Fairfax 2026

[djokovic1]

50 minutes ago, Marco Van Basten said:

Where are you getting average annual fixed income return = 4.2% over the past 40 years?  That seems extraordinarily low.  Even AI claims it averaged 4.61% for a 5 year treasury.  Given that Fairfax seems to have bought corporate debt and mortgages, my guess is that the return on the fixed income portfolio was north of 6% per annum over the past 40 years.  

Do you have a more accurate number? Where does the 6% come from precisely? Even if 5 year bond yields average 4.5% over the last 40 years, Fairfax has always run much shorter duration than that. How do you factor that in to your analysis?

 

Btw, even with your 6% assumption the equity returns are 12%+ which beats the S&P over 40 years.....can't argue with that?

 

 

47 minutes ago, Marco Van Basten said:

f we assume that those returns are indeed correct, then why didn't Fairfax raised a twenty billion or forty billion hedge fund from endowments and ran it with the incentive and management fee going to the company?

 

Why didn't Buffett do this?

Edited yesterday at 12:07 AM by djokovic1

[Viking]

I think it is pretty much impossible to determine a rate of return on Fairfax’s equity book over the past 40 years. What do you include? CDS? Equity hedges? TRS? Some of these were risk management positions, others were investments? 
 

But we do know with certainty one thing: Fairfax’s share price compounded at 19% for the past 40 years (US$; dividends reinvested). That is elite performance. Clearly, Fairfax is doing a number of things exceptionally well. 
 

Fairfax has two businesses: 

• Insurance
• Investments. 

We also know returns from the insurance business were weak pre-2010. 
 

As a result, I suspect investments have been an important driver of long term results.
 

What did they return each year?
 

No idea. And I don’t need to know… All I have to do is look at the stock price to know what I need to know. (Fairfax’s total performance has been elite.)

 

I am a quickly becoming a Thordike desciple…. Capital allocation is the most important thing over the long term for a company like Fairfax. Capital allocation drives per share results over the long term more than anything.

But assessing capital allocation is not easy - so most investors/analysts ignore it. Instead they focus on other things that are easier. 

Edited 20 hours ago by Viking

[Txvestor]

4 hours ago, djokovic1 said:

Do you have a more accurate number? Where does the 6% come from precisely? Even if 5 year bond yields average 4.5% over the last 40 years, Fairfax has always run much shorter duration than that. How do you factor that in to your analysis?

 

Btw, even with your 6% assumption the equity returns are 12%+ which beats the S&P over 40 years.....can't argue with that?

 

 

 

Why didn't Buffett do this?

Thats what I was saying about being invested in the model than the stock/equity picking acumen of HWIC. If they manage to do that, it would be icing on the cake. Even a mere 5% annual gain on the equity book get you to double digit equity returns. And we already currently have billions in unrealized gains in the trunk(almost worth around 15% of total equity worth). 

Edited 19 hours ago by Txvestor

[djokovic1]

3 hours ago, Viking said:

I think it is pretty much impossible to determine a rate of return on Fairfax’s equity book over the past 40 years. What do you include? CDS? Equity hedges? TRS? Some of these were risk management positions, others were investments? 

I agree it’s hard to be precise but I find it an interesting endeavor. Personally, I would include the CDS, hedging, TRS all of it and not exclude them. They were decisions taken by management at the time to maximize risk reward of the model. So I would first calculate the all-in number and then if you wanted to strip out specific items you can.

 

I don’t think it would be impossible to approximate a rough FI returns using historic financials (though it will take time and I may make an attempt), and if you have that number you will also have a directionally correct equity return.  It would be nice if management did the hard work for us as I’m sure they have the numbers

 

~15% CAGR of equity compounding for 40 years is mind blowing and so is 19% of bvps and share price compounding for 40 years.

[Maverick47]

2 hours ago, Txvestor said:

Thats what I was saying about being invested in the model than the stock/equity picking acumen of HWIC. If they manage to do that, it would be icing on the cake. Even a mere 5% annual gain on the equity book get you to double digit equity returns. And we already currently have billions in unrealized gains in the trunk(almost worth around 15% of total equity worth). 

