site banner

Some thoughts on the trajectory of Elon Musk

Naturally prompted by the current Twitter situation, I've come to the point where I just have to write down my thoughts.

I have no doubt that Elon Musk is a genius, both of thought and action. He can formulate visions and execute them. He has two truly epic feats under his belt - starting a viable car company from scratch (the first since the 1930s) and bringing about the next generation of space technology and exploration, after a long, long winter. This is definitely not the work of an "emerald mine heir, just investing his money."

He is however not an infallible genius, which is particularly noticeable in areas outside of his core expertise. And that includes social networks. In some sense, it might be the kind of venture least amiable to an engineering, top-down approach. The product is made of a fickle, unpredictable human mass and there are no good instruments or levers to make it do what you want.

The first thing about the whole Twitter situation which really gave me a pause was the fact that Musk had apparently waived due diligence as a part of the $44B takeover bid. This is completely incomprehensible to me. From an M&A perspective, it's like a story of someone who picks up a skank at a seedy dive bar and proceeds to raw-dog her. Incredibly irresponsible. Are you sure you don't want to use a condom? Things might seem easier in the moment, but the potential for future regret is rather alarming! The rebuke I've heard was that Dorsey had already told him all the important stuff anyway, but that's just not how the process works. For one, the due diligence could have given him a way out of the bid (and boy, wouldn't that turn out to be handy...) It's not guaranteed, but rare indeed is the DD that doesn't uncover some sort of irregularity or dubious representation that could have served as ammo in the lawsuit. Secondly, the DD would have mapped out the exact internal structure, external relations, responsibilities and exposures. Even if (or rather precisely because) the plan was to mow through the ranks, this would have been extremely useful to have. If you're going in with an axe, you should at least have a map of the areas you intend to clear-cut. The whaling system deployed by Musk might have been effective at selecting for a combination of competence, drive and vision alignment (and/or desperation) - but that's not the same as critical institutional knowledge. Twitter is vast and something like 80% of the people who knew what went where and why are gone. The sole irreplaceable value of Twitter is in its existing user network - but this is inextricable from the pulsing, living IT snarl containing the accounts and their connections, which is in turn inextricable from the human apparatus building it and maintaining it. With cars or rockets, as long as you have the tech packages, you can always just bring in new competent engineers to continue the work. But there isn't any objective singular blueprint of Twitter. No single person has the whole picture. It's dubious whether it can even be successfully cold-reset. It's just... why go about it that way? Why not put on the condom?

The second incident was the checkmark fiasco: 1. Blow up the old and opaque verification system 2. Concoct an $8/month pay-to-play scheme 3. Discover why the verification system had been there in the first place 4. Clumsily return to a variant of the old opaque verification system. I'm sure the advertisers were thrilled. How am I not looking at an impulsive, poorly though-out spiteful action here? There are people stuck with GIANT PENIS handles to this day...

The thirds aspect is Musk ostensibly sleeping over at Twitter HQ, wildly coding into the night with the bros. The problem is that either his ethos of "You can't put in less than 80 hours a week and expect a thing to work." is wrong or Tesla and SpaceX are getting the shaft here. And the stock price sure seems to indicate the belief in the latter. More than half of the value gone, YOY, as of the time of this writing. And heaven knows what's happening to Neuralink or the Boring Company. Precisely to the degree that Musk is an irreplaceable genius, the Twitter stunt is coming at the expense of projects he himself considers vital for the survival of human consciousness. What are the priorities here?

The further unmentioned elephant in the room is stimulant abuse and, even worse, the attendant lack of sleep. At this point, it would take a lot to persuade me he isn't up to his gills in some Chinese designer hyper-opti-MegaAdderall regimen, which just appears as both the likeliest cause and result of his recent actions and decisions.

The historical parallel I'm most reminded of is Napoleon. Certainly no rando of middling qualities - but also somebody who, past his initial bout of success and innovation, slumped into the belief in his own brand of unerring radical decisions, with well-known consequences.

So I'm out. Not that it should matter to anyone in any practical terms, but my confidence in Elon Musk's process and vision is gone. At this point, it mostly looks like the driver's seat is occupied by erratic hyperconfidence. I'm not expecting Twitter to disappear any time soon, in fact I still consider it somewhat more likely than not that the company will ultimately stabilize. It's not that any single action had caused irreparable damage - but the series of unforced errors, starting with the bid itself, isn't inspiring any future confidence in me. I will not be getting on that rocket to Mars, thank you very much.

22
Jump in the discussion.

No email address required.

This rush to adjustment is absurd.

(1)

"But it looks like Musk is doing a bad job at Twitter." This is usually a reasonable predictor of failure. However, everything Musk succeeds at looks like an absolutely terrible idea until all of the sudden, it's not. So hold your horses.

