dartmonkey

Thanks. Additional question: what this means is that retrospectively, the losses for the year in which the reserve was taken ended up being less than expected, so the combined ratio reported in that previous year (or several previous years) was higher than it would have been, had they not taken so much reserve. Of course the reserves are just a guess, but once they have the benefit of a longer period to assess losses, do they go back and reassess past years’ combined ratios, or does this year’s favourable adjustment make current year’s combined ratio look better?

The calculation is actually even easier than that: 16%= $160m means 100% = $1000m, no need to multiply 160 by 6 and round.

i see now, sorry to be slow: 44.4% p.a. for FaIrfax, 44 and 40% over 4.25y for the indexes. Probably all these numbers are without considering dividends, right? This wouldn’t change much, as the dividend rates of each are similar, 1.4% for the S&P and about 1% for FFH.

What does this mean? Total return for Fairfax or for the indexes should include dividends. Are you just saying Fairfax’s return is better than either index, even ignoring Fairfax’s dividends? At 41% p.a. (in USD), it is WAY ahead of either index (8.9% and 8.2%), but I don’t know if the latter include dividends or not.

Yeah, roughly. While I think 10% p.a. would be acceptable, I think Fairfax is likely to do a fair bit better, unless the market really plunges, and then the RELATIVE returns will be excellent and no one will worry about hitting 10% or not. But the more likely scenario, if there is a soft insurance market, is that Fairfax doesn’t fight for market share and lets its float sink a bit, instead of marching up year after year as it has in recent years. If they get 3% of book on underwriting, and 3% on the bond portfolio (after a few years, when the current higher rated bonds have to be rolled over), and say 5% on the equity holdings, that makes about 3%+2*3%+5%=14%, since they have about one times their equity in insurance premiums, 2*equity in bonds, and 1*equity in socks, so 14% on equity, so at price=1.4*book they might hit 10% return on their current price. Add a smidgeon for opportunistic sales like pet insurance, Stelco, etc, and maybe we get 12%, pre-tax, or 10% after tax. Hard to find something else with a fairly reliable 10% earnings yield even through a soft insurance market and lower bond yields, two things that don’t usually go together anyways.

I am greedy enough to happily accept a 76% premium, despite also agreeing that it is probably worth a bit more. It would take my CAGR to 12% over 10 years, not brilliant, but a hell of a lot better than the current 5% (or maybe a little higher, since I bought a fair number of shares at $12-14 in the last 2 years). But a bird in hand, and all that. And I couldn't care less whether I get cash or FFH shares; getting cash instead of shares just means I have to spend about 30 seconds more and pay less than $10 in commissions, each of which I can live with. And given my high concentration in both FFH and FIH, I would probably also by a little bit of something else that's down more, like GOOG and IBKR.

This sounds reassuring for 2025 because they are hedged but maybe not for subsequent years.

Do we know what procedure they have offered to their employees who are 50 and over? Screening for risk factors and offering target therapies to reduce risk factors, to people identified as being at high risk, is standard medical practice. Fairfax is probably doing more than this, perhaps testing Lp(a), or doing a scan for coronary artery calcification, or doing an exercise stress test on a treadmill. These 3 screening techniques may not be cost effective, but it is hard to see how they could lead to worse outcomes, and it is quite plausible that any of them, or all 3, would lead to better outcomes, even if they are not recommended as cost effective by organizations like the USPSTF.

Speaking of Poseidon, p. 16 has an oddly titled table called 'Common stock holdings as of December 31, 2024'. In that table there are companies who have common stocks that are marked to market like Orla Mining and Occidental Petroleum, and associates (20-50% ownership) that are equity accounted, like Eurobank and Quess. The table gives values for what they call 'Carrying value per share' and also 'Share price', often somewhat higher for the associates. But there are also companies that, as far as I know, are not common stocks, like Poseidon. For instance, for Poseidon, they give a carrying value per share of $14.08, with a share price of $15.50. But where are they getting this share price?

That means they weren't added to the TSX 60. They announce all the changes in the same press release, so no news about the TSX 60 means there were no changes. It would look like this:

Right, we discussed this a few days ago. Seems like the answer is, indefinitely! Fairfax made about $2.4b in net earnings APART from investment gains last year (underwriting, interest income, share of earnings of associates), so it is at about 13 times earnings even without its investment gains (or 9 times earnings if you include the investment gains. On the thesis that they will never pick good equity investments or make any money from them except by luck, they're still cheap, half of Intact's multiple. And Fairfax's might occasionally make a gain on investment too...

Then it's not too late - at $1420 the price is a little lower than the average price over the last 2 weeks. There's no discernable influence from the timing of the possible index add. I bought a few more shares, based on what seems like a reasonable possibility that they will be worth 5-10% more next week.

