dartmonkey Posted March 11 Posted March 11 33 minutes ago, Hamburg Investor said: So this a $887mn onetime earnings push for q2, isn‘t it? Yes, it seems that way, about $890m pre-tax, plus an increase of almost as much in the difference between carrying value and market value of the other half.
Crip1 Posted March 11 Posted March 11 13 minutes ago, dartmonkey said: Yes, it seems that way, about $890m pre-tax, plus an increase of almost as much in the difference between carrying value and market value of the other half. That's a question for accounting folks. Poseidon was carried at $2B and had a fair value of $2.5B. After selling half for $1.9B, does that mean the remaining half is still to be carried at $1B or is it carried at closer to $1.9B? One would think thaty the Fair Value would be closer to $1.9B but not sure of carrying value. Ultimately, as an owner it's not terribly meaningful as FFH still owns the same percentage of the company...I simply do not know the accounting for this. Would very much appreciate anyone who could enlighten us.
dartmonkey Posted March 11 Posted March 11 15 minutes ago, Crip1 said: That's a question for accounting folks. Poseidon was carried at $2B and had a fair value of $2.5B. After selling half for $1.9B, does that mean the remaining half is still to be carried at $1B or is it carried at closer to $1.9B? One would think thaty the Fair Value would be closer to $1.9B but not sure of carrying value. Ultimately, as an owner it's not terribly meaningful as FFH still owns the same percentage of the company...I simply do not know the accounting for this. Would very much appreciate anyone who could enlighten us. I believe the rules for an associate holding, where Fairfax has between 20% and 50%, mean that it is neither consolidated (as it would be if it owned more than 50%) nor marked to market (as it would be if they owned less than 20%.) Since thery are going from 45% ownership to 22%, the same treatment should apply, and the fact that they just sold for twice their carrying value means that fair value goes up but carrying value stays the same. 1
Sparky627 Posted March 11 Posted March 11 1 hour ago, Hamburg Investor said: So this a $887mn onetime earnings push for q2, isn‘t it? That’s correct. It should translate to $41.50 per share in capital gains, and the remaining 22K Poseidon shares get marked up to 28.30/sh.
Viking Posted March 11 Posted March 11 Poseidon: Fairfax’s Latest Capital Allocation Home Run Fairfax has been delivering a capital allocation master-class over the past five years. That may sound like hyperbole. It is not. All one has to do is look at the facts — at what management has actually done, not what the market says about it. The latest example arrived yesterday morning with Fairfax’s partial sale of its Poseidon investment. This is a significant transaction. Not only because of its immediate financial impact, but because of what it reveals about Fairfax’s business model, its investment framework, and management’s increasingly impressive record as capital allocators. A Big Transaction in a Big Holding Poseidon is Fairfax’s third-largest equity holding. At December 31, 2025, Fairfax carried its investment in Poseidon at approximately $2.0 billion, or $15.50 per share, while the position’s market value was approximately $2.6 billion, or $20.00 per share. That meant excess of fair value over carrying value stood at roughly $600 million. Yesterday, Fairfax announced that it had agreed to sell approximately 50% of its Poseidon position — 67.6 million shares— at $28.30 per share, for aggregate proceeds of approximately US$1.91 billion. That sale price changes the picture dramatically. At $28.30 per share, Fairfax’s full Poseidon stake is effectively being valued at approximately $3.7 billion. Compared to year-end 2025, that is an increase of roughly: $1.7 billion above carrying value $1.1 billion above market value That is a very large uplift in value on one of Fairfax’s largest holdings. How the Value Shows Up What makes the transaction especially interesting is that it surfaces value in two different ways. First, Fairfax will record a realized investment gain on the shares being sold. On the roughly half of the position that is being monetized, the company is expected to book an investment gain of approximately $866 million when the transaction closes in the second quarter. Second, the sale price establishes a fresh market-based valuation benchmark for the half of the position Fairfax continues to own. After the sale, Fairfax will still retain an ownership interest of approximately 22.1% in Poseidon. Using the transaction price, the remaining common equity stake would have a market value of about $1.8 billion. Excess of fair value over carrying value on that remaining position rises to approximately $824 million, up from $600 million at year-end. Put differently, this one transaction surfaces roughly $1.1 billion of additional value for Fairfax shareholders between the realized gain and the higher implied value of the continuing stake. That is an outstanding result. Fairfax has, in effect, managed to have its cake and eat it too: it has crystallized a very large gain while still retaining a significant ownership interest in the business. A Home Run, Made More Powerful by Size The investment performance since inception is remarkable. Fairfax first began investing in Seaspan in 2018. Over a period of roughly four years, it invested a little over $1 billion to build a position of approximately 132 million shares, representing about 43% of the company. Very rough math suggests Fairfax paid about $7.90 per share on average. Today’s sale price is $28.30 per share. That means an initial investment of roughly $1 billion has now become a position worth approximately $3.7 billion, before considering the dividends Fairfax has received along the way. Include those dividend payments and the total value created likely approaches $4 billion. On very rough math, Fairfax has generated close to $3 billion of profit from this investment over an eight-year period. Yes, the percentage return is excellent — roughly 300%. But what makes the outcome truly exceptional is not simply the return. It is the size of the bet. Back in 2018, Fairfax’s common shareholders’ equity was approximately $11.8 billion. A $1 billion investment was not a small side position. It was a major capital allocation decision. It was a high-conviction bet. Stanley Druckenmiller once observed that what matters in investing is not simply whether you are right or wrong, but how much money you make when you are right and how much you lose when you are wrong. Poseidon is a textbook case. Fairfax was right, and the position was large enough to matter. Mr. Market Still Does Not Quite Get It As is so often the case with Fairfax, the market’s reaction was curiously muted. Despite the scale of the transaction and the value it surfaces, Fairfax shares finished the day up only about 1%. That reaction suggests that many investors still do not fully grasp either the immediate financial implications or the broader significance of what management has accomplished. This is not new. Fairfax has often monetized assets at attractive prices with surprisingly little market recognition (initially). One is reminded of the sale of the pet insurance business in 2022, which generated a gain of roughly $1.2 billion and was likewise