Fairfax 2025

[djokovic1]

2 hours ago, SafetyinNumbers said:

I think you are willing to make a forecast while the more pessimistic holders are basing their decision on historical valuation. 

 

Yes.  

 

And also, interestingly, prior to 2000, in the 1990's FFH mostly traded at a P/B of 1.7 to 3+. To be fair they were growing consistently at a high rate during that time.

 

 

But since the 2000's their P/B multiple has always been around one. Largely due to patchy performance either from underwriting, or investments or cat's. 25 years is a long time, so it’s understandable that most investors think ~1x is the right multiple based on historical valuation.

 

But if you do think its a new and improved Fairfax (which I do) since the cancellation of equity hedges and better underwriting, then going forward a 1x P/B assumption is wrong and too low.

 

Edited November 23, 2025 by djokovic1

[SafetyinNumbers]

14 minutes ago, djokovic1 said:

 

Yes.  

 

And also, interestingly, prior to 2000, in the 1990's FFH mostly traded at a P/B of 1.7 to 3+. To be fair they were growing consistently at a high rate during that time.

 

 

But since the 2000's their P/B multiple has always been around one. Largely due to patchy performance either from underwriting, or investments or cat's. 25 years is a long time, so it makes sense that investors think ~1x is the right multiple based on historical valuation.

 

But if you do think its a new and improved Fairfax (which I do) since the cancellation of equity hedges and better underwriting, then going forward a 1x P/B assumption is wrong and too low.

 

 

They had very high ROE in those early years but I think the multiple really exploded because investment to equity leverage increased significantly.  They were buying float so cheap and analysts couldn’t fathom how much of that float would get paid out over the years so forward ROE was likely overestimated. Still great deals but not as good as they looked. Now underwriting and investments are contributing to returns. If ROE were based on economic returns, it’s much higher than the accounting returns the last 4 years as Viking”s work highlights and a big part of the reason I have such high confidence in accounting ROE for the next 4 years. 
 

 

 

 

[djokovic1]

Yes agree with all of that. And the best part is the accounting ROE is mostly locked in for the next few years with fixed income, investment in associates and stable profitable underwriting (as shown the past 10 years). The mark to market equity returns (over the long term) + TRS (over the long term)+ buybacks are all additive to that.

Edited November 23, 2025 by djokovic1

[Viking]

1 hour ago, djokovic1 said:

 

Yes.  

 

And also, interestingly, prior to 2000, in the 1990's FFH mostly traded at a P/B of 1.7 to 3+. To be fair they were growing consistently at a high rate during that time.

 

 

But since the 2000's their P/B multiple has always been around one. Largely due to patchy performance either from underwriting, or investments or cat's. 25 years is a long time, so it makes sense that investors think ~1x is the right multiple based on historical valuation.

 

But if you do think its a new and improved Fairfax (which I do) since the cancellation of equity hedges and better underwriting, then going forward a 1x P/B assumption is wrong and too low.

 

@djokovic1, great analysis. Looking at its evolution over the past 40 years, it looks to me like Fairfax keeps morphing. 

• 1985-2000: Start-up? Build out P/C insurance business via aggressive acquisitions. Investment management business performing well. 
• 2000-2006: turnaround. Massive losses in P/C insurance business (under-reserved) almost sink the company. Investment management business continues to perform well.
• 2007-2010: Hedge fund? Reap massive gains from CDS/equity hedges.
• 2011-2020: investment management lost is way (equity hedges/shorts and poor equity investments) masking a slowly improving insurance business.
• 2021-2025: After a decade of work, high quality insurance business emerges. Investment management fixes its investment framework. And deals with poorly performing equity holdings. Result is high quality investment management business. 

It is like Fairfax has been on a journey as a company - to find the ‘right’ business model. It has continuously tried different things. Importantly, it has kept learning from its mistakes (first with insurance and later with investment management). The company that exists today has never existed before. Comparing it to past versions (and multiples) makes no sense to me. I think that is the point you are making - and I whole heartedly agree. 
 

