of course, it has now been proven to have been crazy low. 

 

What are some of the lessons looking back on the 2023 forecast? 
1.) Valuing a turnaround is tough. 

2.) It is important to keep an open mind - when the facts change… update the model.

3.) Don’t get trapped by dogma (things that ‘have to happen’).