not met with the kind of re-rating one might have expected. That the stock did not sell off this time may count as progress. Earnings Estimates Are Going Higher This transaction also has a near-term consequence that should be obvious: analyst estimates for 2026 are likely headed higher. Between the Poseidon sale and the recently announced sale of Eurolife’s life insurance business, Fairfax is now set to book approximately $1.2 billion of investment gains in the second quarter alone: Poseidon: ~$866 million Eurolife life insurance business: ~$350 million That is about $53 per diluted share after tax. The amount is material, and it is not reflected in analyst models today. Upward earnings revisions are coming. More importantly, we are only about ten weeks into the year. There are still more than forty weeks to go. Fairfax may well surface additional value before 2026 is done. Against that backdrop, an earnings estimate of $190 to $200 per diluted share for 2026 is beginning to look conservative. At a share price of roughly $1,685, Fairfax is trading at less than 9 times normalized earnings. For a company with Fairfax’s collection of assets, balance sheet strength, management quality, and capital allocation track record, that remains a remarkably low valuation. The Hidden Value Flywheel For some time now, we have been making the case that Fairfax is carrying a very large amount of hidden value on its balance sheet. At year-end 2025, a conservative estimate placed this figure at more than $4.5 billion. To some, that may have sounded overly optimistic. Transactions like this are a useful reminder that hidden value is not imaginary. It is simply unrealized — until management chooses to realize part of it. That is an important distinction. The Poseidon sale did not create value out of thin air. It revealed value that had already been built inside the business and its investment portfolio. And importantly, this transaction only modestly reduces the broader reservoir of unrealized value that still resides within Fairfax. That is why this matters so much. The sale of Poseidon is not just a one-off gain. It is evidence that Fairfax’s hidden value flywheel is real and that it can become a meaningful source of future reported earnings over time. Fairfax Sells Assets — That Is Part of the Model One of the recurring mistakes made by investors analyzing Fairfax is to treat asset sales as unusual or non-recurring. They are neither. Monetizing assets is a normal part of Fairfax’s business model. It has been for decades. Why does Fairfax sell assets? First, to realize strong returns. Second, to improve the quality of the company by recycling capital into new opportunities with superior prospective returns. This is what good capital allocation looks like. It is not passive. It is not static. It is a continuous process of building value, harvesting value, and reallocating capital. Will the gains be lumpy from quarter to quarter or year to year? Of course. But over a two- or three-year period, the pattern is much smoother and much easier to understand. Fairfax’s investment gains should not be viewed as random windfalls. They are better understood as the periodic monetization of value that has been compounding quietly beneath the surface. Poseidon is an excellent example. New Information Matters The other important thing this sale provides is information. Fairfax has been steadily increasing the size of its private equity holdings. By definition, these positions are harder for outside investors to value with confidence. Public marks are limited. Accounting carrying values can understate economic value. As a result, many observers default to skepticism. Asset sales help solve that problem. They create market-based reference points. In this case, Fairfax sold approximately half its Poseidon position to three separate buyers, including one existing shareholder increasing its stake and two new strategic investors. It is reasonable to conclude that these are informed counterparties and that $28.30 per share is a credible estimate of current fair market value. That matters not just for Poseidon, but for how investors should think about Fairfax’s broader portfolio of privately held assets. “Bet on the Jockey” Perhaps the most interesting aspect of the Poseidon investment is what it says about the evolution of Fairfax’s investment framework. My long-standing view is that Fairfax made an important adjustment around 2018. Management quality — the “jockey” — appears to have become an even more central part of the decision-making process. Poseidon was one of the earliest and clearest examples of this shift. Fairfax invested in Seaspan because of David Sokol. That was, at the time, a contrarian call. Sokol had once been viewed as Warren Buffett’s heir apparent at Berkshire Hathaway. After the Lubrizol episode, however, he was clearly out of favor in many circles. Fairfax looked past the controversy and focused on the underlying record and capability of the operator. That is classic contrarian investing: buy low, not only in securities, but sometimes in people’s reputations. It has worked out fabulously well. Over the past eight years, David Sokol and CEO Bing Chen have done an outstanding job building the business. Fairfax’s 2025 annual report makes this plain. Since 2017, the fleet has grown from 89 ships to 184, revenue has risen from $831 million to $2.5 billion, net income has increased from $175 million to $727 million, and earnings per share have climbed from $1.33 to $2.49 despite a much larger share count. Those are not the results of financial engineering. They reflect real operating execution. Fairfax’s gain on Poseidon is, in the end, the direct consequence of partnering with capable people and backing them with meaningful capital. 2018 Was a Very Good Vintage It is also worth noting that Poseidon was not the only example of this pattern. In 2018, Fairfax also made a major investment in Stelco, partnering with Alan Kestenbaum — another clear “bet on the jockey.” That investment, too, became a major success. Fairfax’s roughly $193 million initial investment ultimately delivered a total return of approximately $568 million, or about 294%, over six years. Some of that capital was likely recycled into Orla Mining in late 2024, which itself has already become another highly successful Fairfax investment. That sequence is worth studying. Fairfax identifies strong operators, invests at attractive valuations, supports them over time, monetizes when appropriate, and recycles the capital into the next above-average opportunity. This is not accidental. It is a system. Final Thought The Poseidon sale tells us several things. It tells us that Fairfax continues to uncover and monetize value in size. It tells us that the company’s hidden value is real. It tells us that management understands how to recycle capital intelligently. And it tells us that Fairfax’s growing emphasis on backing exceptional operators has produced excellent results. Most importantly, it tells us that Fairfax is still being underestimated. The market tends to focus on reported earnings in the next quarter. Fairfax is building value across a much longer time horizon. That difference in perspective continues to create opportunity for shareholders willing to do the work. Poseidon is not just a successful investment. It is a case study in how Fairfax compounds capital.
