Liberty

What is your estimate of fair value, where would you buy? We can get this information from the president's letters, which I recommend you read. In 2013, they got their 8 analysts to do DCFs of Constellation Software: So we know that based on the assumptions made in 2013, the consensus model shows that their model price approximately matches the share price. At that time: - Share price( Dec 31, 2012): $108.09 - Share price( Dec 31, 2013): $208.24 - Share price USD( Dec 31, 2013): $196.10 (1.062 CAD = 1 USD) - Revenue per share (2013, USD): $57.13 - Multiple of revenue: 3.43x - Revenue CAGR (2004-2013)= 30% - Cash flow CAGR (2004-2014)= 30% - ROIC + Organic Net revenue growth (3 years, 2011-2013): 43, 37, 39 Now: - Share price (Jan 10, 2020): $ 1356.6 - Share price USD (Jan 10, 2020): $ 1038.25 (1.307 CAD = 1 USD) - Revenue per share (LTM, USD, 09/30/2019): $158.79 - Multiple of revenue: 6.54x - Revenue CAGR (2008-2017)= 26% - Cash flow CAGR (2004-2014)= 31% - ROIC + Organic Net revenue growth (3 years, 2011-2013): 35, 32, 33 Revenue growth has slowed, from 2018 letter: So organic growth (2008-2017) has been 2% (actually 1.9%), and in the period of 2004-2013 (which the DCF was based off of), it assumed an organic growth rate of 4.9%. How sensitive is the share price to that value? In the 2013 letter: So a subtraction of 3% cuts the intrinsic value of the shares by more than half. He also thinks that current ROIC is going to continue to drop, from 2017 letter: He also writes (2013 letter): So to return to the same revenue multiple as 2013, we need to drop by roughly 40-50%. Then we adjust 50% again for the 3% drop in organic growth. And then adjust again for potential slipping of ROIC. Then adjust again to account for drop in growth to 12%...where do we end up? If you think I am a madman for suggesting such a big disconnect between the current share price and their DCF model (even though they wrote it!), consider his discussion of JKHY in the 2015 letter (Section: Great Companies Are Not Always Great Stocks), he writes: Going back to our table, his idea of extreme overvaluation from the 2001 tech bubble is in the range of 5-6x revenue, and CSU is currently trading above that, at 6.54x revenue. Finally, I will leave you with this piece of wisdom: Hey, thanks for sharing your thoughts. I'll share some of mine. Mark Leonard is probably the most financially conservative person in the world. On top of that, he has an incentive in talking down the price of the stock to allow his employees to buy it lower, which is a talent retention tool. He's also said at the AGM that his sales of the stock this year (Mark Miller also said it) were signalling to try to drive down the price of the stock. Mark used the analyst consensus model because it's probably the less controversial thing to use, but does it mean that the analysts get it right? We can now know with hindsight that they had it totally wrong ever since IPO, by a lot. At almost any time you could've paid double (maybe triple or more, depending on when) what the analysts said was "fair value" and still performed very well. Also, aggregate organic growth hides a lot. The company is composed of multiple revenue streams that are worth different things. Hardware is worth the least, then professional services, then licenses, and then maintenance/recurring. The latter, which is where most of the value resides, has been growing organically at 4-5%. Another factor is that they buy distressed assets and runoffs. This makes organic growth seem lower even if these purchases create a lot of value (high IRRs) because they buy them at such low prices and are pretty good at turning them around or at least slowing down the melting and improving margins. Some periods have higher organic growth, maybe because they have fewer distressed assets in the acquisition pool, maybe because it was during the period when they invested more in organic initiatives (they reduced that investment because they found that they got better returns from acquisitions). Another thing I'd point out is that using multiples of sales can also be very misleading. Not all revenue is the same, right? You won't compare a dollar of revenue of Visa to a dollar of revenue of a grocery store or a profitless startup. "his idea of extreme overvaluation from the 2001 tech bubble is in the range of 5-6x revenue, and CSU is currently trading above that, at 6.54x revenue." If a business with no profits and a very uncertain future is trading at 6x revenue, that's very different from a business with high FCF margins, very sticky cashflows, and a very high internal diversification by industry verticals and geography is trading at 6x revenue. The tech bubble took place in a very different context (nascent commercial internet) with different kind of companies (growth at any cost without profits or barriers to entry for most of them). At a normalized FCF margin of 30% (which isn't too extravagant for this kind of software -- it can seem lower because they keep acquiring things that have lower margins, which has some transaction costs, and then they drive up margins over time), 6x sales is 20x FCF. Is 20x FCF that high for CSU when the SP500 trades where it trades and many other high quality businesses trade at 50x FCF? (look at TYL or MKTX or whatever). If CSU is supposed to trade at 3x sales, that's like 10x FCF at 30% margins. Or 15x FCF at 20% margins. Does that make sense? As for the JKHY example, if you go and look, JKHY peaked at 11.5x sales and 38.4x EBITDA in 2000, and yet it only just barely underperformed the SP500 by 1% over the next 10 years. That seems a positive data-point to me. I'd say that we're not in a 2000-like tech bubble right now. Mark is used to buying VMS at 1-2x sales, so 6x sales sounds very high. But CSU isn't a VMS. These businesses don't have reinvestment opportunities, they are mostly cash cows that grow at maybe 1-2x GDP. CSU has a long-track record of high ROIC re-investment, which makes its cashflows worth a lot more than the cashflows for a single VMS. As an evergreen buyer of software businesses, I'm sure he's very honest in saying that he thinks prices are too high... Also, the market has been discovering the economic value of software over the past decade, and valuations have risen generally (as interest rates have been falling, also increasing valuation multiples), which has also been a tailwind to CSU. It doesn't mean it'll keep going, but it can mean that software was undervalued 10 years ago, rather than assume that the valuation 10 years ago was correct (right after the bigger crisis in almost a century, back when the internet was much smaller and the total spend on software was much lower?). So yeah, ROIC might go down and shareholder returns almost assuredly won't be what they've been, but don't go too far in the other direction and be careful parsing what Mark Leonard says. He's like Buffett saying in 1980 that Berkshire was now way too big and wouldn't outperform like it did. He's eventually right, but it doesn't mean the game is quite over. The company has been improving its coverage of small deals, expanding to new geographies (more deals in scandinavia, europe, australia, south america, japan) and shown that it can scale up capital deployment (2019 was a great year with no big and few medium deals.. past years that have been close in size all had much bigger deals), and they've become more competitive with larger deals with the use of fenced-off leverage, so we could wake up one morning to a decent-size deal. They've only deployed something like $3bn in acquisitions in their history, so if they ever do a 500m or 750m deal, it'll be a pretty big event. These are some of my thoughts. I could be wrong, though.

