Hamburg Investor

Where would you look for opportunities, if FFH would get to expensive? Me personally I’d look at MKL, BRK, META, BN, DHR, Protektor Forsikring. JOE. Meta is the only tech stock I have; for me it’s less of a tech idea and more like a network or infrastructure perspective. I look at it and recognise similarities to the newspaper (Washington Post) investment(s) of Buffett.

Fingers crossed, that FFH will not go to 3.0 pb - and if so, that you (and me) find better opportunities. At present valuations I’d have to go to less high conviction opportunities. Maybe I’d go there, maybe not. I imagine taxes being a much bigger threat for me in Germany than for you, so I am just more focussed on longterm and less trading.

Sorry, but although I agree with some of general Ausführungen, ich sehe eben Deine Ausführungen nicht angewendet: High multiple?: I am not assuming a ROE of 20% over 20 years – that was @TwoCitiesCapital. I am assuming much lower ROEs. Longterm? Regarding your view on "longterm": You seem to imply that long-term predictions about returns are generally not possible. At least, that's how I understand your statement: "You have to be right in your assessment for 10+ years with a high multiple. It's hard enough to be right for 5 years." My view would be: If I couldn't find longterm winners, I would invest into an etf. That's how I e. g. see software companies, etc. In my view, that's exactly why Buffett didn't invest in tech. His comment that tech was outside his circle of competence was, in my view, always an attempt not to offend anyone who thought tech was within his/her circle of competence. There may be very, very few people who really have a tech circle of competence; in my view, many of today's winners who bet on tech were simply in the right place at the right time on the stock market. There are few tech investors who have consistently found the right tech investments over 30, 40 or 50 years. But there are examples of successful investors in other industries. High multiple again: Where can you find quality companies with a PE ratio of 22 and a high long-term (20+ years) ROE of 15% (i.e. a PB ratio of 3; and I am, of course, always talking about the proceeds compounding at 15% as well)? If "tech stocks" is your answer, then sorry, I don't follow you. Tech is disruptive per se; a fact that Buffett and Munger both saw very clearly. Tech disrupts tech. Nobody knows with any degree of certainty what AI, quantum computers and all the other things we are not even aware of coming down the road between 2021 and 2030 will do to the moats of the Mag 7. IBM, Konica, Fuji, Kodak, Nokia, Blackberry – people were so sure about the big long-term moats of all of them. They were wrong. This risk of moat disruption is way higher for most other stocks when compared to FFH, BRK and MKL. BRK obviously has the lowest risk of the three, but also has the lowest return expectations going forward. Anyway: An example of a high conviction 15% compounder at a PE ratio of 22 would be great. I don't see it. The math: If you do the math, you will notice: A very long period of outperformance with just a few percentage points above the average of 10% per annum is a very good way to beat the market. At least it worked for a lot of investors buying BRK in the 1980ies or 1990ies or 2000s and I read a lot of people complaining they haven't bought back than. Which makes me think, that a lot of us are happy with a CAGR above the market of 5% or even a bit less ov er the longterm. And with that I am not saying, FFH being the next Berkshire. Obviously, BRK, MKL and FFH have been good opportunities to achieve precisely that in the past. The question is whether this is considered a coincidence or a result of their internal structure and similarities. If you add the three together, you get an average CAGR of 17% to 19% over 140 years on average. The three will probably achieve less than that looking forward, as they are all larger than they were in the 1990s and that's a drag to returns. If that outperformance in general is coincidence for you (which is of course totally fine; everybody should follower his/her paths)), I totally understand, that you think, outperformance of the three isn't predictable and than PB Ratio 3 is way too expensive. But than we just have different assumptions. Others thought, that ROEs above 30% for software companies would hold forever; that's where I differ. I first bought FFH in 2013 or so and gradually expanded that. Of course, you had to hold on to the company for a long time before it paid off. Of course, the management mistakes were a big burden. But still, I always assumed that they would fix it at some point. And there were other things (soft market, low bond yields, growth beating value), that were another drag to returns. We don't have to agree, but I consider FFH to be an outperformer over a very long period of time and a (lumpy) compounder. For that, a PE ratio of 22 is not perfect, but okay.