Excellent point.  The prospect of double digit returns without a sizable risk of loss of capital is an ideal investment for me personally.  For more than ten years now I have set my personal expectations for investment returns at somewhere in the range of 6 to 8% annually.  That’s the level that should allow me to reach my reasonable goals for income in retirement.  I would be well satisfied with that result.

 

Finding a candidate for inclusion in my portfolio that has outpaced this target for the last five years, with a realistic prospect of doing so for the future five years as well, is already “icing on the cake” for my personal needs, and as you note, the business model of Fairfax and current store of unrealized gains may well produce results skewed to the upside.

 

Fairfax under Prem’s stewardship for 40 years has produced stellar annualized returns of roughly 19%.  I don’t need Fairfax to produce anything like this for the next 40 years in order to achieve my personal wealth goals,  and in fact, even the achievement of their current target of 15% would be far in excess of what I would hope for.

 

I am reminded of Charlie Munger’s comment regarding Berkshire Hathaway and Warren Buffett’s stewardship thereof, made sometime near the turn of the century:

 

"I think the top guy won't be as smart as Warren. But it's silly to complain, 'What kind of world is this that gives me Warren Buffett for 40 years and then some bastard comes along who's worse?'

[djokovic1]

3 hours ago, Viking said:

am a quickly becoming a Thordike desciple…. Capital allocation is the most important thing over the long term for a company like Fairfax. Capital allocation drives per share results over the long term more than anything.

 

And I 100% agree with this. I am lucky  to have spent some 1-1 time with him and discussed investments with his team. will also be seeing /meeting him at a conference tomorrow. Excited!

 

His book is my investment bible and assessing capital allocation is the most important aspect in my investment decision making.

[This2ShallPass]

https://x.com/ihtesham2005/status/2067336730693509150/video/1

Bill Ackman video, first half is a great pitch for owning Fairfax and second owning JOE (he talks about neither though)

[Marco Van Basten]

13 hours ago, djokovic1 said:

Do you have a more accurate number? Where does the 6% come from precisely? Even if 5 year bond yields average 4.5% over the last 40 years, Fairfax has always run much shorter duration than that. How do you factor that in to your analysis?

 

Btw, even with your 6% assumption the equity returns are 12%+ which beats the S&P over 40 years.....can't argue with that?

 

 

 

Why didn't Buffett do this?

Dude, you stated that the company owned a portfolio of 5 year bonds.  So that's what I went on.  Recently in the last couple of years, Fairfax disclosed or somebody stated on this board it was buying mortgages via KW at 8-10% yields.   So in my opinion, your assumption of 4.2% annual return from fixed income portfolio is untenable and cannot be supported.  

Over the last 40 years, investment grade bond index returned 6.16% annual compounded return.   So my assumption of 6% does not look unreasonable.

 

My view which I am clearly articulating poorly is this:

 

a) We have no idea how well the company's equity portfolio have performed over the long term.

b) Using reasonable assumptions it is possible to show that the equity portfolio has underperformed the S&P 500 over the past 40 years, and similarly it is possible to show that it has outperformed.

c) The equity portfolio is very unlikely to have materially outperformed the S&P over the past 40 years.

[Viking]

1 hour ago, Marco Van Basten said:

Dude, you stated that the company owned a portfolio of 5 year bonds.  So that's what I went on.  Recently in the last couple of years, Fairfax disclosed or somebody stated on this board it was buying mortgages via KW at 8-10% yields.   So in my opinion, your assumption of 4.2% annual return from fixed income portfolio is untenable and cannot be supported.  

Over the last 40 years, investment grade bond index returned 6.16% annual compounded return.   So my assumption of 6% does not look unreasonable.

 

My view which I am clearly articulating poorly is this:

 

a) We have no idea how well the company's equity portfolio have performed over the long term.

b) Using reasonable assumptions it is possible to show that the equity portfolio has underperformed the S&P 500 over the past 40 years, and similarly it is possible to show that it has outperformed.

c) The equity portfolio is very unlikely to have materially outperformed the S&P over the past 40 years.

Can you please define what you mean by equity portfolio? Are you talking stocks? Are you talking CDS/equity hedges (did well), equity hedges/shorts (did poorly) and TRS(did well)? 
 

What exactly are you trying to measure?