In the case of Musk, the claim "it looks like he's running company x like a total clown" isn't good evidence that he doesn't actually know how to run company x. It is likely that he would look like in idiot even in the case that Twitter eventually becomes a trillion dollar company. Everything we are seeing is weak evidence of what the future outcome will be. The nature of market inefficiencies is that they are not obvious, else someone would have already exploited it. Finding major market inefficiencies almost requires you to see things in a way that would seem absurd to everyone else.

This is just the beginning. There are so many things that Twitter can become. I don't understand how people are writing him off so easily. How would you expect a world where Twitter is a huge success to look different than it does right now.

Did you think that Musk was going to do things that everyone thinks are totally reasonable, just as everyone expected based on his clear operational excellence a couple of weeks after buying it, it does really well? That was never going to happen.

I would expect a success to look something like "Musk is a total idiot" to "see I told you he was a total idiot" to "OK, it looks good now but it's just a blip on the radar" and eventually "Ya Musk was successful but it's because of his workers not him. Any rich guy could do this" and "did you hear about his fathers emerald mines?".

I don't understand how you are distinguishing a world where Twitter goes bankrupt from a world that it's absurdly successful. The expectation is that everyone would say that he was an idiot. Everyone would say that he has no idea what he's doing, and eventually people would say how it was obvious in retrospect (and had nothing to do with Musk). I don't understand why you would expect Musk to run his business in a way that doesn't seem stupid. Everything he's done for the last 20 years has seemed stupid. Until it wasn't. And people still think that every new thing he does, at those same businesses, is stupid.

Surprise, when people exploit market inefficiencies it looks stupid. If it was obvious sthen omeone would have done it already. Musk has a long history of doing stupid things that are totally reasonable in retrospect.

I don't know if Twitter will succeed or fail. I do think that if Musk does spend a lot of time at Twitter (at the expense of Tesla and SpaceX) he will make a tremendous return. I do think that there is a very large chance that Twitter will eventually make an absurd amount of money and that when it does so it will look nothing like it does today; I think that Twitter the micro blog platform will serve as a launching point for other more profitable businesses.

Things currently look exactly like we would expect a Musk success to look; they also look exactly like we would expect a Musk failure to look. As of yet we don't have sufficient information to determine whether this is a winner or a total flop.

(2)

Twitter is a far more plausible business than either SpaceX and Tesla. Tesla and SpaceX were terrible businesses, the idea of them succeeding is absurd, but we are so used to their success it seems like Twitter is uniquely terrible; the reality is that Twitter operates on far more fertile soil than Tesla/SpaceX. Tesla/SpaceX were pie in the sky companies that were going to fail with near certainty. Conversely, Twitter is a company that should be worth more, it was expected to be worth more, but has failed to live up to its potential. In a vacuum, Twitter is a far more plausible than SpaceX/Tesla.

(3)

There are a lot of claims that Musk is just some luck business guy. This is absurd. The odds that someone would hit gold so often, on companies that seemed certain to fail is incredibly unlikely. His first 2 businesses Zip2 and X.com (X.com acquired paypal and later rebranded as Paypal) might have been luck. He was a smart guy, but there were lots of smart guys making lots of money like that before the dotcom bubble. So it's impressive, sure, but that's it.

However, Musk then picks two companies that were near certain to fail, just terrible ideas, even if they had adequate funding, which they didn't, and turns them into no only functioning businesses, but implausibly large successes. I don't think you appreciate how absurd this is, and how this, in combination with his two software startups, suggests that there is property of Elon Musk that makes a company likely to succeed. It would be improbable that he was so consistently lucky in not only business, but in what were the WORST businesses; these businesses were terrible ideas. You would have to be looking to loose money to touch them. The nearest competitors to both businesses have some of the largest revenues in the world and don't touch either of those markets. The improbability of succeeding at all these things is implausibly improbable.

He had 180 million before tax, and wanted to start a rocket ship company and a car company, despite not having nearly enough capital to make a weak attempt at either.

There hadn't been a mass manufacturing automotive company since the early 1900's. Starting one should have required a ton of capital and time. You might not know this but Tesla has the highest automotive gross margins of any car company not named Ferrari, despite Musk's silly automation and manufacturing ideas being mocked as the whimsical overconfidence of a fool. He was a complete idiot until maybe 2 or 3 years ago, and turns out all those stupid ideas created the highest automotive gross margins and companies are tripping over themselves to incorporate them into their assembly lines. Tesla has higher gross margins than Toyota (the king of not only auto manufacturing* but of all manufacturing*), Mercedes, and BMW, despite the current standard cost of manufacturing electrics cars currently being $10,000 higher than that of ICE vehicles. This is absurdity upon absurdity.