Thank you for the post. Just some questions, that come to my mind and maybe you could help improve my learning curve: 1. ROE 20% for 20 years for FFH - how do you get there? If a company is priced for a 20% ROE over 20 years, than just a small deviation brings big problems to the outcome. Couldn't agree more. You need to have a big moat for getting such a high ROE over such a long timeframe. But is 20% ROE over the next 20 years really your (or anybody's?) base assumption for FFH as of today? I don't think so; would be interesting to learn how/why you get to that. 2. If ROE20% over 20 years: Why expensive with PB Ratio 3 / PE Ratio 15? Anyway, let's stay with your assumption to come to my second question: ROE20% over 20 years - that's around a 40 bagger of equity over 20 years, isn't it? I'll leave out dividends and taxes on them, etc. for now; let's assume that dividends are reinvested in FFH without taxes (I'm only interested in the rough dimensions anyway, so I prefer to do rough calculations in my head and may be off a bit; but I think it doesn't change my following argument a lot). Okay, so staying within the framework of your assumptions, a pb ratio of 3 and a ROE of 20%, that's a pe ratio of 15, right? So my second question would be: If you're making the case for any ROE20 company, why not buy that at a PE of 15? I mean, that'll be really cheap in my eyes. Even assuming ROE20 for most years and not all, the valuation would still be really okay. I mean, Munger made the case, that over long periods of time your returns will bei close to ROE, and if ROE is high, it's hard to overpay; and I'd add, that pb ratio of 3 wouldn't be high in my eyes. Maybe you could add some color to that?! 3. My base case for FFH for comparison: Way lower than yours - but with a loooonger view Now here's my assumption; would be keen to learn what you think about it. In a nutshell: ROE of 18%+ over the next 5 years ROE of 15%+ over the next 20 years (that includes the first 5 years, doesn't it? So even ROE13% or ROE 14% for year 6 to 20 would bring us to ROE15% for the 20 years, if 18%+ happens in the first 5 years, right? after that ROE of 12% to 15% for another 10 to 25 years and after that "some outperformance in comparison to the average S&P500 stock" (yes, I know, FFH isn't part of that). Maybe ROE 12% or 13% against 10% or 11% longterm ROE for the average US business? Doesn't seem very ambitious in my eyes. (it would be helpful if the average price of FFH would stay below intrinsic value, as buybacks could be done as a. that would push ROE immediately and b. that would help to bring an ever growing cash flow to work (look what happens to BRK since one and a half decades...) at low risk and c. that would help FFH to stay smaller for longer, which would help ROE to stay higher (again BRK: the investment universe just shrank so much... It just became so big) 4. Why should FFH outperform forever (at least a bit)? Where‘s FFHs enduring moat? My general assumption is, that BRK, MKL, FFH share similarities that should help them to outperform the market over many, many decades. Not in each decade, but on average (low yields, growth outperforming value, soft insurance markets, bad capital allocation haven't helped FFH in the 2010 years; but my base assumption clearly isn't that with a forever view). What are the similarities, where do I see the enduring moat? - The decentralised structure, - (better than) cost free float leverage, - leaders clearly being "value investors from Graham and Doddsville", how Buffett would call them - the will to invest a bigger portion of equity into businesses (in contrast to Bond investing) than average insurers, - culture, - many different cash flows and the possibility of reinvesting proceeds from weaker-growing businesses in better-growing ones Those are some important basic similarities. Putting all that together, that’s an enduring moat. At least it has helped the 3 companies to reach a balanced average yield of somewhere between 17% and 19% I guess over 140 years (BRK: 60 years and the other two each 40 years). With that view I don't see why to sell FFH even at pb ratio of 3 tomorrow (so that would be a pe ratio of 22 or so for a business with ROE 15%; that's an okay valuation in my eyes, as I am pretty sure FFH being an outperformer over the long run AND I'd have to pay a lot of taxes (around 30% on profits and I have a lot of those...), when selling (and than hoping the price to erode and buy back cheaper; and than that doesn't happen; that's a typical error, happening all too often). The S&P500 is valued at a pe ratio of 30 AND that average S&P500 company will clearly underperform 15% ROE; I think 20 years compounding with 15% CAGR defines the best 5% or so of all S&P500 companies). So in a nutshell I don't see ROE20% over 20 years for FFH as may base case assumption - but outperformance against the market over the very looooong run. My view: You can buy a business at a high valuation either on the assumption of a very high ROE over a short timeframe and after that assummng falling back to average (okay your case with 20 years isn't short; but a 50+ years view is longer... ) or if you have a high conviction in a lower outperformance over a longer timeframe ("forever" being the north star). I am clearly not aiming for 20%+ returns over decades after tax with that; but I want very high conviction for an outperformance and a rock solid investment over time (so don’t translate „rock solid“ with „unlumpy returns“). Even bought at 3 pb ratio I think CAGR 15% (OR an outperformance of 5% to the market) being possible over 2+ decades, but not for sure. Depends on the 15% „+“ above, which I see over 2 decades and on the question, if after that a CAGR of 12% or 15% kicks in. And with that CAGR thinking I am clearly not reflecting to Mr. Market but to development of intrinsic value over time. And don’t get me wrong: After all I am much more happy with a pb ratio of 1 or below over the next decade or so than pb rstio of 3 or above; the lower the shareprice, the more buybacks, the higher the ROE, the smaller FFH stays for longer, which helps FFH management moving the needle with smaller investments… All that’s better for longterm growth of intrinsic value and we only get that push, if price stays below intrinsic value. You yourself made the case for Constellation Software; I haven’t analyzed it, as the software only focus of the business is just too narrow for me. Would be happy to learn, what you think about my assumptions and approach and reasoning about why pb ratio of 3 tomorrow would be okay.