It’s like Fairfax has been on a 40 years journey - trying to create a great company (able to exploit the P/C insurance business model). I think they just (finally) figured it out/cracked the code. Mr. Market just doesn’t see it yet. That is a great set-up for a long term investor (getting a compounding machine but not having to pay for it).

 

Another way to look at Fairfax is through the lens of a person. And their evolution and growth over time. Today, Fairfax is a talented, experienced and mature company - it is just entering its prime. 

Edited November 23, 2025 by Viking

[SafetyinNumbers]

51 minutes ago, Viking said:

It is like Fairfax has been on a journey as a company - to find the ‘right’ business model. It has continuously tried different things. Importantly, it has kept learning from its mistakes (first with insurance and later with investment management). The company that exists today has never existed before. Comparing it to past versions (and multiples) makes no sense to me. I think that is the point you are making - and I whole heartedly agree. 

I don’t think the insurance acquisitions were mistakes. They were risk adjusted bets that worked out on an economic basis but didn’t look great on an accounting basis after the initial acquisition. As for the investing side, we should expect them to be wrong a third of the time. We just don’t which ahead of time or the sequence of the mistakes. Otherwise, I fully agree!

[glider3834]

1 hour ago, SafetyinNumbers said:

 

They had very high ROE in those early years but I think the multiple really exploded because investment to equity leverage increased significantly.  They were buying float so cheap and analysts couldn’t fathom how much of that float would get paid out over the years so forward ROE was likely overestimated. Still great deals but not as good as they looked. Now underwriting and investments are contributing to returns. If ROE were based on economic returns, it’s much higher than the accounting returns the last 4 years as Viking”s work highlights and a big part of the reason I have such high confidence in accounting ROE for the next 4 years. 
 

 

 

 

another consideration looking at price to book, ROE and what is appropriate multiple is to consider over the last 7 years the goodwill & intangibles portion of book value for Fairfax's 5 largest insurers has actually been reduced in aggregate, due to acquisition related accounting eg amortization of customer relationships - even though these businesses have grown significantly. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Edited November 23, 2025 by glider3834

[SafetyinNumbers]

5 minutes ago, glider3834 said:

One thing to bear in mind when looking at price to book, is that over the last 8 years the goodwill & intangibles portion of book value for Fairfax's 5 largest insurers has actually been reduced in aggregate, due to acquisition related accounting eg amortization of customer relationships - even though these businesses have grown significantly. 

Great point. The buybacks above book value also reduce equity. Both actions are boosting forward ROE which in theory should boost the forward multiple. The conservative valuation of the equity portfolio and the high reserves also both act to boost current ROE but over time they make their way into earnings.

[Parsad]

2 hours ago, djokovic1 said:

 

Yes.  

 

And also, interestingly, prior to 2000, in the 1990's FFH mostly traded at a P/B of 1.7 to 3+. To be fair they were growing consistently at a high rate during that time.

 

 

But since the 2000's their P/B multiple has always been around one. Largely due to patchy performance either from underwriting, or investments or cat's. 25 years is a long time, so it’s understandable that most investors think ~1x is the right multiple based on historical valuation.

 

But if you do think its a new and improved Fairfax (which I do) since the cancellation of equity hedges and better underwriting, then going forward a 1x P/B assumption is wrong and too low.

 

 

You can pay what you want for Fairfax, but when it hit its peak in 1998, any shareholders who had bought then didn't see break even again until 2014...that's 16 years with no gain unless you were averaging in over 16 years.  Margin of safety serves a real purpose...it avoids irrational exuberance!  Cheers!

[benchmark]

4 minutes ago, Parsad said:

 

You can pay what you want for Fairfax, but when it hit its peak in 1998, any shareholders who had bought then didn't see break even again until 2014...that's 16 years with no gain unless you were averaging in over 16 years.  Margin of safety serves a real purpose...it avoids irrational exuberance!  Cheers!