djokovic1 Posted March 11 Posted March 11 Thanks @Viking Love your summaries. So insightful. I think we are all a little bemused with the share price performance given the string of strong fundamental news. But their capital allocation is what makes Fairfax such an 'easy' investment. Even if the stock price doesn't move today we know management is buying back stock aggressively. It's a question of when not if, the stock price catches up to intrinsic value. I am very curious to see what the pace of buybacks are in March (as discussed earlier, early indications are that they are buying back 4-5x what they were in Jan/Feb at a pace of 15% annualized)
dartmonkey Posted March 11 Posted March 11 1 hour ago, djokovic1 said: I am very curious to see what the pace of buybacks are in March (as discussed earlier, early indications are that they are buying back 4-5x what they were in Jan/Feb at a pace of 15% annualized) Yes, me too. If we have to have the low share price (around 8 times earnings), at least it allows for really moving the needle with share repurchases. They won't have the cash from the Eurolife and Poseidon sales until Q2, but now that they know they will have it soon, they can be much more aggressive with the buybacks, and still have plenty of cash to repurchase whatever insurance stakes they want to. (I don't know what the time limits on these shares is, I haven't been able to find the Offer Documents from the 2021 sale of 9.9% of Odyssey to OMERS and CPP, along with whatever other minority shares they have the right to repurchase, and the press release just said they would be able to buy back the Odyssey stakes, but not until when: "After closing, Fairfax will retain the flexibility to repurchase the interests of OMERS and CPPIB Credit Investments in Odyssey Group over time." With what looks like the winding down of the hard insurance market, there should be lots of cash going from insurance subsidiaries to the holding company anyways, but with an extra billion, along with a share price that has gone nowhere since May 2025, and 2.2m shares authorized in the current normal course issuer bid, we might see the repurchase numbers move a lot higher.
Viking Posted March 11 Posted March 11 (edited) 1 hour ago, djokovic1 said: Thanks @Viking Love your summaries. So insightful. I think we are all a little bemused with the share price performance given the string of strong fundamental news. But their capital allocation is what makes Fairfax such an 'easy' investment. Even if the stock price doesn't move today we know management is buying back stock aggressively. It's a question of when not if, the stock price catches up to intrinsic value. I am very curious to see what the pace of buybacks are in March (as discussed earlier, early indications are that they are buying back 4-5x what they were in Jan/Feb at a pace of 15% annualized) @djokovic1, I appreciate the comment. there are so many interesting angles to the Poseidon sale. A big one for me is the hidden value theme (that it is much more than most people realize) and that it will become a more important driver of future reported results (as positions get monetized). Both of those ideas are being confirmed with the Poseidon sale. In terms of Mr. Market, there has been one constant in the 25 years I have been following Fairfax - Mr. Market just doesn't like the company, its leadership and its business model. (The 'proof' is the stock price.) There are lots of reasons for this. While it is a little annoying at times (like today), it really is a very good thing for Fairfax. And that is because it will give them the opportunity to continue to buy back ~1 million shares every year at a very attractive valuation. 2026YE P/BV = ~1.15, and an investor gets all the hidden value for free (it will provide a material boost to future earnings like we are seeing this year with Poseidon). That is crazy cheap for a company like Fairfax. 1H 2026 is shaping up to be a very active and interesting period for Fairfax: Sources of capital: Debt Issuance: $475M ($292 + $183) Eurolife sale: $945M Poseidon sale: $1.9B Dividends from insurance co's: ? Uses of capital: Dividend payment: ~$300M YTD Share buybacks: $384 ($1,692/share) Some share buybacks will be for compensation programs AGT IPO: $400M (~$250 conversion + $150 new) Under Armour: ~$150M (increase in position) Kennedy Wilson take Private: $1.65B Allied World: $1B? (buy out minority partner) IDBI Bank: $8B? (no idea on if this happens or amount) Bottom line, Fairfax has a lot of uses of capital. So it is not surprising to see them monetize 50% of Poseidon. I would not be surprised to see them monetize one or two more positions. ---------- As @glider3834 pointed out: 'Subsequent to December 31, 2025, the company purchased for cancellation 226,694 subordinate voting shares under the terms of its normal course issuer bids at a cost of $384.0' Edited March 12 by Viking
Txvestor Posted March 12 Posted March 12 5 hours ago, Crip1 said: That's a question for accounting folks. Poseidon was carried at $2B and had a fair value of $2.5B. After selling half for $1.9B, does that mean the remaining half is still to be carried at $1B or is it carried at closer to $1.9B? One would think thaty the Fair Value would be closer to $1.9B but not sure of carrying value. Ultimately, as an owner it's not terribly meaningful as FFH still owns the same percentage of the company...I simply do not know the accounting for this. Would very much appreciate anyone who could enlighten us. Not just that but they also have 12M x $25 series J preferred notes with a 7% coupon that starts adjusting up by 1.5% each year from this June onwards. Im thinking those might be redeemed by Poseidon before the end of Q2. I'm not sure how those are accounted either.