+1 When starting to write my focus was more on margin of safety than on returns; but you‘re right; it’s about both. Just another add: I think the more layers a company has, the more value investors are biased to put a margin of safety on each layer they find within a conpany („this wholly owned business could earn between x and y; let’s be conservative and… If we now sum up the earnings streams of all wholly owned business it‘s a bit above xyz, let’s be conservative and… Which multiple should we now apply to all income streams of the wholly owned businesses, having a wide range of ROEs? Let’s be conservative and… ). But that‘s wrong! For someone with a hammer everything looks like a nail. For some value investors every layer they find is given a Margin of safety. If you‘re applying margin of safety like that, you are not focussing margin of safety on risk, but you are biased away from holdings and to investments with less (say: one) income stream; while it should be just the other way around (all other things equal). It’s all about risk. If there’s more risk, the margin of safety should be bigger, and if there’s less, it should be smaller - that’s the point which should define the magnitude of the margin of safety.

+1 And isn't it like that: All things being equal any company with multiple cash streams is less risky than one with one income stream. So two companies each with ROE 15% (same balance sheet strength, same culture, same ...), one having one business, one with multiple (and each business having the exact same strength of moat), than the one with multiple income streams is less risky. Why? Because the risk of failure at the company having only one cash stream is higher. If it looses its one cash stream, it hardly can invest into other businesses. Berkshire Hathaway itself being a very rare exception from the rule, as it transferred from a textile business to an insurance business with relatively high equity investments; but e. g. Kodak, Polaroid, IBM, BlackBerry, Nokia, Sears, newspapers in general etc. weren't succeeding. To a lesser degree even a company with multiple cash streams, each having a slightly lower moat, is much more able to survive disruptions over time than one cash stream business; the return of the more diversified cash stream in many cases will be higher on average over long periods of time (as the one income stream business fails more often on average). So from a mathematical standpoint, you are more happy having 10 cash streams with a moat of 80% perfection (all else, e. g. ROE, being equal), than one stream with 90% perfection. And that's true, even if you were not able to shift cash from one business to another. But it applies the more if you are having a decentralized structure and are able to invest the proceeds of the dying (or not-so-good-delivering) business into other income streams, that are easily accessible to management. In reality moats are volatile over decades and returns in any business go up and down. So having the possibility to invest proceeds of a business returning only ROE 10% at a time into another, which at a time brings 18%, is a structural advantage, and it's good for survival chances. Per definition tech is a sector, which is driven more than any other sector by progress and disruption. Nokia had a big moat, just as Blackberry had or Kodak. It is no coincidence that Coca Cola, Wells Fargo, Anheuser-Busch, Hershey and Altria are among the most successful long-term stocks, if you're going back over 100 years. There are virtually no technology stocks within that century category. Technology as a sector stays, but the businesses within that sector change more often. Even though technology changed the world more than any other sector, tech companies haven't given the best returns. they disrupt each other more often. A lot of people think, that Buffett and Munger were 100% making the case for concentrated bets and were against diversification. But that is wrong or at least greatly shortened and only half the story. The best proof of this is Berkshire Hathaway itself: It is built on dozens of independent cash streams. This kind of structural setup – many cash streams, decentralised reinvestment of proceeds – has a lot to do with safety (and with returns). It's not the perfect setup, but it's very very good and has a bigger chance of lasting very, very long. Failure just doesn't kill you so quickly. Munger: "You don’t have to be brilliant, only a little bit wiser than the other guys, on average, for a long, long time."

I I know; it’s just, that you wrote yourself before, that you excluded the TRS; now you write the opposite. Anyway, the logic you apply is just as good; you‘d just have to adjust for the cost of the TRS (I don’t know, what they pay for that, do you? Always wanted tobask that question, as I am not so used to TRS) and taxes, I guess. Might be a little bit more work as if FFH would just have bought back 8% more of its stock than holding the TRS. Regarding selling equities in general (and high cash accumulation from earnings; that might come within the next years; FFH begins to get a cash machine…): I think it’s good to consider capital allocation within look through earnings. If you believe that management will reallocate cash to attractive returns in the short or medium term, you can assume higher future look-through earnings in a scenario; if not, remain conservative. At least, I did that and found it helpful in other contextes; nothing for FFH for now I think, but wouldn’t surprise me, if cash on hand would get more volatile within the next years.