Good point, but I assume you are not saying that there is irrational exuberance for fairfax now -- it's trading at 1.3x book atm? you are just saying that you would not pay more than book for it for your own margin of safety. 

[Parsad]

16 minutes ago, benchmark said:

Good point, but I assume you are not saying that there is irrational exuberance for fairfax now -- it's trading at 1.3x book atm? you are just saying that you would not pay more than book for it for your own margin of safety. 

 

Yes, correct.  Some people are assuming that Fairfax's long-term p/b should be around 2...I'm just saying that at that price, there might be some irrational exuberance.  I think fair value is around 1.3-1.5...if you can buy at a discount to that range, you will do well over time.  Cheers!

[Munger_Disciple]

1 hour ago, Parsad said:

 

You can pay what you want for Fairfax, but when it hit its peak in 1998, any shareholders who had bought then didn't see break even again until 2014...that's 16 years with no gain unless you were averaging in over 16 years.  Margin of safety serves a real purpose...it avoids irrational exuberance!  Cheers!

 

Excellent point about the all important "margin of safety". It's easy to overlook its necessity & importance in the case of Fairfax because of recency bias where everything went well for them (no major surprises in the insurance operational side or investing side). That won't always be the case. 

[SafetyinNumbers]

1 hour ago, Parsad said:

 

Yes, correct.  Some people are assuming that Fairfax's long-term p/b should be around 2...I'm just saying that at that price, there might be some irrational exuberance.  I think fair value is around 1.3-1.5...if you can buy at a discount to that range, you will do well over time.  Cheers!

A multiple between 1.3-1.5 suggests a very high return expectation vs long term equity index expectations of 10%. 
 

I think there are a few ways to triangulate intrinsic value and all get me well above 1.5x BV. I also don’t think Fairfax would be buying stock back at fair value. They would want a discount.

[Thrifty3000]

@Viking Buffett used to add intangible amortization back to cash flows to estimate owner earnings. Do you factor intangible amortization into your estimates? I assume it’s another $5 or $10 per share.

[mengan]

This is how I look at FFH economic earnings. Example for 2024:

 

Cash from operating activities before changes in working capital

4,509,700,000.00

Depreciation, amortization and impairment charges

(551,400,000.00)

Amortization of share-based payment awards

(164,900,000.00)

Other, incl. depreciation of right of use assets Interest Paid Interest paid on lease liabilities  Free Cashflow to Firm Before Change in Working Capital Allocated to Increase of Working Capital   Free Cashflow to Firm (FCFF)

(202,400,000.00) 508,800,000.00) 57,300,000.00)   5,224,200,000.00  (1,067,100,000.00) 4,157,100,000.00

 

Applied to multiple years, you get the following graph per share.

 

 

This assumes no adverse (huge) insurance event that undermines the reserving.

 

Reduce Interest Payment to get to FCF to equity owners.

Edited November 24, 2025 by mengan

[steph]

P/B is a very important metric for these type of businesses, but in the end it is cashflow and the reinvestment of this cashflow that will make this a good investment or not. 

FFH should easily do 150-180 usd EPS this year and the following years.  All in all this should continue to grow nicely.

So you pay 10 times conservative earnings for a business that has compounded at 19% over the last 40 years and that aims to compound at 15% over the long run.  To have done this over a very long term this business must have a very serious moat. 

Buying such a high quality business at such a low multiple is to me quite unique.  I am searching for these opportunities in the market but can’t find them.  

In my humble opinion at these levels I still have a very important margin of safety. 

 

[thowed]

12 hours ago, Parsad said:

 

You can pay what you want for Fairfax, but when it hit its peak in 1998, any shareholders who had bought then didn't see break even again until 2014...that's 16 years with no gain unless you were averaging in over 16 years.  Margin of safety serves a real purpose...it avoids irrational exuberance!  Cheers!

