Baron Capital holds about $25 billion of SpaceX and $5 billion of Tesla. Its next thirteen positions put together come to roughly $15.5 billion.
Most managers with a concentration like that talk their book carefully. Ron Baron does the opposite: he says his firm has made clients $71 billion in profits, that $30 billion of it came from one founder, and that the next decade will produce several times more.
"Of the 71 billion in profits, 30 billion came from Elon, 40 billion from other people."
Baron started the firm with $10 million under management in 1982 and ran it to $69 billion by June; he first bought SpaceX in 2017, and Baron Partners Fund is, on his account, the best-performing mutual fund in the United States since it became a fund in 2003.
The full segment is covered here so you can skip it.
Here are the 12 numbers that matter.
👤 Guest: Ron Baron, founder, CEO and Portfolio Manager of Baron Capital, who has held SpaceX since 2017 and whose Baron Partners Fund is his firm's flagship
🎙️ Hosts: Becky Quick, Joe Kernen and Andrew Ross Sorkin, who present CNBC's Squawk Box
🧩 Other segments: Cameron Berg of Reciprocal Research, on whether artificial intelligence systems have anything like a mind
📰 Published: 16 September 2026 on YouTube (CNBC Squawk Pod)
🔴 YouTube | ⏱️ length not available
Key Takeaways
Two Musk companies are about two-thirds of the firm's equity value, and Baron is not trimming
$25B of SpaceX and $5B of Tesla against $15.5B in the next thirteen positions combined
The cost of putting a kilogram into orbit has fallen from $18,700 in 2010 to about $1,500, and the next rocket takes it to roughly $150
The Starlink satellite target went from 20,000 to 100,000 because of expected demand from AI agents, not from consumers
Only about a third of the revenue he models comes from subscribers; the rest is government, enterprise and mobile
He accuses the three US wireless carriers of announcing a pricing cartel in a press release
His worst investment ever was a price-fixing scandal at an auction house, which is why he watches for it
He is not giving much weight to humanoid robots, which he calls dystopian, even though Musk thinks they are the biggest thing at Tesla
A data center in space costs $2 and nothing a year, against about $10 of capital and $2 a year on Earth, on his figures
He says SpaceX builds one in four months against two and a half years for everyone else
He thinks inflation has run 4% to 5% for his whole life, which is the reason to hold equities at all
1. $71B, and Number One
The interview opened on the size of the concentration, which one of the hosts read off the firm's latest disclosed holdings.
The show's own figures: about $25 billion of SpaceX as of the end of June, more than $5 billion of Tesla, and roughly $15.5 billion across the next thirteen positions combined. Baron added that the SpaceX stake is three billion more than Nvidia holds.
His account of the firm's history: "So when we started our business at $10 million under management in 1982, hundred million in ' 92. In June it was 69 billion and we made our clients 71 billion in profits."
"Of the 71 billion in profits, 30 billion came from Elon, 40 billion from other people."
"But Baron Partners Fund in part because of Elon Musk is the number one mutual fund in the United States since 2003 when it became a fund. Not 1%, number one out of all the funds."
He also said the exposure is where he wants it. Asked whether he is buying more or is already fully exposed at roughly $30 billion, he said that is enough.
2. The Car Stopped for Ducks
Before any of the analysis, Baron told a story about his own car, and it is the clearest description of Tesla's self-driving software anyone gave on the show.
He bought one of the last 250 Model S cars, a special run that cost twice what the earlier ones did. He put the price at 160 or 170 against 80 for a standard car, and said used examples now change hands above what he paid. He is not selling it.
"And all the cars have self-driving. If you want it, you pay $100 a month."
He overslept one Sunday, typed the name of the restaurant rather than an address, and let the car drive. It opened the garage, reversed out, drove down a long driveway, opened the gate, looked both ways, turned onto the street and worked its way to Main Street — and then stopped short.
"And then I look out and I'm watching the front of the car and there's a large duck walking across the road with six little ducklings behind it." "It stopped for the ducks." The car then drove on and parked next to the restaurant.
"This car is unbelievable." He offered to lend one to a host to try in New York traffic.