+1 That way of thinking resonates a lot with me. It's all about intrinsic value; the rest will take of itself. Just some random thoughts regarding what yo wrote: 1. Regarding your point within the downsides "lower returns from equities": Low share price helps covering that! Yes, that can happen. One there's Mr. Market and two management has to find good investments; and they can fail. All things equal, investing into equity is a risk; Prem could be wrong, as he has been (I don't blame him for that; even Buffett has been wrong multiple times). But (and that's my add): As long as the share price is below IV buying back shares is a high return, low risk alternative. And management uses it heavily. So a share price below IV reduces the risk of being wrong AND gives an extra push to returns. So each dollar flowing into buybacks helps keeping the course. Let's hope for low stock price! 2. Multiple 15? Than IV $3.240, not $3.000 If $200 is your expected earnings you get to $3.000 with a multiple of 15. But when you want to get to IV, shouldn't you add the TRS. If TRS are 8% of FFH (is it 8% Not sure), than you could add another 8% of $3.000 of that. So with P/e 15, FFH IV would be $3.240. I don't say you'r right or wrong; it's just a "If you think FFH earns $200, than isn't it $3.240?" You haven't added the gains; so maybe it's even higher... 3. Is 15 a good multiple for FFHs Earnings? I don't know. What I can say: A multiple of 15 on look through earnings has been a steal for BRK, FFH, MKL, when you held them a decade or two most of the time. It's obvious, but I think we are talking rarely about it: If a company is able to deliver ROE of 15% (or even more) over a loooong time, than you can pay (absurdly) high p/b ratios and still get a great return. Imagine you could have bought BRK in 1970, 1980, 1990, 2000 at such a price or FFH/MKL 1990, 2000. The years with ultra low bond rates and growth (vs. value) winning and soft markets have destroyed that pattern a bit. But maybe that time was an anomaly? Still the 3 are all bigger today etc., so I am not saying ROE of 25% or more should come again. But still. The main question for me is therefore: Will FFH (and MKL) be able to deliver above market average ROEs of 15% over long time frames (decades)? If yes, than P/E 15 (so around 2.0 pb ratio) isn't high, as long, as the company is able to reinvest with that same 15%. Using the rule of 72, ROE 15% means a double in equity every 5 years ... and the equity 16-folds in 20 years. If you buy at pb 2.0 and the multiple get cuts in half at the end (so p/b ratio of 1, p/e 7.5 for an investment with ROE 15%, you still have an 8-fold return and above 11% CAGR in stock price. That's okay as a return in a pessimistic scenario. Could be valued with pb 2 or even pb4 (p/e ratio30) after 20 years; than you'd have a 16 bagger (CAGR 15%) or 32 bagger (CAGR 18%). And your return would get better, if FFH would be able to buy back stock, or would deliver ROE 16%, 17% ... or if it would be able to deliver that 15% over 20+ years. That's al upside scenarios.

+1 And please let me add (from Buffett): „We don’t buy and sell stocks based upon what other people think the stock market is going to do (I never have an opinion) but rather upon what we think the company is going to do.“

I agree that this is unlikely to happen again. However, isn't it better for FFH's earnings over the next 5 years, if Treasury Yields would just stay were they are today? Than FFH would have a full 5 years of 4% returns (or: an average yield of 4%) – instead of 1 year of around 1.5%, another year in which the US10 rose from a yield of 1.5% to 3.5%, and 3 years with an average of 4%; that's what lies behind us? We are not looking back on five years with an average yield of 4%, don't we? I know FFH had and has not only US10 but a lot of shorter running bonds, so maybe US10 ist not the best comparison; but you get my point. It's not important, if Treasury Yields go up; it's important, that they are (somewhat) high on average. And FFH had short maturities (three months? Six months? I don't remember...), but even those cost something when things went uphill so quickly; and I don't see that happening in the next five years (although anything can happen, as always), so that headwind might not show up again. The increase helped FFH to leave the insurance market behind, but a consistent 4% bond yield over the next five years would actually give FFH a higher average return than it has had in the past 5 years imho, not a lower one. Personally, I don't find this argument very convincing as an evidence that FFH's float would be less worth looking through the windshield than through the rearview mirror. Or do you mean bond yields will be worse over the next five years than they have been in the past? I've been reading for three to four years now that yields will fall significantly again. But that's not happening, and I don't know anyone who could predict it certainly. In any case, 4% doesn't seem implausible to me over the next five years. In any case, it seems more plausible to me than 2% on average or 6%. Maybe the average yield of the past 5 years was between 3.0% and 3.5%; so even if that would be again the average of the next 5 years, that wouldn't be a "new" headwind to FFHs. I also think that your two points could be related; isn't it precisely because bond yields rose that FFH was revalued higher over the last five years. Not only because of that, but also because of that?

What if you could win a Fairfax stock within this context? You could either deny to bet (no win) or you could win it by being on the right side of over/under?

I think you're changing the perspective away from my "price versus intrinsic value" approach (which is fine), but if you do that, you should value buybacks and other opportunities by the same yardstick - AND you should value the full picture. CAGR: You forget the reinvestment opportunities here. Yes, 10% is the "static" return of FFH at 1.5 P/B Ratio with 15% ROE. But that is only the yield you get, when buying. But this 10% can grow at around 15% per year over long periods, as long as the business can keep earning 15% on reinvested capital. So the CAGR of this investment into FFH (or any other stock with P/B Ratio 1.5 and ROE 15%) is not 10% - it's 15%, assuming (a) 1.5 P/B is fair value and (b) the market applies the same multiple at the beginning and at the end. You can apply the same logic to higher‑ROE businesses like MSFT. Those are the usual Munger assumptions – stable economics, no multiple crush, no dilution, etc. Fair value: Why are you referring to "a fairly valued acquisition" only in relation to the alternative investment, and not to Fairfax stock itself?

