dartmonkey

Yesterday's close was 1562.45, with an intraday high of 1575.31. Closed today at a new high, 1,571.02, with a higher intra-day/all-time high of 1580.01. Previous Close 1,562.45 Open 1,560.74 Bid 1,568.56 x 0 Ask 1,572.00 x 0 Day's Range 1,559.77 - 1,580.01 52 Week Range 932.00 - 1,580.01 With Fairfax very actively repurchasing in April, we should not be happy about this, but...

I guess this might make for a 'successful' IPO, successful for investors at least, even if selling shares to the public below what someone might have paid for them is not usually what the other shareholders would want. At least it's not too dilutive, given that it is only selling about 42m shares, out of 874m shares outstanding, or a little less than 5% of the company. But if they are selling 40% less shares, at a lower per-share price, how can they expect to raise Rs 1125 instead of 1250 crore? If they sell at the middle of their target range, Rs 265/share, they would need to sell 42.45m shares to raise Rs 1,125 crore. If that represents a 40% cut in the number of shares, that would have been 70.75m shares they previously intended to issue. If they expected to raise Rs 1,250 crore, that means they intended to sell them at just Rs 177/share, which doesn't jive with the idea that they have marked down the price to get an IPO pop. Anyways, most of these questions will be resolved in a week when we have a share price on the NSE and BSE.

Apparently, they have announced the price range, today: The company announced a price range of Rs 258-272 on Friday, May 10, 2024. Read more at: https://thearcweb.com/article/go-digit-ipo-valuation-fairfax-prem-watsa-fyeU0WSqN1fQY2G6 I don't know WHERE they announced it, but it doesn't seem to be just a rumour.

Can someone explain to me how 49% of the equity is only worth $152m (carrying value)? Or only worth $476m ('fair value', according to the Q1 report.) It clearly can't be that $2.092b worth of preferreds will only convert to an addition 19% of the equity. Explain it to me like I'm an 8-year old.

Maybe you get both equity AND fixed income exposure with Fairfax? I posted this to the Berkshire board, confusedly thinkiing I was talking to Fairfax shareholders. I'm copying it below because, while the comparison between Fairfax and Berkshire is only of moderate interest to most Berkshire shareholders, it is of great interest to Fairfax shareholders! ====== Posted to Berkshire board today: "Both Fairfax and Berkshire are constrained by regulators and by common sense in what they can do with their float, equity obviously being preferable if you have enough surplus capital to do it. Undoubtedly, Fairfax is considerably more constrained, but it isn't just a question of surplus capital,, it's also a question of how much float they have, in relation to their equity. And Fairfax has way, way more. Quoting my post here 2 days ago: With Fairfax, you have float of $33b* with $22b of equity (2023 year end numbers), whereas for Berkshire, you have float of $169b with equity of $561b. So $1 of equity is increased to $2.50* of investable assets with Fairfax, whereas with Berkshire, $1 of equity is increased to $1.23 of investable assets. Fairfax is twice as leveraged by investment float. So if you think the key to success of Berkshire was the float leverage, Fairfax is a much better setup. Thinking about this further, the above way of framing the float actually understates the difference. Perhaps a better way of looking at it is that,, for each dollar of equity invested, Berkshire invests another $0.23 of float. For each dollar of equity invested by Fairfax, you get another $1.62* of float invested. It is unsurprising that Berkshire can invest a lot more of its 23c of float in equities, compared to Fairfax. Berkshire has $568b in book value plus $169b in float, and invests $383b in equity securities and equity method investments. Fairfax has $21.615b in book value plus $35.1b* in float, and invests $15.5b in equities (mark to market, equity accounted and consolidated). So one way of looking at it that Fairfax has 15.5/21.615 = 72% of its book in equities and Berkshire has 383/568 = 67%. Yes, Fairfax has a way bigger bond portfolio, in proportion to its float, but this is just because its float is so much bigger. Fairfax actually has MORE of its book invested in equities than Berkshire, with the rest of its enormous float in fixed income because of regulatory requirements. So I am making the case that Fairfax is even more exposed to equities than Berkshire, and also gets the additional leveraged value of the much bigger fixed income portfolio. And as an investor paying 1.1x equity for Fairfax rather than 1.4x for Berkshire, this difference is further magnified. Does this make sense? Thoughts? What would Bloomstram think of this argument? *My number in the Wednesday quote was wrong, for some reason I said Fairfax had $33b in float, meaning $1.50 in float for every $1 in equity; the actual number is $35.1b in float, or $1.62 in float for every $1 in equity."