SafetyinNumbers Posted March 12 Posted March 12 10 minutes ago, Viking said: @djokovic1, I appreciate the comment. there are so many interesting angles to the Poseidon sale. A big one for me is the hidden value theme (that it is much more than most people realize) and that it will become a more important driver of future reported results (as positions get monetized). Both of those ideas are being confirmed with the Poseidon sale. In terms of Mr. Market, there has been one constant in the 25 years I have been following Fairfax - Mr. Market just doesn't like the company, its leadership and its business model. (The 'proof' is the stock price.) There are lots of reasons for this. While it is a little annoying at times (like today), it really is a very good thing for Fairfax. And that is because it will give them the opportunity to continue to buy back ~1 million shares every year at a very attractive valuation. 2026YE P/BV = ~1.15, and an investor gets all the hidden value for free (it will provide a material boost to future earnings like we are seeing this year with Poseidon). That is crazy cheap for a company like Fairfax. 1H 2026 is shaping up to be a very active and interesting period for Fairfax: Sources of capital: Debt Issuance: $475M ($292 + $183) Eurolife sale: $945M Poseidon sale: $1.9B Dividends from insurance co's: ? Uses of capital: Dividend payment: ~$300M YTD Share buybacks: $384 ($1,692/share) Some share buybacks will be for compensation programs AGT IPO: $400M (~$250 conversion + $150 new) Under Armour: ~$150M (increase in position) Kennedy Wilson take Private: $1.65B Allied World: $1B? (buy out minority partner) IDBI Bank: $8B? (no idea on if this happens or amount) Bottom line, Fairfax has a lot of uses of capital. So it is not surprising to see them monetize 50% of Poseidon. I would not be surprised to see them monetize one or two more positions. ---------- As @glider3834 pointed out: 'Subsequent to December 31, 2025, the company purchased for cancellation 226,694 subordinate voting shares under the terms of its normal course issuer bids at a cost of $384.0' I think we should be careful about mixing up the insurance subsidiaries holdings and the holding company although FFH did hold a few hundred million of Poseidon at the holdco, most of the proceeds are going to the insurance subsidiaries. Fairfax has about $4b they are allowed to dividend up from the insurance subsidiaries. The last few years they have sent about half of that up and used all of it for buybacks. That’s what I’m expecting this year too. Adding in the $400m of proceeds from Poseidon, they might have $2.4b to spend. I assume they have been out of the market for at least Monday and Tuesday post release of the annual report but maybe they got started again today.
