Bottom line: Tesla has no actual product to justify the stock's sky high price.
GM, Ford, Toyota etc. all have a price to earning ratio (P/E) of less than 10.
Tesla has a P/E of 120.
But, but, but --- Tesla is more than just an auto manufacturer. Again, there is little in the way of actual product to justify this viewpoint. 90 percent of their income is from autos.
But, but, but --- Tesla is a "growth stock" with huge future potential. Again, the actual data shows otherwise. Sales in 2024 actually declined.
The only reason Tesla has a P/E of 120 is because the so called "smart money" (aka institutional investors) remain bought into Musk's fantasy in a big way.
Musk's best skill is con artistry and Wall Street is one of his biggest marks.
The best way to justify that high p/e would be for them to use it to buy out the competition.
It costs Tesla ~<6% of market cap (780b) to acquire Ford (40b) or GM (44b) (with a friendly ftc); so for 15% of their shares, that could become the only "American" manufacturer - then they'd probably strip those 2 for parts, end the dealer model, bust the unions, maybe keep a few of the popular models.
Some of this might be inevitable anyway, as autos become more expensive (tariffs, inflation, chips, unions), and more reliable (dealer maintenance model, fewer sales)
I don't know much about A&Ms or frankly business, but it seems kinda strange that this hasn't happened yet. We got the Stellantis Car borg before the Tesla car borg.
You spin off a company with all those obligations and pay it a pittance to take them on as supposed assets, soon after it declares bankruptcy. Now your new megacorp is free of all those pesky contracts.
I believe this too, but I’m also pretty nervous about trying to cash in my beliefs by selling it short or buying put options. The old adage “the market can remain irrational longer than you can remain solvent” still holds true.
It’s expensive to hold a long term short position against Tesla. Options expire, leveraged ETFs decay, opportunity cost. I’m sure there are some Michael Burray types that are long on short positions but as a retail investor I can’t do it.
The stock has show amazing resilience in face of continuous bad news. To summarise just the recent bad news:
- Sales are down 15% in California
- Cybertruck production has been reduced in Texas due to poor demand
- Two models have been pulled from sale in China
- RoboTaxi delayed again, this will probably never ship
You can write put positions against a short position to have a positive carry short.
It's not an awful approach when a stock already has fear and high volatility. To step it up a notch you can check the ratio of recent implied to realized volatility, and look at option skew (often Puts are extra expensive relative to calls in times of stress) as well as volatility smile shape (far out of the money tail risks can run very expensive in times of stress). All of that is easy for a retail participant, or I should say, anyone expecting to make money from shorting should probably be able to do the "middle school" stuff like this (brokers like ThinkOrSwim will have this info a few clicks/15 seconds away from typing in the ticker).
If you truly know what you are doing, short shops will do things like amplify their exposure and then use software to precisely hedge correlations (short a bank? Long a precise basket of other financials to cancel most of the broad exposure out). For others, the approach above can be nice, because in the case where you are not right about your view but not especially wrong either, the position is just harvesting risk premia which is a proven, core source of return from trading inherent to markets. Of course, if you make a bad bet you're going to get bad results. No avoiding that.
Those institutional investors deserve to lose every penny if they truly believe AI and robotics are going to justify that valuation. Musk and the investors are just keeping up appearances to avoid losing their money. And he appears to be out of ideas for futuristic projects, judging by his recent, overly generic promise of “abundance”.
The legacy automakers have been stagnant innovation-wise for a long, long time.
Tesla builds their own equipment to build the cars, which gives them an advantage in design flexibility. They could (or could have) conceivably eat(en) the market share of many of the legacy automakers before they could possibly catch up. I feel like most of the valuation is and was speculation in that direction. It’s not the existing auto makers to compare them to, it’s the entire auto market.
Unfortunately Elon can’t stay focused and thinks he can solve every major problem in parallel himself.
I hope the board can at some point realize that he needs to take a break from Tesla, and later come back if and when he can focus.
But, but, but --- AI and robotics. Again, Telsa has no real product to show that they can do AI and robotics any better than they do autos. It's pure wishful thinking at this point.
But, but, but --- robotaxis. Ditto --- same as above.
The really ominous "death cross" signal for Tesla --- their sales are declining (down 9% last quarter in the US) while EV sales overall continue to grow at a robust pace (up 11% last quarter in the US).