I'd say it's the other way around: If the FFH price is below IV, theoretical it's hard to find another investment better than buybacks. At least at times. Not only for FFH, but for most enterprises. Thought experiment: If Fairfax hypothetically achieves a ROE of 15% over long time horizons, then that means nothing other than: It has the ability to generate $15 in profit from $100 in equity. It does not generate $10 or $20, but precisely $15. So if FFH does everything as it has done in all its activities over the years, i.e. investing in equity and bonds, etc. (sometimes with good results, sometimes less favourable) and with average leverage, etc., then 15% is what it could achieve (... and at FFH, this includes the totality of investments in a wide variety of investment opportunities). Now let's assume that FFH should be worth $3,000 (intrinsic value), but could be bought at $1.500. Than here's the decision to be made: 1. Either investing every $100 of new equity within FFH and getting $15. 2. Or buying back that exactly same business, at 50 cent per dollar. I would take option 2. So if FFH buys its own shares for LESS than $3,000, it achieves a return of MORE than 15%, while within FFH the outcome would be strictly 15%. In other words: Buybacks below IV per definition achieve more than the business is capable of achieving on its own. Of course, again, this is theoretical. It's a thought experiment. Just another perspective. But it has practical implications. Everytime FFH can buy itself at a discount to IV, that may be the best choice - not always, but sometimes. Depending on the other opportunities at the time.

Sorry, I don't understand why and what you are referring to. Absolutely. But I didn't claim that either. I'm just saying two things. These are two separate points. They support each other, but I would never say that they are directly dependent on each other: Share price below intrinsic value = +1 opportunity to invest in. Share buybacks are just another investment opportunity for FFH, but ONLY if the price is below intrinsic value. So it's like the same opportunities to invest in PLUS buying FFH. If the price is above intrinsic value, this opportunity isn't existing. If you agree, that more opportunities to invest are a good thing, than a share price below IV is helpful from this perspective. The lower the price, the higher are implied returns on the repurchased shares. Do you agree? Circle of Competence and informational edge. With any investment, it is important to really understand it. Buffett calls it the circle of competence. And yet, there is no such thing as 100% knowledge/understanding. But you should be close. In addition – and this is very important – it is rare, but very helpful to have an information edge over the market and to know the current dynamics of a company as an investor. Viking is doing an amazing job; and yet we are certainly not overstepping the mark when we assume that Prem knows Fairfax better and sees more details of that mosaic "Fairfax Financial" ( @Viking often refers to parts of Fairfax that we do not see at all. Key words: "hidden value" - e. g. the pet insurance business a few years ago, It was sold for above 1bn dollar and no one was even aware of it - but Prem). This means Prem has a better understanding of the intrinsic value, is able to assess the value more accurately, and repurchases are simply safer than most other (equity) investments. Am I wrong? Haven't I made my points clearer before?

If FFHs stock price is down, it just gives Prem & Co an additional investment opportunity, that they wouldn’t have, if FFH would be priced above intrinsic value. How can any additional opportunity be a bad thing? One of the big advantages for BRK, FFH and MKL is, that by design they can invest into stocks, bonds, they can buy complete businesses, they can invest worldwide, or invest into float growth or they can buy out minority shareholders, TRS, … and - if they are lucky and the share price is below intrinsic value - they also have the opportunity to repurchase shares. All else equal a lower FFH share price gives a higher return to repurchases, right? There’s no guarantee in returns and you, Prem or anyone can be right or wrong with any investment. But it’s intuitive, that Prem doesn’t know any other equity investment better than FFH itself and he and his team understands the dynamics better than any other investor. So if there’s one stock he really has an Information edge, than it’s his own. So even though there clearly is no guaranteed return for any investment FFH does; FFH itself comes closest to that north star of a „guaranteed return“ for Prem. So the lower FFHs share price, the higher the „guaranteed“ repurchase return. And of course any other investment opportunity has to get over that guaranteed hurdle rate. Why shouldn’t we as shareholders be happy, if that hurdle rate stands at e. g. 22% instead of 15%?