SafetyinNumbers Posted March 12 Posted March 12 1 hour ago, Viking said: A big one for me is the hidden value theme (that it is much more than most people realize) and that it will become a more important driver of future reported results (as positions get monetized). Both of those ideas are being confirmed with the Poseidon sale. This is such an important point. We are seeing the examples one after another for years and it makes sense because of IFRS accounting. The gains are more predictable than they appear because we can value a lot of the equity portfolio ourselves and see how cheap it is. We can see that intrinsic value in these businesses is growing while the market value or reference price may not reflect it because the market isn efficient. I can get up to $10b in unrealized gains that we know are likely to be realized over time. It’s not a static number and might be growing 5-15% a year depending on underlying performance of the portfolio. This alone can boost ROE dramatically in any given year, like Poseidon and Eurolife will do in 2026. I trained as a CA and my favourite part of training to write the UFE (old Canadian qualifying exam) were the cases on valuation. I wrote case after case analyzing an agreement where a controlling shareholder has an agreement to buy minority shareholders out based on a fixed multiple. The case would require us to give advice to the minority holders. In a way, the Fairfax annual report, is like a long UFE case. The controlling shareholder is buying back stock based on a multiple of price to book and we know the book value is understated because of the nature of IFRS accounting for significant influence and control positions. Fairfax didn’t start buying positions above 20% in operating companies until 2012, from what I can gather. If you know otherwise please let me know. Under the accounting rules, bad news leads to immediate layoffs. Good news results in book value going up by FFH’s share of earnings, not by a multiple of those earnings and certainly doesn’t reflect multiple expansion. Eurobank is a clear example and a rare one where we have a liquid observable market price. It’s easy to see how this keeps book value constrained but boosts operating ROE significantly. Fairfax understates book value in other creative ways too. At Fairfax India, they use high discount rates. Much like at the mothership, now that we are 11 years in, it’s hard to hold back the gains. BIAL, the most important piece has been written up for the last 2 quarters rolling the DCF forward as it is now above the Siemens purchase price. That sort of book value momentum in theory should be attractive to value investors although lately they seem happy to sell both FFH and FIH. On the insurance side, they are consistently releasing reserves because they are over reserved. Excess reserves are like deferred equity i.e. future book value. Coming off a hard market we can have especially high confidence that it’s going to show up in ROE over the next few years too. The short term reaction from the market to the Poseidon deal tells us a lot about market structure. Stock prices have either have higher correlation to revenue growth or at the very least the marginal Canadian institutional investor thinks they do. The last hard market our multiple went from 0.6x to 1.6x from October 2020 to July 2025 as institutions chased revenue growth. So far in the soft market we have retraced to 1.3x and we could certainly go lower but since the company has bought as high as 1.5x BV, my guess is during this soft market we will eventually get back to that multiple. The next hard market is when we could see significant multiple expansion. The hard part will be holding on as FFH may head to multiples we haven’t seen since the late 90s.
Crip1 Posted March 12 Posted March 12 5 hours ago, Viking said: Mr. Market Still Does Not Quite Get It As is so often the case with Fairfax, the market’s reaction was curiously muted. Despite the scale of the transaction and the value it surfaces, Fairfax shares finished the day up only about 1%. That reaction suggests that many investors still do not fully grasp either the immediate financial implications or the broader significance of what management has accomplished. This is not new. Fairfax has often monetized assets at attractive prices with surprisingly little market recognition (initially). One is reminded of the sale of the pet insurance business in 2022, which generated a gain of roughly $1.2 billion and was likewise not met with the kind of re-rating one might have expected. That the stock did not sell off this time may count as progress. No question about it. "Not quite getting it" is an understatement. I was looking at the FFH Value Line update dated 2/27, shortly before earnings came out. Here is what happened since 27-Feb: Actual F/Y EPS of $213.78 vs $201.35 Value Line estimate – 6.2% higher Actual Y/E BV/Share of $1,260.19 vs $1,238.65 Value Line estimate – 1.7% higher Actual shares outstanding are 20.9M vs 22.0M Value Line estimate – 5% lower Sale of an asset which will produce a gain of $40+/Share while also increasing Fair Value over Carrying Value Objectively, this is a pretty damned impressive 2 weeks by any measurement. Mr. Market's take is that shares are 2.4% (US) and 3.5% (TSE) lower than they were at market close 27 Feb. Without question, they don't get it. But, this is what makes a market...those of us who DO get it can smile. -Crip
Duke In Shadows Posted March 12 Posted March 12 6 hours ago, Viking said: Stanley Druckenmiller once observed that what matters in investing is not simply whether you are right or wrong, but how much money you make when you are right and how much you lose when you are wrong. Poseidon is a textbook case. Fairfax was right, and the position was large enough to matter. Fairfax Sells Assets — That Is Part of the Model One of the recurring mistakes made by investors analyzing Fairfax is to treat asset sales as unusual or non-recurring. They are neither. Monetizing assets is a normal part of Fairfax’s business model. It has been for decades. Why does Fairfax sell assets? First, to realize strong returns. Second, to improve the quality of the company by recycling capital into new opportunities with superior prospective returns. This is what good capital allocation looks like. It is not passive. It is not static. It is a continuous process of building value, harvesting value, and reallocating capital. Will the gains be lumpy from quarter to quarter or year to year? Of course. But over a two- or three-year period, the pattern is much smoother and much easier to understand. Fairfax’s investment gains should not be viewed as random windfalls. They are better understood as the periodic monetization of value that has been compounding quietly beneath the surface. Poseidon is an excellent example. two things. this is Soros not Druck: "it's not whether you are right or wrong, but how much money you make when you are right and how much you lose when you are wrong" the part on selling assets is interesting. When you listen to the CEO of Sleep Country Canada around the time Fairfax acquired the company, he explained that they chose Fairfax over several private equity firms partly because of Fairfax’s long-term approach. They were looking for a permanent home, and that played a big role in their decision. The deal flow Prem and company get over the next decade will depend on how they treat their private businesses. If they simply flip companies like Peak or Sleep Country in two years, they won’t have the same reputation as long-term owners.