[LC]

I'm not as bullish on Fairfax's equity picks but over 40 years I agree w Viking the share price is a good indicator of the performance of the underlying earnings streams (investments, insurance). Hard to argue that, IMO.

[SafetyinNumbers]

1 hour ago, LC said:

I'm not as bullish on Fairfax's equity picks but over 40 years I agree w Viking the share price is a good indicator of the performance of the underlying earnings streams (investments, insurance). Hard to argue that, IMO.

The equity returns were strong when underwriting was weak. Then the equity returns were weak when underwriting was strong. That generated 18%+ CAGR in BVPS. Now they are both strong. It will be interesting to see if returns are better than the <10% built in to the share price. 

[Maverick47]

4 hours ago, Marco Van Basten said:

Dude, you stated that the company owned a portfolio of 5 year bonds.  So that's what I went on.  Recently in the last couple of years, Fairfax disclosed or somebody stated on this board it was buying mortgages via KW at 8-10% yields.   So in my opinion, your assumption of 4.2% annual return from fixed income portfolio is untenable and cannot be supported.  

Over the last 40 years, investment grade bond index returned 6.16% annual compounded return.   So my assumption of 6% does not look unreasonable.

 

My view which I am clearly articulating poorly is this:

 

a) We have no idea how well the company's equity portfolio have performed over the long term.

b) Using reasonable assumptions it is possible to show that the equity portfolio has underperformed the S&P 500 over the past 40 years, and similarly it is possible to show that it has outperformed.

c) The equity portfolio is very unlikely to have materially outperformed the S&P over the past 40 years.

Fair enough, @Marco Van Basten!
 

We know that there are points in the history of the company when the company performed admirably on the investment side of the business model and points when their performance was msignificantly subpar.  And as was probably the case with Berkshire Hathaway, it may well be the case that the stretches of significant outperformance for the company on a per share book value basis occurred earlier on, when the company was quite a bit smaller so history is not likely to be the best guide to the future.

 

What will really matter to us is how we expect them to perform in the near to intermediate future if we have included them in our own investment portfolios.

My simple way of looking at it is that if the company is able to earn 5% on their total investments, which are leveraged bout 3 to 1 for investments to shareholder equity,  also earns positive returns from underwriting profits, and doesn’t have too much of a drag from corporate overhead, interest expense and taxes, that a 15% or greater return on equity is eminently achievable.

 

If they earn closer to their long term historical return on total investments of 7.7%, then even if underwriting profits were to completely disappear, we’re still sitting at around the 15% ROE level.

Only if interest rates plummet and the total investment return drops well below 5% do we have a situation where a 15% ROE would appear to be out of reach without dramatically higher underwriting profits to act as an offset.

 

I don’t think we need to assume stellar equity performance (or even outperformance relative to the index) to make it likely that the company will produce good results for its shareholders.  But if anyone does wish to include in their estimate of intrinsic value the expectation that the current stable of equity investments will match or outperform the index going forward, that would in my opinion lead them to believe that an even greater margin of safety exists at the current market price.

 

My personal rough expectation is that there is a greater than 50% probability that the company will meet or exceed a 15% ROE over the next five years, and I believe the downside probability is quite limited.  Sort of a heads I win, tails I win a bit less and almost a zero chance of loss compared to current market price over the next five years.

[LC]

24 minutes ago, SafetyinNumbers said:

The equity returns were strong when underwriting was weak. Then the equity returns were weak when underwriting was strong. That generated 18%+ CAGR in BVPS. Now they are both strong. It will be interesting to see if returns are better than the <10% built in to the share price. 

 

My base case is a step-up in underwriting and middle-of-the-pack equity/investment returns. I don't see whatever it is that Prem et al see in Underarmor, Mattresses, and KW. That said I also would not have jumped into the mining business but Orla has done well. And there is stuff I do align with: Atlas Corp/Poseidon for instance (these guys took me under but hey what are ya gonna do) . And on the bond/fixed side I like what I see. 

 

That's essentially my 100ft view of why I am still holding here. 

[SafetyinNumbers]

25 minutes ago, LC said:

middle-of-the-pack equity/investment return

What’s middle of the pack numerically?

 

 

[LC]

@SafetyinNumbers I don't have a great way to quantify that but I'd say a lumpy return that matches the index or underperforms by 1-3 percent. 