 

Thanks - that's wild - I didn't know that.  So pretty much the same as a Tech stock.  Looking forward to the book to get more up on the longer-term history.

 

[hardcorevalue]

Just a friendly reminder that P/B on FFH is losing relevance as it matures...

This is a fantastic company. A top 1% compounder at a single digit multiple with the S&P at 30x. 

I understand conservatism and cyclicality but this thing looks dirt cheap here to me. 

 

I think it's more likely that not that we never see the prior cycle p/b multiples for the rest of my life. 

 

There's a whole class of new shareholders that are going to discover Fairfax as the company continues to execute. The notion of float is not understood by the general public but they will look at it with amazement and bid this up. Index inclusion will come eventually here. 

Anyways, obviously super biased here but this looks cheap cheap cheap. 

[backtothebeach]

15 hours ago, Parsad said:

 

You can pay what you want for Fairfax, but when it hit its peak in 1998, any shareholders who had bought then didn't see break even again until 2014...that's 16 years with no gain unless you were averaging in over 16 years.  Margin of safety serves a real purpose...it avoids irrational exuberance!  Cheers!

 

Damn, I knew about the "lost decade" post GFC, but actually, if you got in right at the peak in 1998, you only had the dividend to show for 23.5 years.

 

https://www.fairfax.ca/wp-content/uploads/1998-Letter.pdf

"After a 9% increase in shares outstanding, earnings per share increased by
51% to $32.63 per share. Book value per share increased by 47% to $185 per share and our share price followed suit, increasing 69% to $540 per share"

 

"We have US$700 million in S&P Index puts"

 

So it peaked at ~3x BV. Does not look crazy in terms of earnings though.

 

This is probably old news for most here, but I think I am going to have to go back and read every annual report that is available, to really understand how this company has evolved over its history.

[dlr1493]

Hi everyone! Excited to read this. I had it on preordered on Amazon but cancelled as it won't arrive to me in time before I travel! However, I have seen it on Audible. From those that have read, am I best waiting to get my hands on a physical copy or do you think it'll still be good audio format? I live in UAE and can't find any physical copies here!

[Santayana]

58 minutes ago, backtothebeach said:

Damn, I knew about the "lost decade" post GFC, but actually, if you got in right at the peak in 1998, you only had the dividend to show for 23.5 years.

Not even that, as the dividend didn't start until 2006 as I recall.

[Jaygo]

17 minutes ago, dlr1493 said:

Hi everyone! Excited to read this. I had it on preordered on Amazon but cancelled as it won't arrive to me in time before I travel! However, I have seen it on Audible. From those that have read, am I best waiting to get my hands on a physical copy or do you think it'll still be good audio format? I live in UAE and can't find any physical copies here!

 

Audible would be fine. Other than some pictures of Prem, his family and staff there are not really any charts or graphs

[patterson]

While P/B may be losing relevance for FFH for many reasons (for example, as FV/CV grows), wasn’t book value always a flawed metric for insurance companies? Wouldn’t a better measure be adjusted BVPS + float per share – goodwill (part of the cost to acquire float)? That number is (very roughly) $2450 usd per share.

[Viking]

21 hours ago, Viking said:

@djokovic1, great analysis. Looking at its evolution over the past 40 years, it looks to me like Fairfax keeps morphing. 

• 1985-2000: Start-up? Build out P/C insurance business via aggressive acquisitions. Investment management business performing well. 
• 2000-2006: turnaround. Massive losses in P/C insurance business (under-reserved) almost sink the company. Investment management business continues to perform well.
• 2007-2010: Hedge fund? Reap massive gains from CDS/equity hedges.
• 2011-2020: investment management lost is way (equity hedges/shorts and poor equity investments) masking a slowly improving insurance business.
• 2021-2025: After a decade of work, high quality insurance business emerges. Investment management fixes its investment framework. And deals with poorly performing equity holdings. Result is high quality investment management business. 