I'm enjoying this interview with Jim Keller (engineer working on CPU design). This is what a clear-thinking engineer who naturally thinks in systems and first principles sounds like. https://podcasts.apple.com/us/podcast/jim-keller-moores-law-microprocessors-abstractions/id1434243584?i=1000464730369

Season 1 was great, definitely made me think this was the successor to BSG. But I thought the tone and writing changed after that and the quality dropped, which is too bad. Haven't seen the latest season yet.

Or that the starting valuation was very wrong. In hindsight, obviously, it either was overpriced then, or it wasn't too badly priced if they had executed on the future value creation that was expected. Everybody was screaming how cheap it was at the time and they had compounded capital 40% CAGR since the company was created and so on. They had all these call options on potentially huge homegrown projects that had been generated at low cost, and almost no downside risk because of the huge cash pile on the balance sheet and the royalty model that capped the downside and let others spend on capex, the smartest management in the space, would create lots of value in a downturn, etc. So what happened, right? Turns out that there are benefits to the model, but it's not magic, and doesn't automatically allows them to get high ROIC on the whole capital base.

New acquisition at Vela: https://www.fogsoftwaregroup.com/fog-software-group-announces-acquisition-of-asc-software/ "Zoominfo shows 144 employees and $27.7M in sales" h/t @pearnick

I'm not saying the move makes sense or whatever. I'm saying that the same price means two completely different things depending on how you look at it, which is a nice demonstration of anchoring.