Hsmpanl Posted March 12 Posted March 12 1 hour ago, SafetyinNumbers said: This is such an important point. We are seeing the examples one after another for years and it makes sense because of IFRS accounting. The gains are more predictable than they appear because we can value a lot of the equity portfolio ourselves and see how cheap it is. We can see that intrinsic value in these businesses is growing while the market value or reference price may not reflect it because the market isn efficient. I can get up to $10b in unrealized gains that we know are likely to be realized over time. It’s not a static number and might be growing 5-15% a year depending on underlying performance of the portfolio. This alone can boost ROE dramatically in any given year, like Poseidon and Eurolife will do in 2026. I trained as a CA and my favourite part of training to write the UFE (old Canadian qualifying exam) were the cases on valuation. I wrote case after case analyzing an agreement where a controlling shareholder has an agreement to buy minority shareholders out based on a fixed multiple. The case would require us to give advice to the minority holders. In a way, the Fairfax annual report, is like a long UFE case. The controlling shareholder is buying back stock based on a multiple of price to book and we know the book value is understated because of the nature of IFRS accounting for significant influence and control positions. Fairfax didn’t start buying positions above 20% in operating companies until 2012, from what I can gather. If you know otherwise please let me know. Under the accounting rules, bad news leads to immediate layoffs. Good news results in book value going up by FFH’s share of earnings, not by a multiple of those earnings and certainly doesn’t reflect multiple expansion. Eurobank is a clear example and a rare one where we have a liquid observable market price. It’s easy to see how this keeps book value constrained but boosts operating ROE significantly. Fairfax understates book value in other creative ways too. At Fairfax India, they use high discount rates. Much like at the mothership, now that we are 11 years in, it’s hard to hold back the gains. BIAL, the most important piece has been written up for the last 2 quarters rolling the DCF forward as it is now above the Siemens purchase price. That sort of book value momentum in theory should be attractive to value investors although lately they seem happy to sell both FFH and FIH. On the insurance side, they are consistently releasing reserves because they are over reserved. Excess reserves are like deferred equity i.e. future book value. Coming off a hard market we can have especially high confidence that it’s going to show up in ROE over the next few years too. The short term reaction from the market to the Poseidon deal tells us a lot about market structure. Stock prices have either have higher correlation to revenue growth or at the very least the marginal Canadian institutional investor thinks they do. The last hard market our multiple went from 0.6x to 1.6x from October 2020 to July 2025 as institutions chased revenue growth. So far in the soft market we have retraced to 1.3x and we could certainly go lower but since the company has bought as high as 1.5x BV, my guess is during this soft market we will eventually get back to that multiple. The next hard market is when we could see significant multiple expansion. The hard part will be holding on as FFH may head to multiples we haven’t seen since the late 90s. What are we expecting to happen with the soft market? Do insurance earnings disappear or does float/premium growth just stall? I imagine the 0.6x book was because the market lost faith in Prem as an allocator, and barring something like equity hedges or another blackberry it’s hard for me to see this things trading back down below $1,000 per share. I’m having trouble keep myself from backing up the truck here, which is probably a good sign that it has lower to go.
Viking Posted March 12 Posted March 12 1 hour ago, Duke In Shadows said: two things. this is Soros not Druck: "it's not whether you are right or wrong, but how much money you make when you are right and how much you lose when you are wrong" the part on selling assets is interesting. When you listen to the CEO of Sleep Country Canada around the time Fairfax acquired the company, he explained that they chose Fairfax over several private equity firms partly because of Fairfax’s long-term approach. They were looking for a permanent home, and that played a big role in their decision. The deal flow Prem and company get over the next decade will depend on how they treat their private businesses. If they simply flip companies like Peak or Sleep Country in two years, they won’t have the same reputation as long-term owners. @Duke In Shadows, yes, I was thinking about how I wroded that part... I changed my original to: "When describing what he learned from George Soros, Stanley Druckenmiller has said that what matters in investing is not simply whether you are right or wrong, but how much money you make when you are right and how much you lose when you are wrong. Poseidon is a textbook case. Fairfax was right — and the position was large enough to matter." I can't change the actual post (I am not able to edit my longer posts - I usually get an error message when I try). When it comes to the associate/private holdings, Fairfax has been doing a very good job of working closely with management - and this includes coming up with a mutually agreed to exit strategy. Especially for the large positions. That is much easier for Fairfax when the management team is very good.
SafetyinNumbers Posted March 12 Posted March 12 24 minutes ago, Hsmpanl said: What are we expecting to happen with the soft market? Do insurance earnings disappear or does float/premium growth just stall? I imagine the 0.6x book was because the market lost faith in Prem as an allocator, and barring something like equity hedges or another blackberry it’s hard for me to see this things trading back down below $1,000 per share. I’m having trouble keep myself from backing up the truck here, which is probably a good sign that it has lower to go. In a soft market, gross premiums are slowing. The company is using less reinsurance so net premiums are growing faster. Excess capital is being used for buybacks so net premiums per share is growing even faster. Reserve releases should kick up in the next few years as prior experience justifies not releasing casualty policy related reserves too quickly from the last hard market. I think underwriting profit per share is going meaningfully higher. Of course cats could make it volatile. i think market structure is why the multiple got as low as it did and also why we recovered when we did. It will be hard for the multiple to stay down because of buybacks which is why I think we’ll move towards 1.5x BV.