In other words, the competition is eating Tesla's lunch. But for some less than obvious reason, Wall Street continues to favor fantasy over facts.
Does Toyota, GM, Ford have an FSD that’s just as good?
Bad question --- Tesla doesn't actually have FSD. What they actually have is Level 2 automation that requires constant supervision just like the others.
Do they have charging stations all over the country?
Yes, they do. They are all adapting to use Tesla chargers.
There will be no 'Volkswagen' tesla, the model Y is going the way of the GT-R, and Tesla's other ideas are vaporware. Lidar is a superior approach. Refusing to use it, while claiming superiority, reminds me of how a child copes with losing at their favorite video game.
It doesn't need to be superior, it just needs to be good enough. It's something that has been proven in many technology battles. FSD has made an amazing progress over the last two years. If they solve last 1% of cases they will be in good position to beat Waymo on price as they scale. With ride sharing price is everything. People just go for cheapest rides. That's why cybercab is two-seater, they are trying to minimize cost per mile. Smaller car means cheaper car and less energy per mile.
I agree on your point about good enough. I don't see how they defeat foggy weather and the like with cameras alone however, so I'm wondering if they will ever get that last 1% as things stand. I think they need more sensors to stand a chance at that.
My model Y has trouble with sunny weather today, so even that isn’t quite solved. Driving east into the morning sun constantly gives warnings that the camera needs to be cleaned.
I'm disappointment how unrigorous that lawsuit is. Their argument basically relies on the fact there's a patent filing that vaguely mentions this, and that the odometer sped up even though the plaintiff's driving habits haven't changed. Those are certainly reasons to be suspicious, but aren't exactly smoking guns. Plenty of things are patented but aren't implemented in reality, and we only have the plaintiff's word that there's no other change in driving behavior that could have accounted for the extra mileage. If they're spending all this time drafting a legal complaint, surely they can do a carefully controlled empirical test?
The lawsuit has low chances of succeeding but not on the merits. It exposes a regulatory gray area, both in the US and in the EU. Existing laws focus on tampering, not manufacturer accuracy.
"...Besides the disappointing result in the range test, the Tesla Model 3 had quite an unusual issue–its onboard trip meter was way off and essentially lied about the distance covered...." - https://insideevs.com/news/747548/ev-winter-range-test-norwa...
This is just more of the same - vague anecdotes. Why dig up 2 year old reddit posts to try to prove your case, when you can definitively prove it with a day's worth of test drives?
>"...Besides the disappointing result in the range test, the Tesla Model 3 had quite an unusual issue–its onboard trip meter was way off and essentially lied about the distance covered...." - https://insideevs.com/news/747548/ev-winter-range-test-norwa...
Is "trip meter" referring to the odometer, or the "estimated range" meter that's also displayed?
Given this hinges entirely on what is meant by "trip meter", and the little tidbit is only mentioned in passing with no elaboration, I think it's fine to question what exactly the author meant. That quote works equally as well if they were complaining about the range meter rather than the odometer. If the range meter said there was 200 mi left, then said 100 mi left after only driving 60 mi, I'd be pretty miffed, and might even claimed that it "lied".
"...This analysis identifies over fourteen distinct mileage calculation methods employed by Tesla—ranging from predictive telemetry algorithms to energy-consumption-based estimations—documented explicitly in patent filings and regulatory disclosures. These methodologies influence not just warranty expiration timelines, service monetization, insurance premium structures, leasing mileage penalties, resale valuations, residual-value forecasting, and regulatory credit calculations but also underpin battery health indicators, vehicle range, recorded energy efficiency, and much more. Consequently, Tesla's business model exhibits a unique systemic reliance on odometer accuracy, rendering mileage reporting a fundamental rather than peripheral concern..." -
>This analysis identifies over fourteen distinct mileage calculation methods employed by Tesla—ranging from predictive telemetry algorithms to energy-consumption-based estimations—documented explicitly in patent filings and regulatory disclosures
This is the "they have a patent!" argument all over again. Just because such algorithms exist, doesn't mean they're used in the odometers. It's not even suspicious that they have such algorithms. Batteries are far more complicated than a fuel tank, so even if they weren't trying to fudge odometer ratings, you'd expect them to come up with such algorithms. Finally, you still don't have a response to my main objection: why haven't done an empirical test? They spent all this time making data requests, looking up tesla patents, and digging up old reddit posts, but can't spend a day driving a tesla around to compare what the odometer says vs google maps/another car. What gives?