Thank you again, @Viking. I really like the relative valuation approach! At the same time, I think it makes sense to compare any investment to the market as a whole. In other words, FFH and insurers as a sector compared to the market. As is so often the case, the most complete picture emerges when you consider multiple perspectives. It's not about right or wrong. Each approach has its own value, and only when you take everything together do you achieve a deeper understanding. What would be the point if insurers were trading at a P/E ratio of 30, and FFH would bei cheap with P/E 20, while the market was at a P/E ratio of 10, and the markets expected forward ROE would be higher than the insurers? Market comparison: insurers and Berkalikes vs. S&P500 There are good reasons not to use the S&P500 as a benchmark in its entirety. Not least the imbalance caused by the Mag7. And there are good reasons not to lump FFH, MKL and BRK together (size, different float levers, different strategies). But still the concept is similar (float investing, value investing approach, investing a relatively high percentage of assets into equity, investing into wholly owned businesses, decentralized structure, ...). Anyway: Which Yardstick could be claimed "perfect" on its own? Having said that, I would just like to point out that insurers – and in particular Berkshire and the Berkalikes, i.e. those companies that have relatively high "hidden value" – are very cheaply valued: FFH, MKL and BRK all have P/B ratios of 1.4 to 1.5 and P/E ratios of 9 to 16 (from you, see above) S&P500: P/B ratio 5.4 https://www.multpl.com/s-p-500-price-to-book, P/E ratio 29.3 https://www.multpl.com/s-p-500-pe-ratio). Okay, so compared to FFH, BRK and MKL as a group, the market is therefore somewhere in the range of factor 2 to factor 4 in terms of valuation. Historically, the three have achieved CAGR returns of 15% to just under 20% over 40 to 60 years and are among the top 1% (or top 10 among thousands of stocks) in their markets; looking ahead, one can perhaps expect 12% (BRK) to 15% (MKL) and above (FFH) as ROE over many years and the returns of reinvested capital should lie in that range too. So the reinvestment returns clearly lie above that of the market (over 100 years I think 11% has been the average return, right?). So in a nutshell, the average S&P500 company is valued around three times higher than FFH, MKL and BRK, while it will in no way be able to bring similar ROEs. Back to your question: Perpetual undervaluation would help FFH immensely. FFH is increasingly becoming a cash machine. A thought experiment: Fairfax's share price remains where it is – for all eternity. Fairfax buys back shares for $3 billion every year. Fairfax's market value is $38 billion. You own a single Fairfax share. Then, after just under 13 years, Fairfax would belong to you, even though you didn't even pay $2,000 for the share. The $5 billion+ (and it would probably be much more, but never mind) that Fairfax would henceforth generate annually as pre-tax profit would belong to you. Will it happen that way? Of course not! Someone would get there first and buy up shares en masse. The price would therefore rise. And yet this extreme example shows how lucrative buybacks are at today's cheap prices. It's like a rubber band: the tighter you pull it, the further it flies when you let go. The more undervalued a share is and the more aggressive the buybacks, the more the price will eventually skyrocket. The disadvantage of a high price is that you can issue shares and reinvest the proceeds. But the company is growing faster. What do we gain if FFH becomes as big as BRK? In any case, less than if FFH remains small and correspondingly smaller reinvestments still move the needle.