nwoodman Posted March 12 Posted March 12 1 hour ago, Duke In Shadows said: the part on selling assets is interesting. When you listen to the CEO of Sleep Country Canada around the time Fairfax acquired the company, he explained that they chose Fairfax over several private equity firms partly because of Fairfax’s long-term approach. They were looking for a permanent home, and that played a big role in their decision. The deal flow Prem and company get over the next decade will depend on how they treat their private businesses. If they simply flip companies like Peak or Sleep Country in two years, they won’t have the same reputation as long-term owners. Interesting take. It strikes me ONE was the exit strategy from the start. Seaspan sits right inside the liner ecosystem and ONE was already one of the major counterparties. The consortium structure when Atlas Corp was taken private always looked like strategic operator + capital partners. FFH provides the balance sheet and stability to get the deal done, lets the business compound privately, and eventually the strategic owner consolidates. That doesn’t look like flipping assets to me, it looks more like brokering the asset into its natural long-term home while earning an appropriate return on capital along the way. The fact that many of their private positions appear to have pre-wired exit liquidity is one aspect of the Fairfax model the market seems to completely miss. A similar logic could eventually apply to Sleep Country Canada, although my sense is that if management continues to meet Fairfax’s hurdle rate they will simply keep backing them. So I think in the case of Poseidon, allowing a gradual buyout by the natural owner, only reinforces their reputation as patient capital.
Haryana Posted March 12 Posted March 12 Their Chinese shipbuilder is one of the big buyers in this. https://www.morningstar.com/news/dow-jones/2026031012011/yangzijiang-shipbuilding-plans-to-buy-10-stake-in-poseidon-for-around-826-million China-based Yangzijiang Shipbuilding plans to buy a 10% stake in the holding company that owns containership owner and operator Seaspan Corp. for $825.7 million. ... The investment is expected to enhance alignment between vessel demand and Yangzijiang's production planning and yard development strategy, Yangzijiang said. ... Yangzijiang also disclosed that Hengyuan Asset Investment, a private investment vehicle controlled by its executive chair and chief executive Ren Letian, is buying a 5.0% stake in Poseidon as part of the wider deal.
Redskin212 Posted March 12 Posted March 12 7 hours ago, nwoodman said: Interesting take. It strikes me ONE was the exit strategy from the start. Seaspan sits right inside the liner ecosystem and ONE was already one of the major counterparties. The consortium structure when Atlas Corp was taken private always looked like strategic operator + capital partners. FFH provides the balance sheet and stability to get the deal done, lets the business compound privately, and eventually the strategic owner consolidates. That doesn’t look like flipping assets to me, it looks more like brokering the asset into its natural long-term home while earning an appropriate return on capital along the way. The fact that many of their private positions appear to have pre-wired exit liquidity is one aspect of the Fairfax model the market seems to completely miss. A similar logic could eventually apply to Sleep Country Canada, although my sense is that if management continues to meet Fairfax’s hurdle rate they will simply keep backing them. So I think in the case of Poseidon, allowing a gradual buyout by the natural owner, only reinforces their reputation as patient capital. +1 And I think Sokol may have been the architect of this plan
nwoodman Posted March 12 Posted March 12 1 hour ago, Redskin212 said: +1 And I think Sokol may have been the architect of this plan That would make sense, and it lines up well with @Haryana’s post identifying the new players. Jumping around a bit, this approach also looks consistent with what we’ve seen with Foran and Eldorado. Paraphrasing @Viking here, it still amazes me that many of us can join the dots on the capital recycling/allocation portion of Fairfax’s business, yet analysts continue to treat these as discrete, almost accidental, events that surely aren’t repeatable. If this investment book sat inside a private equity wrapper, I suspect the market would value it very differently (or at least would have 12 months ago ). Its arguably even more valuable now
netcash1 Posted March 12 Posted March 12 Disruption in the fixed income market should be a win for Fairfax.
dartmonkey Posted March 12 Posted March 12 7 hours ago, nwoodman said: it still amazes me that many of us can join the dots on the capital recycling/allocation portion of Fairfax’s business, yet analysts continue to treat these as discrete, almost accidental, events that surely aren’t repeatable. Yes, it's a hard element to model. We've got (i) underwriting, (ii) dividend and interest income, (iii) the equity portfolio (I never know whether dividends from the equity holdings are included in (ii), does anyone know?), and (iv) earnings from wholly owned non-insurance subs like Recipe or Sleep Country. But what about (v) the occasional rabbit pulled out of the hat when bonds or stocks or an associate equity holding like half of Poseidon are sold? Clearly the $870m Poseidon capital gain and the $825 increase in market value of the stake that is retained are not adequately accounted for by just taking Fairfax's share of Poseidon's earning, and although the timing is not predictable, this is pretty clearly another source of income for the company. Last year, Fairfax had $3151m in pre-tax investment income, a big part of their $6440 in pre-tax income; in 2024, investment income was $1067m out of total pre-tax income of $3875m. Many people I have talked to about this investment way, "well, yes, but you can't count on having that kind of gains every year." Obviously this is true, but we already have another $866 from Poseidon and $350m from Eurolife for 2026, and the year is young. For a $37b market cap company like Fairfax, trading at less than 6x last year's pre-tax net income, if you don't model in a billion or two of gains on sale of investments every year, you get the kind of nonsensical valuation the market is giving Fairfax right now.