His basic idea is that Fairfax invests in high risk equity bets that sometimes work and often don't work. Recent bets have worked, but going forward, investors are at risk of having another bad 10-year run like the one we lived through starting about 15 years ago. Call it the long Blackberry/short tech era. It's not a crazy idea, or it wouldn't be, if Fairfax had not in the meantime developed excellent underwriting, expanded premium volume from savvy acquisitions and repurchases from financing partners like OMERS, huge streams of fixed income earnings locked in for several years, and 2 great insurance startups in Digit and Ki. And on the equity investment side, if there had been just one or two high-risk bets like Stelco and Resolute that worked out, you might worry about the others, but with Eurobank, the Fairfax swaps, Orla Mining, the Bangalore airport, Thomas Cook India, etc. etc., it's looking more like Farifax has the Midas touch than just beginner's luck. It's good for Fairfax's repurchases, though, the longer this Fairfax=Blackberry pessimism persists, and it's good for entertainment value! I can just imagine how Horn feels, after every one of his justifications for a low share price keep falling, one after the other. Maybe the next one will be the index inclusion. Or the Ki or the Bangalore Airport IPO? Or higher income from corporate bonds, if we get a market swoon?

It certainly has been a wonderful investment, after a long slow start, but a 10-y CAGR of about 11% is great when you're investing other people's money (float)! I think the market has been slow to recognize all the good things happening at Eurobank, and I'm amazed the stock is not up more, after the stellar earnings announcement. They are now trading at 6.6x last year's earnings, when we know that the big acquistion of another 37.5% of Hellenic Bank in November, and, soon, the remaining 6.5%, is going to boost 2025 earnings substantially, apart from the underlying growth in Eurobank itself. My one quibble is that you can't count the $129m future dividend, which is already baked into the present share price. This is not material to your estimate, but I believe the total return should be $2342-$129=$2213. Since the July 2024 dividend and the forced January 2025 share sale are both recent, the timing of these returns is not material but ideally, the total return should be adjusted for the timing of the different components of the return stream, and this will become material as Eurobank continues to issue dividends over the coming years.

Exactly. BV itself is a very poor indication of intrinsic value, and the ratio of IV:BV will be quite different from one company to another. I would argue that FFH merits a higher ratio than Berkshire, just because it has such a large float available for investment (about twice equity) in relation to Berkshire where that float is a small fraction (about a quarter of equity.) But in both cases, the INCREASE in BV from year to year, over the long-term, is a good indication of how quickly IV is increasing. The following is a bit dated, I'll update it when we get float numbers from Berkshire and Fairfax in a few weeks, but it hasn't changed much and illustrates the point that Fairfax can do a lot more investment with a given amount of book value: Berkshire float, year end 2022: $164b; market cap $859b; float leverage 19%; Fairfax float, year end 2022: $31.2b; market cap $23.8b; float leverage 131%.

Great, thanks. Seems like information that should be on their web page but the like to keep it bare bones.

Quibble: this would be a 4-bagger, not a 3-bagger. Starting Dec 31st 2020 at $340.94, when it gets to $3409.40, it will be a 10-bagger, i.e an increase of 900%. Factoring in dividends: according to Yahoo Finance, FRFHF was at an adjusted price of $314.96 at the end of 2020, so it is now at 471% of its price just over 4 years ago, almost a 5-bagger.

John Chen took over in 2013 and I thought the hypothesis was that the patent portfolio was worth a fortune. Ten years later, I doubt it. I guess having Blackberry in that 13-F investment profile probably helps us keep the share price low and maximizes the value of share repurchases. But I would still be happy to see it gone, even if the repurchase window closes, and maybe now that Watsa is off the Blackberry board (Feb 15 2024) it can happen this time...

I hate to bring up this painful subject, but... What the hell is happening at Blackberry? Recap from last year's AR: We got our money back on our convertible ($167 million in 2020, $183 million in 2023 and $150 million in 2024) plus cumulative interest income of approximately $200 million. Our common stock position as of 2023 ($162 million or 8% of the company) which was acquired at a cost of $17.16 per share was valued on our balance sheet at $3.54 per share. Another horrendous investment by your Chairman. At $3.54 at the end of 2023, this would mean they owned 45.8m shares, which is what they reported on p.15 of the report. Dataroma says they have 46.7m, maybe because Watsa owns some personally, but let's take that 45.8m, assuming they have neither bought nor sold any shares. A year later, the share price was much the same, $3.68, after trading as low as $2.01 during the year 2024. But in the last 6 weeks, the price has soared, to $6.20 right now, or up 68% year to date, in a flat market. Ok, that's only a gain of $115m, which is small fry for Fairfax nowadays, but still, it's something. And I can't see any news that would justify this: Blackberry just sold off their cybersecurity business (Cylance) at a huge loss, and are focusing on Internet of Things (IoT), whose revenues are growing a bit (13% last year, to all of $62m...), and now they are trying to give away the software (QNX) to engineers in India and elsewhere in the hope that it will become a standard. Blackberry's current market cap is $3.7b, hard to fathom. Is there speculation they might be acquired? As a long-suffering Fairfax shareholder, I can only dream...