Posted 18 hours ago No one except for Bloomstran who thinks BRK’s surplus capital gives it a return advantage over FFH since it can invest more of its capital in equities which by definition have a higher return potential than fixed income. Of course, he ignores the leverage and one would think a quickly increasing surplus capital might be highly correlated to the multiple, but I digress. My question is, how much surplus capital do you think FFH could deploy into quality equities if there was a giant dislocation that created a big opportunity. Both Fairfax and Berkshire are constrained by regulators and by common sense in what they can do with their float, equity obviously being preferable if you have enough surplus capital to do it. Undoubtedly, Fairfax is considerably more constrained, but it isn't just a question of surplus capital,, it's also a question of how much float they have, in relation to their equity. And Fairfax has way, way more. Quoting my post here 2 days ago: With Fairfax, you have float of $33b* with $22b of equity (2023 year end numbers), whereas for Berkshire, you have float of $169b with equity of $561b. So $1 of equity is increased to $2.50* of investable assets with Fairfax, whereas with Berkshire, $1 of equity is increased to $1.23 of investable assets. Fairfax is twice as leveraged by investment float. So if you think the key to success of Berkshire was the float leverage, Fairfax is a much better setup. Thinking about this further, the above way of framing the float actually understates the difference. Perhaps a better way of looking at it is that,, for each dollar of equity invested, Berkshire invests another $0.23 of float. For each dollar of equity invested by Fairfax, you get another $1.62* of float invested. It is unsurprising that Berkshire can invest a lot more of its 23c of float in equities, compared to Fairfax. Berkshire has $568b in book value plus $169b in float, and invests $383b in equity securities and equity method investments. Fairfax has $21.615b in book value plus $35.1b* in float, and invests $15.5b in equities (mark to market, equity accounted and consolidated). So one way of looking at it that Fairfax has 15.5/21.615 = 72% of its book in equities and Berkshire has 383/568 = 67%. Yes, Fairfax has a way bigger bond portfolio, in proportion to its float, but this is just because its float is so much bigger. Fairfax actually has MORE of its book invested in equities than Berkshire, with the rest of its enormous float in fixed income because of regulatory requirements. So I am making the case that Fairfax is even more exposed to equities than Berkshire, and also gets the additional leveraged value of the much bigger fixed income portfolio. And as an investor paying 1.1x equity for Fairfax rather than 1.4x for Berkshire, this difference is further magnified. Does this make sense? Thoughts? What would Bloomstram think of this argument? *My number in the Wednesday quote was wrong, for some reason I said Fairfax had $33b in float, meaning $1.50 in float for every $1 in equity; the actual number is $35.1b in float, or $1.62 in float for every $1 in equity.

This is the key to the argument, I think. With Fairfax, you have float of $33b with $22b of equity (2023 year end numbers), whereas for Berkshire, you have float of $169b with equity of $561b. So $1 of equity is increased to $2.50 of investable assets with Fairfax, whereas with Berkshire, $1 of equity is increased to $1.23 of investable assets. Fairfax is twice as leveraged by investment float. So if you think the key to success of Berkshire was the float leverage, Fairfax is a much better setup. If you think the key to Berkshire's success was investment savvy (an argument that it would be hard to support from the last 20 years of performance), then it's less clear that Fairfax is a good investment. Since I favour the view that it is Berkshire's structure that has been the key to its success, at least in the last 30 years, I really like how Fairfax is positioned right now.

It is amazing to me that they don't mention, in this outlook summary, what jumps out at me in their numbers, which is the fiscal deficit. Combining federal and state deficits, this was 9.2% of GDP in 2023 and is expected to come down slightly to 8.9% in 2024 and 7.9% this year. In the USA, federal deficit was 6.3% of GDP in 2023, or $1.7T, with another $1.2T from state and local governements, so the number is comparable. Both are unsustainable, and it would seem worth mentioning that this stimulus can not go on forever. In India's case, there is some consolation from the fact that accumulated debt is not as high as the levels typical in Europe and North America. For the USA, this is 121% at the federal, with another 14% from state and local governments, for a total of 135%. In India, fortunately, debt levels are lower, 61% at the federal level and 30% at the state level, for a total of 91%. In addition, the 9% or so combined deficit is not far from the real GDP growth rate of about 8%, compared to the USA's growth rate of less than 2%, so India's real indebtedness as a percentage of GDP may at least remain at about the same level as a proportion of GDP.