That's factually true. But the point of investing is to make money, no? So I'm not sure it helps to say that other things also did poorly. The index is literally the default, something you can invest is for a few 10s of bps. So you should definitely compare yourself against it, at least in part, if you're an active investor. Not over-performing it all the time is normal. Under-performing it by 150% over a decade shows that something is definitely problematic with the approach. And in the case of Lin, he said that 100% of his portfolio was in this one stock, and I was trying to warn him might not be good risk control and that the stock might not be getting good enough ROIC on the whole capital base, as well as had reinvestment and terminal value challenges, to perform well going forward. I may be wrong, but it's certainly something to consider and discuss, and it wasn't much looked at in this thread.

I find this situation a good practical demonstration of anchoring. Because the stock fell from peak, a lot of people feel like the stock is down and doing poorly right now. But you would've told anyone a few months ago that Tesla would be above $700 and they wouldn't have believed you. So it's all about where you anchor and how far you zoom in or out.

No surprise you timed it so well. I think fondly of your kind and hyper-rationale comments back in May 2006 that gave many of us including me the confidence to buy FFH options and make incredible gains. Thank you again. What a contrast to the quality of this board today when I post at the end of October my conviction that the shorts were failing to consider the disruption entailed in the recent series of positive developments and Tesla was likely entering into an exponential phase. Instead of support and discussion what I got in response was "ok Boomer". Perhaps in the future we all can realize that if we support and aid each other we will all be rewarded. I've got $600 in unrealized gains at the close today. Woohoo! More gains tomorrow? Losses tomorrow? This is fun stuff. So are you out of the trade, or do you think you can grow that $600 bucks into the one comma club?

If you go to the bottom of the about page of Overcast, you can see how much time the silence-shortening feature has saved you (without counting other speed adjustments). So far I've saved 426 hours. And I usually listen between 1.75x and 2x, so that's probably another couple thousand hours there. (if you try to go from 1x to that it'll sound crazy, but go to 1.25x for a few weeks, and then 1.5x... your brain gets used to it and for most conversational shows where the information density is fairly low, it's totally fine... might not work for dense audiobooks, but just adjust for what works for you.) That's a lot of extra material thanks to these features... More than pays for itself (time is your most valuable resource).

People complain that they need to subscribe to Netflix and Hulu and HBO and such to watch the shows they want to watch... Wait until you need to pay Spotify and 3 others (or at least create accounts on and use their clients) to get all the podcasts you want to listen to...

2019 presentation: https://assets.pershingsquareholdings.com/2020/02/05122230/PSCM-Annual-Investor-Update-Presentation.pdf

Yeah, asking questions on a discussion forum is baaaad. How's your cup of coffee today? Good or bad? There's a 50% chance it's either one, right? Please. Did he bring some of the stuff on himself? Sure. But did you take pleasure on egging him on? I would bet so. Who are you talking about, SD or Lin? And what egging him on? I don't know what you mean by that. All I did was challenge some unchallenged claims and then discuss the answers, which is exactly what a discussion forum should be. When everybody is sitting on the same side of the boat, bad things tend to happen. It's healthy to have diverging opinions, it sharpens thinking. I'd rather that newcomers and lurkers who come to a thread get various takes and make up their minds about which makes the most sense than have everybody all singing kumbaya and saying that everything is amazing while the stock underperforms the index by 150% over a decade.

I use the Overcast app. My next choice would probably be the Castro app. I dislike Spotify's entry into the field because most of the innovation and dynamisms comes from the openness, like the open web. Once it gets all walled garden and fragmented, it'll likely stagnate and become worse for listeners.

Yeah, asking questions on a discussion forum is baaaad. How's your cup of coffee today? Good or bad? There's a 50% chance it's either one, right?

Of course. I don't think anyone will disagree with the point that timing is an important component to making money (you can be wrong and make money too!!). I'm just saying that in these really volatile situations, one can mentally think about outcomes that aren't binary, and using other means to express that view. Perhaps that's not worth the energy expenditure for you - totally get it. I happen to think the opposite, but that's not relevant to my point. Sure, that's true (though not in the specific case of Bitcoin in 2017, as options weren't available at the time, IIRC). But in other situations, you can usually decide to cap your downside and upside via options and limit risk.. But if you do that and things go massively in your direction, you might feel the pain of the missed gain too, which can be just as real as a loss to many.