Yes, but I have another question regarding disruption, addressed it to Perplexity and got to this: My personal take is, thinking in three dimensions and maybe it's even chronological (?): Run for more efficiency: Big Insurers win against smaller in the " ("1st phase") Every insurer invests into AI, saves costs, margins get better... BUT: The smaller have to invest the same amount than the bigger ones - but their profit from winning efficiency is smaller. => This creates an advantage for the large companies over the small ones (THINK: "strengthening the moat" and "Higher barriers of entry"). This widens the moat of big insurers against smaller "classical" insurance companies and barriers of entry FFH is special because it can combine the advantages of small insurers with those of large ones if it performs well (the costs of learning and change can be shared; at the same time, smaller units are more agile and quicker to implement changes than large ones, AND FFH learns much faster overall than a centralised insurance unit due to the large number of entities and regions. Think "learning company". => This should help FFH even more than BRK, MKL and most others etc. BUT: Competition eats away margin gains through AI efficiency gains – either completely or partially ("2nd phase") If every big insurer has similar efficiency gains, competition will once again lead to falling prices and margin gains being eroded. This applies more to highly competitive, transparent and large insurance markets than to small, intransparent, "separated" ones, where e. g. other insurers don't have any data to compete (THINK: There's more competition in car insurance than in insurance of a models or footballer's legs in a special insurance deal). => Here, for example, I would see MKL in front, FFH in the middle and BRK at the end. But FFH has at least a special extra advantage: Regional diversification. That might give them more time to adapt in markets, that are not first movers, so they can learn in e. g. Ukraine or South Africa from what happens in the US (my best guess is, that this will start where AI comes from and is adapted the most into the economy...). "Real" AI disruption through AI disruption in the field of "face to the customer" ("3rd phase") AI changes how people and companies research for the best fitting insurer. Old analogue networks are becoming less important and losing their binding effect. The hurdles and time required to compare insurance policies are diminishing enormously, and decision-makers (whether B2C/private or B2B/entity) are being given new tools (whether they are giving better results, I cannot say; but they are used and that's the important thing within this discussion). A 100 pages insurance terms and conditions paper can be proved, understood and compared to others very easily in seconds etc. I just did that with finding an insurance for our new solar roof. You just ask, what's the biggest risk and which one is not so important, if the pricing is better... The question appears: Who needs human estate agents anymore in an AI search world? The role of the agent is under threat, at least in large areas of insurance, I guess. However, this shifts control over the "face to the customer" into motion. The old models (Think: "the estate agent's office on the corner" and "the existing insurance platforms in the digital world") erode. To me - clearly not being an insurance guy, I just know a thing or two about moats and disruption in general - here appears a risk that this upheaval will lead to a greater separation between "face to the customer" and "risk assumption" – and that new players will know how to exploit this to their advantage. You don't want to be the risk taker in the background, that could be changed immediately. You don't have to be the face to the customer yourself, but you do want to control the face to the customer one way or another. In all cases you do not want big parts of the interface to the customer being concentrated "elsewhere" in a room out of your own control. => I don't have any idea, how big this threat really is, but I perceive "change" and that's not necessarily good for moats. At least I am sure, that BRK - being more like a conglomerate owning an insurer, than an insurer owning companies, should be on a less riskier spot than MKL or FFH. FFH again has the regional aspect, so if things get ugly in e. g. the US, they might learn and invest into such a disruption. Still it would change the whole business model in a very meaningful way and the US is their biggest market. Here's - additionally - what Perplexity told me: "Likely disruption patterns: Embedded and B2B2C platforms: Tech and commerce companies integrate insurance as an invisible layer (e.g., at checkout, in mobility apps, in devices, travel), while the actual risk carrier remains interchangeable in the background. Disruption here: Loss of the customer interface, commoditization of risk, margin compression for traditional retail insurers. Vertically digitalized insurtechs: Fully AI‑first players in specific niches (usage‑based motor, cyber, parametric, health) that build underwriting, pricing, and claims around AI/automation from day one Disruption here: They attack profitable sub‑segments rather than the entire universal model – similar to how Tesla initially focused on the premium segment. New risk information: AI‑enabled telematics, geodata, and image analysis (e.g., wildfire risk from property images) shift information advantages to those who have access to these data sources and the models. Complete disintermediation of insurers is less likely due to licensing, solvency capital, and regulation; instead, roles are more likely to be reshuffled: • Who holds the capital and bears the risk? • Who owns the customer interface? • Who provides the AI infrastructure/models? In many scenarios, the insurer still carries the risk but moves a step further away from the customer and has less pricing power." I am not an insurance guy, so I may be off again here and there - but I am happy to learn as you know.

Yes, selling is only the "normal" and tax-inefficient way to realise profits. But Buffett in particular was very creative in this regard. I believe that if you gradually buy shares in a company and do not extract assets from the company in exchange for shares (like the Duracell deal), but instead take over the entire company in a final step, this is another way of avoiding paying taxes on the profit.

My best guess is, that most of us are calculating intrinsic value with too high taxes on current growth (e. g. "growth in book value per year"), while the real tax paid on growth is way lower (below 20% or 15% or even lower?). The difference comes from the deferred tax float, if I get it right. I think, that's another mosaic of the bigger picture: Holdings do have so many layers and all too often people tend to build in margins of safety in each layer, sometimes without even realizing. "If you have a hammer, everything looks like a nail" - the same seems to be true for a value investor with the margin of safety. So value investors typically calculate every part within a holding conservatively. And than the small investments are often completely excluded with reference to margin of safety; you wouldn't exclude any earnings income stream at, for example, Meta or Amazon, would you?). But then there is even less risk with FFH than with others. And there's a lag in paying taxes (not calculated by value investor). CR? Let's be conservative. Bond rates? Let's be conservative. Returns on reinvested cash? Let's be conservative. But why? FFH has completely different investment opportunities (bonds, buying shares or wholly owned businesses, buying insurance, buying in different regions, investments into organic growth of subsidiaries, or into insurance growth, TRS, etc.). I don't see e. g. NVIDIA having as much different investment opportunities. If other chips are needed in the world of quantum computers and they are not supposed to be the first, then what? If FFH has invested big in Blackberry and it doesn't pan out, that's manageable. So from a standpoint of risk it should be just the opposite: If you're having one or a few income stream, like e. g. most of the Magical 7 have, than - all else equal, e. g. equal moats - the risk is higher, than at e. g. BRK, FFH, BAM, DHR. Margin of safety is all important; but riskier businesses should be calculated with a higher margin of safety and good holdings like FFH trying to hide assets etc. the opposite, shouldn't they? Otherwise we do not apply the same standards to investments, underestimating some and overestimating others.