Does anyone know what this ‘Argentina transaction‘ is?

gfp: The profits have been cash settled periodically the entire holding period. There is no lump sum gain at the "end." What we don't know is if these gains are taxable at all. In the United States, "profits" in an issuers own stock are not taxed. There are differences between the TRS position and a share repurchase, like taxes (FFH may pay taxes every year on the gains from the TRS, unless they get an exemption; on the other hand, there’s only a small repurchase tax up front for a share repurchase) and timing of the cash outflows. The latter is probably their reason for putting on the position and holding on to it now. But if they had the cash, I suspect they’d pay less taxes overall by doing the repurchase sooner rather than later, and using the proceeds from share price appreciation to buy back shares at much higher prices. I say it’s a wash because the higher the price goes, the more they gain from the TRS but the more they will have to pay to repurchase the shares. For instance, they put on the trade at $373/share for the equivalent of 1.96m shares . A share repurchase of 1.96m shares would have cost $732.5m. If shares hit $1372, they would have made $1.96b, yippee, but at that new share price, they would have to pay 1.96*1372=$2.69b, which is, not coincidentally, $1.96b more than the repurchase would have cost when shares were at $373. It’s like Alice running in Through the Looking Glass: as fast as the TRSs run, they can never gain ground on the share repurchase that they could have done. It’s a great trade either way, but whatever happens, it’s 1.96m shares they can retire with the returns, minus whatever tax applies.

Yes, it’s pretty unlikely they found counterparties that wanted 2 million shares’ worth of short exposure, when the shares were trading at US$373, so this explanation makes sense. In other words, FFH is long 2m shares and the banks are neutral. It’s basically just a dressed up share repurchase. Which also suggests that counting it as a long position and tracking its returns is a bit odd, since you wouldn’t do that with repurchased shares, if they had structured the transaction that way.

I see your point about raising money - if shares were at 2x BV they could issue shares rather than selling bonds at 8% or selling preferred shares to OMERS and paying them 8%. But I don’t see how the company’s share price has much to do with ROE (except marginally, by decreasing interest cost.) Say the share price rose to $2000, or 2x book, how will that help them increase book more quickly? I guess the ideal would be to get the best of both worlds: low share price for a while to repurchase shares cheaply, then a high price to issue shares profitably. I’m just stating the obvious, buy cheap and sell dear, but it would be good to get a little bit of the "sell dear" part for a while.

It's an investment fund, of course the risk is with investors, not the fund managers. If you think they will do a good job getting you a good return, then you invest and pay the fees. What seems to bother you is that the fees are not determined based on the share price, but based on the book value. It is true that this is not the norm in the investment industry, and I wonder how they justified this choice. To me, it seems more fair, since it is a value that can not be influenced by short term movements of the share price. On the other hand, it is something the manager has some control over (Muddy Waters would have something to say about that), and if the share price trails book value at the end of a given calculation period, I can see how it would seem neither fair nor friendly. I haven't been able to find any discussion about this fee structure (that is, the fact that it is based on book value), from 2014 when the structure was set up.

Yes, I think that's fine, because in the long run, it makes no difference, especially if Fairfax starts taking its performance fee in cash. It is probably true that they didn't anticipate every possible combination of book value, intrinsic value and share price, and now that book value has done well but the share price has not, taking performance fees on the basis of book value makes it seem that Fairfax has taken advantage of FIH investors, but in the long run, the weighing scale aspect of the stock market will prevail, and the share price will track book value as it did in the beginning, and the fees paid on the basis of book value will end up being about the same as if they were paid on the basis of share price. Of course, it's not much fun if you want to sell your shares before that convergence happens, but as an investor, knowing that fees were based on book value, you took that risk.