They're actively hiring teams to diversify away from the CATL dependency.
They've starting hiring teams in India [0] and Malaysia [1] to work with alternative LFP suppliers.
It's most likely Agratas in India (they manufacture the batteries used in Tata and JLR's EVs) and EVE Energy in Malaysia (they began an expansion outside Chin due to EU and US tariffs).
India and the US are about to announce a Bilateral Trade Agreement [0], which was already in the works during Trump 1. Musk visited Modi yesterday [1] and Vance is traveling to India in the next 2 days [2], so the BTA and sourcing from India is most likely the primary topic of conversation.
The Modi admin has also been cultivating Trump and Stephen Miller adjacent advisors [3][4][5] at the Heritage and Hudson Foundation for a couple years now too, by trying to use Walter Russell Mead the same way Graham Allison acted as a conduit between China and the US.
And besides that, the tariffs remain oriented as a de facto blockade against China, and as long as they remain at a rate above 60-70%, Chinese sourced suppliers remain price uncompetitive.
Your arguments come from a place of rationality and observations, without a doubt, but the decisions of material importance are being made by irrational actors. Maybe you’re right, maybe not, too much volatility to say with certainty imho. Might as well play roulette in the current macro.
On a recent APM Marketplace episode, Kai asked one of his guests to predict how this ends, and their answer was, paraphrasing, “haha are you kidding me? I can’t even tell you what the next 24 hours is going to look like.”
You could literally be in the room where the decision is being made, and I still wouldn’t give high confidence in prediction accuracy, based on observations to this point.
The same Trump that negotiated the USMCA as a replacement to NAFTA in his first term, only to return and start harassing Mexico and Canada (!) over made-up grievances? Same guy that flip-flops on national policy depending on the weather - that Trump?
Difference is neither the Trudeau nor AMLO admin made alliances with Stephen Miller adjacent acolytes, nor cultivated relations with the American (and other Anglo speaking) Right, and overindexed on diplomatic protocol with an administration that subscribes to a personalist protocol.
Even when Trump was not president, countries like India, Israel, KSA, Japan, etc tried keeping informal relations with that sphere, which made negotiating much easier from the start.
You're trying to find a pattern where none exists: so, if Canada did not intentionally cultivate Trump admin. insiders, does that excuse Trump being monstrous to them?
Hell, Denmark loans out Greenland as an unofficial American protectorate and the US retaliates by seeking to just steal their territory, despite their almost insane pro-American bias. To the extent of sending the VP there to challenge Denmark's hold on its sovereign territory?
Again - that's the guy you're reasoning with?
That's before you factor in the American Right's virulent hatred for Indians, who they see as parasitic, simply for coming to work in the US on H1Bs. I hope it works out perfectly for all parties involved.
Yep. Everybody with a brain knows that. The richest man on the planet is a tech con man.
And yet, this post will get flagged and downvoted into oblivion.
How did we get there ?
As for those downvoting - nope, I'm not hating for him or anything like that. It's just that I like to tell my kids that good guys can win. And Elon is not that.
I was willing to forgive Elon some of his eccentricities.
Until he repeated, more than once, obviously egregious untruths of Trump's.
I will admit I was late to the realisation.
Elon will say whatever he feels will progress his agenda, no matter how far disconnected from reality. He's jumped the shark and is doing far more harm than good. He needs to shrink into obscurity for his own, and the rest of the world's, good.
GM, Ford, Toyota etc. all have a price to earning ratio (P/E) of less than 10.
Tesla has a P/E of 120.
But, but, but --- Tesla is more than just an auto manufacturer. Again, there is little in the way of actual product to justify this viewpoint. 90 percent of their income is from autos.
But, but, but --- Tesla is a "growth stock" with huge future potential. Again, the actual data shows otherwise. Sales in 2024 actually declined.
The only reason Tesla has a P/E of 120 is because the so called "smart money" (aka institutional investors) remain bought into Musk's fantasy in a big way.
Musk's best skill is con artistry and Wall Street is one of his biggest marks.
reply