He once said that 100% of his portfolio was in ALS (if I remember correctly), so that would both explain not participating in much else and the unusually intense attachment to the company. I guess he's really gone... Didn't even say goodbye

There are other options (no pun intended) to play the situation that may better allow one to tweak the way they think the various scenarios might unfold, and limit risk. Shorting is of course one path, but there are numerous options strategies to express a more nuanced view. Sure, my point wasn't about tactics, but about the larger reality that "being right" at some point isn't enough to make money. There's usually no such thing as a free lunch, other strategies have their downsides and upsides too.

I didn't mean that there's nothing to talk about. I mean it's "interesting" in that there's constantly newsflow and controversy and stuff to find and new products and updates coming out and it's all very emotional. Kind of like how on this board some threads have hundreds of pages and others barely have any replies. Teslaq community is like one of those hundred page threads, which isn't always a sign of good risk-reward.

It's a youtube video rather than a podcast, but I enjoyed this one:

It's like saying that Bitcoin is a bubble in early 2017 after it's up thousands of percents or whatever. Great. But if you could have shorted it at the time, you could then have still lost your shirt... before then being "proven right" when it fell 80% or whatever (and then went back up, and so on). Being "right" is worth very little unless you also time things really well with these types of super volatile and unpredictable situations. As for Tesla, I've mentioned elsewhere in this thread why I didn't think it was a good short. I could see how they could run into problems and implode, but they could just as well pull rabbits out of the hat (as they have in the past) and just power through and do well. There's a bunch of reflexivity on both sides; when things are going badly, they might have trouble raising more capital and the brand might be hurt, which becomes self-fulfilling... But when things are going well, more people are thinking positively about the company and they can raise money at very low cost, which is also self-fulfilling. So it then becomes, do you want to bet against someone who has survived the apocalypse again and again, proven early critics wrong by actually making a space rocket company and an EV company at the same time (while also doing other stuff)? Sounds like a headache to me... Easier ways to make money. Which also brings me to TSLAQ, that quasi-religious clique. These people get together and form this echo chamber mostly because this is an interesting company to talk about, not because it's the best risk-reward. Sears was a good short, as were the negative ROIC energy juniors or whatever, but you don't have a group doing box socials and creating one-issue Twitter accounts to talk about them obsessively. And once you're in an echo chamber and identify your ego with a position, there's very little critical thinking going on (even though they'll all say that they're the only ones thinking critically and everybody else is a brainwashed sheep or whatever -- yeah, that's why they've been wrong for 10+ years on a short that was "obvious" and a company that "can't survive much longer" with "bad products that nobody wants that constantly explode and kill people", with the competition coming out with good EVs "any day now", etc) I still think Tesla can run in trouble and shoot itself in the foot, but I also still think it's low-return-on-brain-damage since it's so unpredictable, and that a much higher stock price (even if it falls 50% from here) reduces their forward risk a lot.

Don't forget the tax shield too. Won't last forever, but it helps.

I have nothing intelligent to say (positive or negative) about Tesla (the company or the car) but the fitted curve intersection end-point corresponds closely to the South Sea historical point when Sir Isaac Newton massively re-entered the trade (after doubling his money before). History says that nobody could ever even whisper the "South Sea" words in Newton's presence when the dust settled and perhaps this explains that. These historical curve fittings (as well as mentions of Enron or whatever) tend to short-circuit real thinking, whether they turn out to be correct or not. You can still get your face ripped off (as lots of Tesla shorts have been finding out recently) even if something eventually turned out to have been a bubble or overvalued...

https://www.cnbc.com/2020/02/04/saudi-arabia-fund-dumped-nearly-all-of-its-tesla-shares-in-the-fourth-quarter-filing-shows.html Perfect.

They broke out Google Cloud but include G Suite numbers with GCP, so we still can't know for sure... I wish they'd just have GCP. Otherwise, it's like MSFT putting Office 365 numbers with Azure, makes things less clear.