Yes, you have a point - although I myself am unlikely to add to my hugely oversized Fairfax position, the company is still repurchasing shares so I suppose we should rejoice in these misconceptions. And along the same line of thinking, there is no reason for us to wish for Fairfax to be included in the TSX 30, either. We get more shares repurchased for the same number of dollars, the longer the share price languishes.

Blackberry 46.725m shares, trading at $2.74, so it's a $128m position Micron 3.912m shares, trading at $111.85, so it's a $438m position Fairfax got all it's money back on its $500m in convertible shares, plus $200m in interest, roughly a 4% annual return in an era of low interest rates, so that is not so bad. But the huge loss on the common ($802m now worth $128m) has to be Watsa's worst ever investment, with a $674m loss. At least this quick $220m gain on the Micron position takes some of the sting out of the BB losses. Still, I would love to see that Blackberry line gone from the Dataroma list, especially since many investors think that these $1.5b in US and Canadian public company holdings accurately represents their $6.9b in total mark to market public equity or their $16.5b in total public equity, or their $92b in total assets. The unfortunate $128m Blackberry position stills looks like a big deal given that it is almost 10% of that $1.5b, but looking at it as context of the $92b in assets, it is only 0.14%, and is fading out of relevance. But the easiest way for that appearance to be corrected would be to just sell the stake and be done with it. Hopefully, now that the convertibles have been sold, Chen is gone and Watsa has left the board, that is something they may do soon.

The mark to market portion of the equity portfolio was $8.7b on Dec 31st, according to the annual report, and on March 8th, Viking estimated (posted here) that it might be worth $9.0b. The other 2 portions, associates and consolidated, were worth $7.1 and $2.8b, respectively, for a total equity portfolio of $18.9, but the earnings from these other 2 portions are already included in the estimate.

There has been some talk about Fairfax (FFH) 'doing the right thing' here by not diluting shareholders of Fairfax India (FIH) by 'taking advantage' of the fact that FIH shares are trading so far below book value, given FFH's 'fair and friendly' motto. As a shareholder of both firms, and given the fact that I own a much higher percentage of FIH than I do of FFH, I have every reason to be happy about this decision, but petec's point has not been addressed, I feel, and that, does FFH not have a fiduciary duty towards FFH shareholders to maximize what goes into FFH shareholders' pockets? I suppose legally they have some wiggle room here, and could plausibly say that it is in the best interests of FFH to preserve FIH shareholders' trust in FFH, and to maintain FIH as a viable investment vehicle for its Indian operations. After all, FFH owns half of FIH, and get a hefty fee (1.5% of book value plus 20% of annual book value gains above 5%.) Feeding the golden goose well is in the interest of making sure they keep getting all these golden eggs.

Ok, so 60.72% of INR 90438 crore would be $10.91b means this would be $6.6b if acquired with no premium. Seems like a very big fish for a minnow like Fairfax India (mkt cap $2b) to swallow, with or without the support of OMERS which would not be wanting to take on a lot of equity risk.

Ok, I see this is true for the book value gains for Odyssey and Allied stakes sold to OMERS in 2021 to raise money for buybacks (what a spectacular trade, by the way!) I'm not sure wh OMERS would really care what price they paid, if they have a side deal that Fairfax is going to repurchase the shares at the same multiple. I'm not saying this is what motivated Fairfax to do the deal; clearly, it was great to get a billion and a half of cash and repurchase Fairfax shares, at a third of today's price. But if you are Muddy Waters and you set out to see the negative side of every trade, the OMERS sales were not really sales, they were loans (an idea that has been discussed here frequently), the price makes no difference if they are just loans, and the impact of the price on Fairfax's book value is sure to be seen as book value manipulation, even if it is only icing on the cake for Fairfax. And for Gulf Insurance in 2023, the book value gain was because the purchase price for the remaining shares was applied to the existing shares: In December, Fairfax Continues its Gonzo Mode by Buying out the Portion of Gulf Insurance it Did Not Already Own at a Very Rich Multiple of ~2.4x Book Value, Taking a ~$300 Million Gain on Existing Shares https://www.muddywatersresearch.com/wp-content/uploads/2024/02/Fairfax-Financial_FFH_MW_20240208.pdf This seems to be confirmed by Fairfax's 2023 AR: Gain on sale and consolidation of insurance subsidiaries of $549.8 in 2023 principally related to the consolidation of Gulf Insurance, which required the company’s previously held equity accounted investment in Gulf Insurance to be remeasured to fair value resulting in a pre-tax gain of $279.9 https://www.fairfax.ca/wp-content/uploads/FFH_Fairfax-Financial-2023-Annual-Report.pdf