I have no idea, why the stock price came down. But it's good, if you like FFH repurchasing shares at an attractive price or if you want to buy it yourself. If you're short-term, than I understand you don't like it. But if you want to stick to this compounder for longer, why on earth should you be pleased if FFH has to pay more per repurchased share? Just some (not entirely serious ) guesses as to why the FFH share price may have fallen sharply today (Edited/added: Since someone just wrote to me and asked about the TSX60 forum: This is NOT meant seriously): - In another forum there's a TSX60 Subsection. They are examining Fairfax for the first time, being new to the index. When they looked at the investments, much of it seemed strange to them. A Greek Bank, a TRS. They were somewhat at a loss, what to do with it. But when they found out Blackberry being an investment, they ran to their desks and began shorting within a minute. - Maybe some of the institutional Fund buyers just wanted to show to their clients at the end/beginning of the new year, that they are well prepared for 2026; so they took the winners of last year into the window without checking to before. Now, as their clients are not looking at what the investors are doing with their cash under the year - and as they don't understand insurance ("boring") and FFH, and the more as it holds Blackberry, they decided to sell again. - Maybe a former investor came back to investing into FFH after (and as) it was put to the index and he read about it again and the many good moves of Prem. Than he rigged deeper and found the last $100+mn, that FFH still has in Blackberry. He became horrified and decided to sell immediately. Folks, I've said it a few times already and I stick to it: I think the BlackBerry investment is Fairfax's most brilliant move. It keeps the price nice and low, so that repurchases remain cheap.

My understanding is, that taxes are to be paid normally only after selling. Buffett: "Berkshire has access to two low‑cost, non‑perilous sources of leverage … deferred taxes and 'float'… Deferred tax liabilities bear no interest … In effect, they give us the benefit of debt - but saddle us with none of its drawbacks.“ I don't think we've written much about this free, low-risk float here so far. Does anyone know how big it is at the moment? Is anyone tracking it here? My best guess is somewhere around $600mn (tax on a gain of $2.5bn "hidden value") Do hidden values drag FFHs MCT Ratio (and as a consequence, its credit rating and its lending rates)? Another question that keeps coming back to me, but to which I can't find a satisfactory answer (I'm not an accountant...): Don't hidden values have the disadvantage that they are not (yet) considered equity – and are therefore not included in the risk assessment? The MCT ratio indicates the ratio of capital available (i.e. equity) to minimum capital required. So if parts of the equity are not reported at all (but are "hidden"), isn't the ratio being kept artificially unattractive? That would then have negative consequences for the ratings by the rating agencies; higher ratios lead to better ratings and lower borrowing costs, if I understand correctly... Or do the rating agencies adjust the equity as part of the MCT ratio calculation? I haven't found anything any hint into such a direction. If the equity relevant to the MCT ratio were to increase after a sale, the rating agencies could otherwise upgrade Fairfax's credit rating again – or am I on the wrong track?

Why? I thought the subsidiaries give the earnings to the holding and they reinvest it. At least that’s my understanding of BRK and FFH. Or isn‘t the named cash „extra“ in the meaning of „they don’t need for operations“? That was my understanding of @Txvestors comment? Yes, I like to look at the same thing from different angles too, and agree, that this is a nice one. In fact this perspective is it, that often brings me back to investing my money into BRK, MKL, FFH and not into other investments. Because with that perspective every of my investments NOT put into these three but in other equities implies, that I must be sure, not only to outperform Prems (Buffetts…) equity investments, but e. g. 2 times the bond interest on top. At least on average. It’s my test, before investing into anything else. I don’t come very often to an investment above those hurdle rate. That’s why 75% of my investments are in insurance companies. I just don’t get those bond returns „for free“ on my investments as BRK, FFH. So why not let Prem & Co do the job with that cost-free leverage on top?

Absolutely right! I tried to be conservative, but as long as we are at these valuation where we are right now, that’s definitely a low hanging fruit. @SafetyinNumbers Aren’t the subsidiaries alowed to buy back stock or give the cash to the holding and they do it or buy something else? I am not an insurance guy, so sorry for the question. But your answer seems to hint into that direction?

Thank you, I had never checked that number (other than at BRK, but WB wrote a lot about it over the years in his shareholder letters; other than Prem). So it's like $5.2B ($6.7B less $1.5B) dry powder and maybe even a bit more as we have yet another quarter behind us and they sold some stuff and assuming they haven't found anything big we don't know of. That's a lot. Put another way, with $5.2B they could put an extra 22.5% of book value to work, getting probably equity like returns on that 22.5%. AND even though not using it yet, they have reached a roe somewhere north of 20% in 2025. Assuming 10% ROE as a return on that 22.5% of book value would push the roe another 2.25% higher, or am I wrong? So might an ever growing cash pile be something that we will have to get used to...? Or are we currently on an unusual peak of cash plus credit line plus cash at subsidiaries, so it would be expected to come down meaningfully? At BRK it grew and grew; makes it very, very safe, but hasn't been exactly good for returns.

I wasn‘t aware there‘s such a fixed number like 6bn to calculate as dry powder?! In fact I wasn’t aware Prem communicated anything about this subject. What has he told us? I think for Berkshire Warren wrote something like 10bn in cash years ago (2010?) and now I just don’t track that any more as it’s just soooooooo much of it and they haven’t got near that level since sooooooo long. Interesting!