Duke In Shadows Posted March 12 Posted March 12 18 hours ago, nwoodman said: Interesting take. It strikes me ONE was the exit strategy from the start. Seaspan sits right inside the liner ecosystem and ONE was already one of the major counterparties. The consortium structure when Atlas Corp was taken private always looked like strategic operator + capital partners. FFH provides the balance sheet and stability to get the deal done, lets the business compound privately, and eventually the strategic owner consolidates. That doesn’t look like flipping assets to me, it looks more like brokering the asset into its natural long-term home while earning an appropriate return on capital along the way. The fact that many of their private positions appear to have pre-wired exit liquidity is one aspect of the Fairfax model the market seems to completely miss. A similar logic could eventually apply to Sleep Country Canada, although my sense is that if management continues to meet Fairfax’s hurdle rate they will simply keep backing them. So I think in the case of Poseidon, allowing a gradual buyout by the natural owner, only reinforces their reputation as patient capital. Asset flipping for sure not. a natural buyer existing does not mean the investment was structured around it. perhaps a feature.
SafetyinNumbers Posted March 13 Posted March 13 4 hours ago, dartmonkey said: the equity portfolio (I never know whether dividends from the equity holdings are included in (ii), does anyone know?) Marked to market (own<20%) dividends would be included in dividend income. When they use equity or consolidated accounting, it would must reduce the carrying/book value. 4 hours ago, dartmonkey said: way, "well, yes, but you can't count on having that kind of gains every year." Obviously this is true, That’s true over short periods but over long periods it’s not. As @Vikinghas written, extending the forecast period to three years makes the forecast easier not harder as it might be for most companies. Another way to model earnings is to use a wide range for expected returns on the equity portfolio of 5-15% annually. The other parts of the model have less variability. Since shareholder’s equity is about the same size as equity portfolio, that’s our starting point to build ROE. Then we can add 2x the coupon of the fixed income portfolio which is 5%. Now we are at 15-25% pre-tax ROE. The rest of the ROE comes from underwriting income less head office expenses including interest expense. We have a pretty good handle on the head office and interest expense. For underwriting income, analysts mostly expect the combined ratio to go up because the market is softening. This is a fundamental misunderstanding of how Fairfax’s culture. It’s in the shareholder’s best interest to book as many reserves as possible to defer tax. The shareholder has the use of more funds and uses less equity. So it shouldn’t be surprising if reserves are available in the case of a bad cat year. To that end I think a range of 91-95 is pretty fair. This is a worthwhile exercise to do to practice modeling. To that end the below RBC model is useful. The investment return is calculated as the 15-25% we calculated above divided by 3 to reflect the leverage. That’s 5-8.3%. For the tax rate, the higher the investment return, the lower the tax rate. A range of 20-25% is fair. If you do model low/high end for 2026 and calculate 2025, please share.
dartmonkey Posted March 13 Posted March 13 11 hours ago, SafetyinNumbers said: 15 hours ago, dartmonkey said: the equity portfolio (I never know whether dividends from the equity holdings are included in (ii), does anyone know?) Marked to market (own<20%) dividends would be included in dividend income. When they use equity or consolidated accounting, it would must reduce the carrying/book value. OK, that makes sense. It does make it a bit more difficult to model, though, since if you took the marked to market equity investments at market value, and then you also included dividends and interest as income, you would be double counting. I guess the solution is to NOT subtract out the marked to market equity portfolio, but to value it as a multiple of interest and dividend, for the part of earnings that are paid out, and as investment gains, which presumably reflect the retained earnings, although in a much more volatile way than if we knew exactly how much dividends came from the marked to market equity portfolio. I really like the RBC model you have posted here on multiple occasions, calculating everything as a multiple of equity, but the one thing I would like to do better is the investment yield, which they have as 7.0% (presumably for 2026?). With the 2.8x leverage they estimate, that 7.0% number is doing a lot of heavy lifting, since most of the final 25.5% in pre-tax operating ROE comes from the 19.9% investment return. It would be great to break up that 7.0% into its fixed income component which we know pretty well (I think they have forecast 5.0% for 2026) and the equity component that we need a fairly wide range for. Looking at 2024, we had $1067 investment gains on $23.0b in equity, so 4.6% (return on total equity, not return on equity investments, as everything in these calculations is based on Fairfax's equity i.e. book value), and with 0.71 investment leverage (to use the RBC term; this is in addition to the 2.00x fixed income leverage), this provided a 3.3% boost to total investment ROE, in addition to fixed income's 10.4%, for a total of 13.7%. In 2025, the equivalent numbers are $3151m investment gain on $26.3b total equity, or 12.0%, with 0.67x investment leverage, for a 8.0% contribution to the investment ROE, on top of 9.3% from fixed income, for a total of 17.3%. You posited a 5-15% range for equity investment returns, and I think that is fair; my number is much smaller just because it is as a percentage of total equity (including fixed income assets). Looking at investment returns as a percentage of equity investments, these were 6.5% in 2024 and 17.9% in 2025, within your range. My next step in doing this is to go back a few more years and see what we have in investment gains as a percentage of equity. Or maybe it makes more sense to use equity investment gains as a percentage of equity investments, and then convert that to equity investment gains as a percentage of all equity, at the last step, since the proportion of equity that is in equity investments will be changing over time. Minor technical question: do you think RBC is using previous year end equity as the denominator, or some sort of average equity, like the sum of beginning of year and end of year equity?
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now