I'm not saying that anyone who sold FFH was necessarily an example of weak hands. I am saying it sounded like weak hands when one particular instance of a seller said this: With such nice gains in a very short period, and no idea of the impact of the short report, I sold. That could be an error to react quickly, because it looks like it is a good quality company. Anyway, with the proceeds I added to existing ones. I am not an expert on insurance, but it’s clear that the book value is aggressively noted, with some assets benefiting from an epic bubble in Indian equities and overvalued US real estate, as well as temporary high interest rates. It does seem that earnings are above the normal trend. The writer acknowledges that he knew little about the company, and proved this when he said that earnings seem to be above the normal trend. You, on the other hand, if your strategy involves taking up shares when they are not in demand, were probably a buyer, not a seller, when the short report came out. If so, you have done well, and it might make perfect sense to have sold back those shares with a quick gain when the share price recovered. But for me, hoping for 100-200% gains from this investment over the next 5-10 years, I would not sell because of a 10% move up or down. It is a lot easier for an investor to hang on if she knew a bit more about Fairfax, and was thus not scared off by the Muddy Waters allegations or fears about their impact.

Because of Digit? By my count, US$26.2 market cap Fairfax owns the following, in India (see p.18 of the annual letter): Fairfax India: $841m (57.6m shares) T. Cook India: $540m (300.3m shares) Quess: $309m (34.5% of the company) Other: $295m (Dec 31st 2023 fair value) Digit: $2265m (ditto) Total: $4173m (my calculation; annual letter says $4247 for Dec 31st) Fairfax has a market cap of $26b, so these Indian holdings represent about 16% of the company, assuming the same degree of undervaluation from book value for Digit that applies to other Fairfax holdings. So to prefer FFH over FIH, to get a stake in India, one would have to really love Digit, Thomas Cook and Quess as investments, as opposed to the Bangalore airport (44% of FIH's fair value holdings as of Dec 31st) and the other medium sized FIH holdings (primarily IIFL Finance, CSBBank and Sanmar Chemicals, which account for another 31%.) And even so, Fairfax Financial only gets you about a 16% exposure to these. I would gladly reduce my somewhat oversized Fairfax India exposure and buy more Fairfax Financial shares (or reestablish my Berkshire position), but for the moment, I think Fairfax India is even more undervalued than Fairfax Financial and gives me a lot more exposure to India's exciting potential. Of course with FIH you get a not insignificant slippage, 1.5% plus 20% of the annual book value growth above 5%. But unless that is a deal breaker, if you really want India, I think FIH fits the bill better.

Yes, sorry for the sarcasm. I don't really believe that Fairfax is primarily motivated by the need to artificially boost its reported book value. On the other hand, when you sell at a high price with a guarantee that you can buy it back at the same high price, with an annual fee, it's not really a sale, it's a loan. But since it is structured as a sale, you can (or perhaps, must) revalue the book value of the shares you still own, which does boost the book value and maybe has some advantages for Fairfax vis-à-vis regulators. Unfortunately, it also opens you up to criticism that you did the deal JUST to boost your book value.

A special kind of pawnshop, that gives you money for your mother's jewelry that you really don't want them to sell, because you want to come back and get it in a couple of years when your finances are better. So you pay them an annual 10% fee to keep the jewelry in a special safe. They can also sell to someone else it if you don't come back to repurchase it within a few years. Muddy Waters would add that there is another side benefit of this deal: you can 'sell' this jewelry to the pawn shop for an inflated price, because it allows you to claim that the rest of the jewelry you own is worth the same amount, reassuring other creditors. So it is a disguised loan, with a 10% interest rate, which can double as a book value adjustment, if needed. Of course, I don't believe this...

Can anyone explain this in plain English? It sounds like FFH has calls allowing them to buy out the minority interests in 2026, 2027 and 2029, but then the minority interests can dump their stakes back to FFH in a variety of was once those call options have expired. Is that right? And I presume there is some formula that says what price FFH has to pay, do we know anything about that?