 

Phrased another way: my expectation is that any future outperformance would more likely be driven by underwriting consistency, superior fixed income investment, and perhaps re-rating of the market multiple driven by more consistency in these two areas.

 

Equity returns I expect to have a much wider range of outcomes (investing in mattresses, stretchy clothes from dying brands, gold miners, CRE and rentals, and emerging market banks kind of gravitates towards that conclusion...). I am not banking on HW to knock it out of the park with those equity investments, but I'd love to be proven wrong!

[SafetyinNumbers]

18 minutes ago, LC said:

@SafetyinNumbers I don't have a great way to quantify that but I'd say a lumpy return that matches the index or underperforms by 1-3 percent. 

 

Phrased another way: my expectation is that any future outperformance would more likely be driven by underwriting consistency, superior fixed income investment, and perhaps re-rating of the market multiple driven by more consistency in these two areas.

 

Equity returns I expect to have a much wider range of outcomes (investing in mattresses, stretchy clothes from dying brands, gold miners, CRE and rentals, and emerging market banks kind of gravitates towards that conclusion...). I am not banking on HW to knock it out of the park with those equity investments, but I'd love to be proven wrong!

I’m not sure the index return has much to do with what Fairfax’s portfolio does. 
 

Do you consider the difference between economic return and accounting returns in your analysis? 

[Viking]

14 minutes ago, SafetyinNumbers said:

I’m not sure the index return has much to do with what Fairfax’s portfolio does. 
 

Do you consider the difference between economic return and accounting returns in your analysis? 

@LC, perhaps another way to ask the question is what cost base are you using when projecting future returns from the equity portfolio?

 

Is it Fairfax’s carrying value? Or market value?

 

The starting point matters. I use CV. Therefore, I think Fairfax will be able to outperform the broad market averages over the next 5 years. 
 

It really is an interesting thought exercise. The up-front assumptions matter a lot.
 

I appreciate you taking the time to comment… this stuff really is complicated. 

Edited 3 hours ago by Viking

[djokovic1]

I agree with most here (strongly so) that the beauty of investing in Fairfax today is that you don't need to rely on above average equity returns for realising great ROE and compounding given that other income streams are locked in and the inherent investment leverage in the business model. For me that point is moot.

I still find it a useful and interesting exercise to separate out the source of the returns as the data is there. At a high level, figuring out fixed income returns vs everything else. The everything else comprises of equity returns including TRS (positive), CDS (positive), hedging (big negative) + realised gains on assets held on book value which are worth more. The below analysis obviously doesn't incorporate the potential future returns from assets for example Poseidon and BIAL held at Book value which will show up sometime in the future when realised at market value.

 

@Marco Van Basten Dude ...if you like to be called that  I have to thank you for your push back because you are right, I was being approximate in my calculations for efficiency and you pushed me to be more specific. I still think you are very wrong in assuming Fairfax's non fixed income investments have underperformed the S&P over the long term, hopefully the below will help. 

 

I was able to go back to 1996 using CapIQ and Claude, so 30 years to calculate specific FI returns by year to 2025. I also have the total investment returns from 1996 to 2025. Using those pieces in the puzzle you can approximate the returns from everything that can't be attributed to fixed income for the 30 year period. Here are the results:

 

1996-2025 (30 year period) geometric annualised returns

Total investment return: 7.0%

Fixed income return: 3.9%

Average 5 year fixed income US treasury yield: 3.2%

Equity return + everything else (TRS, hedging, CDS): 16.6%

S&P Return including dividends: 10.3%
 
Note: this is assuming 75% / 25% fixed income equity split which I think is accurate. If you want to be conservative and use a 70%/30% split, the equity + other bucket return goes to 14.4%, still well well above the S&P return.

 

The fixed income returns are higher than the average 5 year fixed income yields for reasons pointed out by Marco ie higher yielding FI, but not by much and the overall conclusion remains the same as my initial intuition with rougher assumptions. 

For anyone who says Fairfax is mediocre investor, these long term results should provide further evidence over and above the stock price performance that they know what they are doing (clearly a lot of people still don't get it! as they judge Fairfax based on the lost decade and don't realise the handbrake of hedging is gone)