I hate to bring up this painful subject, but... What the hell is happening at Blackberry? Recap from last year's AR: We got our money back on our convertible ($167 million in 2020, $183 million in 2023 and $150 million in 2024) plus cumulative interest income of approximately $200 million. Our common stock position as of 2023 ($162 million or 8% of the company) which was acquired at a cost of $17.16 per share was valued on our balance sheet at $3.54 per share. Another horrendous investment by your Chairman. At $3.54 at the end of 2023, this would mean they owned 45.8m shares, which is what they reported on p.15 of the report. Dataroma says they have 46.7m, maybe because Watsa owns some personally, but let's take that 45.8m, assuming they have neither bought nor sold any shares. A year later, the share price was much the same, $3.68, after trading as low as $2.01 during the year 2024. But in the last 6 weeks, the price has soared, to $6.20 right now, or up 68% year to date, in a flat market. Ok, that's only a gain of $115m, which is small fry for Fairfax nowadays, but still, it's something. And I can't see any news that would justify this: Blackberry just sold off their cybersecurity business (Cylance) at a huge loss, and are focusing on Internet of Things (IoT), whose revenues are growing a bit (13% last year, to all of $62m...), and now they are trying to give away the software (QNX) to engineers in India and elsewhere in the hope that it will become a standard. Blackberry's current market cap is $3.7b, hard to fathom. Is there speculation they might be acquired? As a long-suffering Fairfax shareholder, I can only dream...

I hate to bring up this painful subject, but... What the hell is happening at Blackberry? Recap from last year's AR: We got our money back on our convertible ($167 million in 2020, $183 million in 2023 and $150 million in 2024) plus cumulative interest income of approximately $200 million. Our common stock position as of 2023 ($162 million or 8% of the company) which was acquired at a cost of $17.16 per share was valued on our balance sheet at $3.54 per share. Another horrendous investment by your Chairman. At $3.54 at the end of 2023, this would mean they owned 45.8m shares, which is what they reported on p.15 of the report. Dataroma says they have 46.7m, maybe because Watsa owns some personally, but let's take that 45.8m, assuming they have neither bought nor sold any shares. A year later, the share price was much the same, $3.68, after trading as low as $2.01 during the year 2024. But in the last 6 weeks, the price has soared, to $6.20 right now, or up 68% year to date, in a flat market. Ok, that's only a gain of $115m, which is small fry for Fairfax nowadays, but still, it's something. And I can't see any news that would justify this: Blackberry just sold off their cybersecurity business (Cylance) at a huge loss, and are focusing on Internet of Things (IoT), whose revenues are growing a bit (13% last year, to all of $62m...), and now they are trying to give away the software (QNX) to engineers in India and elsewhere in the hope that it will become a standard. Blackberry's current market cap is $3.7b, hard to fathom. Is there speculation they might be acquired? As a long-suffering Fairfax shareholder, I can only dream...

When is the annual report released? I see that last year, they published 2023 Q4 results on Feb 15, and there is also the letter that accompanied the annual report, with the letter dated March 8 but with no mention in the press releases of when it was actually released. This year, we got Q4 results on Feb 13, and there is also no announcement about when the letter might come out, unless I'm looking in the wrong place. Has the company ever announced when the annual report will be released?

I guess all we know is that they bought it in Q4, so somewhere between $42 and $67, and it is now at $66, so it could be about breakeven if they bought it at the beginning of the quarter, or it could be up as much as $65m if they happened to buy it in December when it was at its lowest. Looking at other mark to market positions like Blackberry, we are not told anything about the cost basis, the way Berkshire used to report its big positiions, so I guess we will probably never know. Still, it seems like a safe thing to own, at 18 times pretty steady earnings. 4.8% dividend yield, buys back a lot of shares, industry that's not going away...

We tend to have a negative view of employee stock grants or stock option grants because some tech companies give them out like candy, cause a lot of dilution to their existing shareholders, and then don't want to account for them as an expense. Buffett has made fun of this in the past, for instance, and said: “Stock-based compensation is the most egregious example,” Buffett said. “The very name says it all: ‘compensation.’ If compensation isn’t an expense, what is it? And, if real and recurring expenses don’t belong in the calculation of earnings, where in the world do they belong?" But a program that wants employees to purchase stock, and gives them part of their compensation to encourage them to do this, and repurchases those shares as they are granted, and accounts for all this appropriately by calling this a compensation expense which reduces earnings, is really doing something completely sensible and, probably, positive for the long-term culture of the company and retention of its employees. It would be good to review exactly how Fairfax is doing this, and it may not be exactly the same in each of the companies it owns (following the principle of allowing managers some discretion in how they do things), but from what I can tell, Fairfax seems to be doing this right.

The price broke through CAD$2000 for the first time in Dec 2024 and touched a high price of $2071.49 on Dec 6, before dropping back below $2000 and finishing 2024 at exactly $2000 (or $1978, adjusted for the dividend paid in January.) The highest price today (which was $2075.81 when youu posted, and $2089.26 now, but the day is young) would be an all-time high as far as I can tell, in absolute terms, but even more so if you adjust for the CAD$21.50 dividend.