It is like Fairfax has been on a journey as a company - to find the ‘right’ business model. It has continuously tried different things. Importantly, it has kept learning from its mistakes (first with insurance and later with investment management). The company that exists today has never existed before. Comparing it to past versions (and multiples) makes no sense to me. I think that is the point you are making - and I whole heartedly agree. 
 

It’s like Fairfax has been on a 40 years journey - trying to create a great company (able to exploit the P/C insurance business model). I think they just (finally) figured it out/cracked the code. Mr. Market just doesn’t see it yet. That is a great set-up for a long term investor (getting a compounding machine but not having to pay for it).

 

Another way to look at Fairfax is through the lens of a person. And their evolution and growth over time. Today, Fairfax is a talented, experienced and mature company - it is just entering its prime. 

 

I love reading about Fairfax's history. It is really interesting. But is it informational for an investor today? I think it is a potential minefield for many investors. 

 

My view is the best way to think about Fairfax's history over the past 40 years is through the lens of 5 completely different companies - each one had a very different business model (what was actually driving earnings at the time). And, of course, earnings is the key. 

 

Looking at past versions of the company (and valuations like P/BV) and then using that as a core input to understand and value the company today is really problematic (IMHO). And that is because too many important things have changed at the company. This exact issue has been tripping up many smart investors for the past 5 years - Fairfax's colourful history is messing them up. It is stopping them from being able to look at the company with fresh eyes - to understand and value it for what it is today (unencumbered by its past).

 

Charlie Munger was the master at understanding psychological biases - and how they constantly trip up investors. Fairfax is a real life example of this - it is tripping up people on this board. Is it any wonder the rest of the investing community is having such a hard time 'getting it' when it comes to understanding and valuing Fairfax? This is such a fascinating topic!   

 

Example: Druckenmiller talked about how he got his big break in his mid-20's. His boss at the time promoted him ahead of older and much more experienced traders. Why? He was young, smart, fresh and - most importantly - unencumbered by his past (he was a newbie). The older much more experienced traders had just been through some viscous bear markets - on the other side, Druckenmiller's boss knew they would be way too cautious. Lots of investors still have this exact same problem with Fairfax today - they continued to be haunted by its past. Remember this when you read Fairfax's history - there is a good change it is going to mess you up. 

 

Fairfax is like a kid who has been trying to figure out what they want to be when they grow up. Over the decades they kept trying different things. Learning what works and what doesn't. Over the past 5 years it appears Fairfax has finally figured it out - what they want to be. They have grown up right in front of our eyes. More mature. More responsible. Enjoying success. (Does it really help to know about their partying years?)

Edited November 24, 2025 by Viking

[SafetyinNumbers]

2 hours ago, patterson said:

While P/B may be losing relevance for FFH for many reasons (for example, as FV/CV grows), wasn’t book value always a flawed metric for insurance companies? Wouldn’t a better measure be adjusted BVPS + float per share – goodwill (part of the cost to acquire float)? That number is (very roughly) $2450 usd per share.

P/B matters for balance sheet businesses but it should be looked at in conjunction with ROE. Assets that are marked below FV but are earning an appropriate return will increase ROE all else being equal. The relationship between the BV and ROE should be exponential. At a 10% ROE a P/B of 1x makes sense for a 10% required return. At a 20% ROE P/B should be much higher than 2x because the compounding effect over time of the return on incremental capital. 
 

For Fairfax, I use BVPS + float, 15x earnings and 2.5x BVPS to triangulate intrinsic value. 

 

[SafetyinNumbers]

46 minutes ago, Viking said:

Fairfax is like a kid who has been trying to figure out what they want to be when they grow up. Over the decades they kept trying different things. Learning what works and what doesn't. Over the past 5 years it appears Fairfax has finally figured it out - what they want to be. They have grown up right in front of our eyes. More mature. More responsible. Enjoying success. (Does it really help to know about their partying years?)

This is really profound. A lot of investors like to invest based on analogs and so they look at charts to get comfort on what might come next even if all of the facts have changed.