SpaceX is just an exercise in implausibility and improbability. The odds of its failure were a near certainty. It's rivals were rich governments, and defense contractors that make bids on Trillion dollar government projects. It would have been a near certain failure with adequate capital, which it didn't have, not anything close to it. The majority of satellites in space were launched by SpaceX. The overwhelming majority

If he would have gone out and looked for business that were most likely to fail a rocket ship manufacturer and mass production automotive company would have been pretty good attempts. You need to go back in time 20 years and consider how absurd these companies were. They were terrible ideas. Certain failures. They both succeeded beyond any imaginable best case scenario.

I don't care that Musk makes cool rocket ships and cars. I don't care that he's "disruptive". That means nothing. What matters is that he picks businesses with the most infinitesimal improbability of succeeding. It wouldn't matter if he had founded a company that makes industrial solvents conditional on its success being so incredibly improbable. T

he salient feature of Musk isn't the cool stuff he makes, it's this strange property that lets him take businesses that are certain failures, and to not only make them profitable, which would be absurd, but absolute giants, that dominate the spaceflight industry (civilian and private) and produce significantly higher gross automotive margins than those of not only BMW and Mercedes, but Toyota as well; all this despite the industry standard that it is $10,000 more expensive to make an electric car. FYI up until the recent supply chain woes, Tesla was increasing gross automotive margins each quarter by a couple of percentage points, despite cutting the prices of the cars by an even larger percentage (that's wild).

Musk is simultaneously the most overrated and underrated person alive. He's overrated because people think he's amazing because he makes cool cars and rockets ships, and because he's super "disruptive". Meh. He's underrated because people don't appreciate the degree to which he puts himself into situations with the most overwhelmingly certain probability of total failure and turns them into implausibly massive successes.

There is some special quality that allows Musk to totally defy probabilities, to an insane degree. At a certain point it's not luck. There is no reason to expect that this weird ability would not extend to Twitter. Everything he does is outside his domain of expertise. Twitter is the closest thing to his first businesses were he wrote all the code, so prima facie, if he has any "area of expertise" Twitter would be his only business that is plausibly related to that area of expertise.

I think that he is very very smart. I think that people dismiss this because fanboys on Reddit praise him for silly superficial things like disruption and making cool stuff. What you should care about, what matters for this conversation, and what is actually impressive about Musk is his ability to give the middle finger to the Gods of plausible outcomes. You hear so much fanboys praising him for silly reasons that one would dismiss the whole thing. He deserves that praise and more, but not for the reason that he gets it.

The whole idea that he's a genius because "cars, rocket ships, and disruption" is silly; it is Musk's absurd ability to shift probability's that stands out. By analogy it's like saying that Tom Brady is the best football player of all time because he's good looking. It totally misses the boat.

Everyone needs to take a step back and wait.

Those are fair points, but in each of those cases Musk was starting those businesses more or less from the ground up, and each of those businesses was based on a big dream that would revolutionize some industry sector. With Twitter he's spending a ton of his own money to take over an established business. And it's not like Twitter was flailing and in danger of going under and he had big plans to save it; he had some minor quibbles with moderation policy and spam filtering. The decision to buy Twitter (and waive due diligence) came across as impulsive, and proved to be so as he spent months trying to get out of it. He then gets rid of half the staff within weeks of taking over and replaces Twitter's verification system with a pay for play model that's so bad it's removed pretty quickly. Terminating staff may have been the right decision, but it's the kind of decision you come to after months of studying the problem since the cost of waiting is much less than the cost of finding out that the company actually needed a lot of the people you fired and the platform goes to shit in the meantime. The blue check thing is exactly the kind of thing that anyone could have predicted had they spent more than 30 seconds considering the possible downsides of giving blue checks to anyone who paid for one. And you can blame the advertisers leaving on ideological reasons but this is a copout—when your business model relies on advertisers then how they'll react has to be taken into consideration. If short-term instability makes it look like advertising on Twitter is a kin to lighting money on fire then advertisers aren't going to play. If Elon can right the ship in the long-term, then great, they'll come back then, but in the meantime he's going to have to make do with what he's got.

I think all these problems are emblematic of why Elon is, at least so far, ill-suited to run Twitter. He's used to starting things from the beginning, relying on investors for funding, and defining the ethos of the companies he runs. With Twitter he's taking over an established business without regard for what made the business successful in the first place, and is violating norms that may cost it. Elon is losing advertisers because Elon never had to rely on advertising revenue. If Elon wants to remake Twitter in his own image that's fine, but he has to realize that that means the company will have the revenue stream of a startup without being able to attract outside venture capital. He's going to have to fund this himself. He won't simply be able to make these kinds of changes while preserving Twitter's prior revenue stream. He wants to have it both ways, but it doesn't work like that.