On whether he is supposed to be driving unattended, he said you look around, and if the car makes a shaking sound you take the wheel. At night in the Hamptons, where there are no lights, he said it sees more than the driver does.
3. Tesla's Share Is Rising
Baron said the widely predicted loss of Tesla's US market share to legacy manufacturers has not happened.
His read of the news of a few days earlier is that Tesla's share is increasing again, because the legacy makers are not going all in on electric vehicles. The rivals lost billions and took shortcuts, he said.
His framing of the difference is a demand argument rather than a product one. A buyer of another electric car is migrating to the technology; a Tesla buyer wants the car itself.
"I don't want to spend too much time on Tesla, but I think Tesla is the time to buy the stock is now." His reason is self-driving adoption: he cited a user base in the low millions, growth of about 55% a year, and 55% of new buyers taking the self-driving option.
4. The SpaceX-Tesla Question
The hosts put the question directly: Musk had been asked on a podcast about folding Tesla into SpaceX and had not knocked the idea down. Baron owns both.
"And so anything he says is better I'm supporting. And I gave him reasons why both it's good and why it's not good." He said he told Musk the reasons on each side and that he is not the one who makes the decision.
Asked what those reasons were, his answer was one line: "I don't want to say."
5. The Robots Look Dystopian
Asked how much of his Tesla thesis rests on the cars against the humanoid robots, Baron separated himself from the founder.
He is not giving much weight to the robots, even though Musk thinks they will be bigger than anything else he has.
His description of seeing them: seven feet tall, doing chores, then walking to a charging post and hoisting themselves up to recharge. "It's really dystopian."
What he does count is the rest of the technology stack. Batteries matter, energy is a large part of the business, and Tesla is a partner with SpaceX in a new facility he described as 100 million square feet — against the Pentagon, which he put at about 5 million.
6. From Minus $15,000
Baron disclosed his own positions when asked whether the firm is overexposed.
His largest personal investments are SpaceX at about $5 billion and Tesla at about $1.5 billion, with a further large holding in his own mutual funds.
He set that against where he started: in 1970 his net worth was minus $15,000 and he did not have enough credit to buy a telephone.
7. Four and a Half Hours
The anecdotes Baron uses to explain why he owns the companies are about how Musk works, and both come second-hand from people he has met.
The head of Intel, a partner in the new facility, visited Baron and described a site visit. Most visitors walk around for an hour and a half and leave. In Baron's retelling of what he was told, Musk stayed four and a half hours and had engineers brought in one after another to answer questions.
A founder on the Forbes 30 Under 30 list showed Baron a photograph of Musk at the wheel of an electric boat she had invented, which flies above the water for military and commercial use. Musk drove it, then handed the wheel over, walked out on deck and lay down to listen to the engines. "And he walks out to the deck and he lies down on the deck and he listens to the engines." "That's his personality."
His recruiting numbers, as he gave them: 130,000 to 140,000 employees, 30,000 new hires a year, and five or six million applicants for those jobs. "He's got the best."
The one person Baron has ever recommended to SpaceX took about nine months to get hired, and was interviewed in her final round by a former head of NORAD who asked a 22-year-old how she would design a satellite and then a network. She had done exactly that at a previous employer that had not given her enough to do.
8. $18,700 to $150 a Kilo
Baron's case for SpaceX starts with a single cost curve.
"Well, SpaceX people sort of viewed it as a rocket company and it's a rocket company because they can get to space more cheaply than anyone else."
"And NASA said that in 2010 to get to space would cost you $18,700 to get to space to get a kilogram into orbit." That was the price when a rocket could be used only once.
Reuse changed it. Repeated testing of the Falcon 9 from 2017 brought the figure to $2,700 a kilogram.
"And now we with a heavy, it now cost you $1,500 per kilogram. We charge other people 3,000 a kilogram, but cost us 1,500."
"And now when we have the new rocket, it's going to cost somewhere around a hundred or $150." He described the new vehicle as the largest ever built and roughly twenty times the capacity of what came before.
What the cost curve unlocks is who can use space at all. In 2010, he said, it only made commercial sense for governments and for large geostationary broadcast satellites; there was no low-earth-orbit constellation to speak of.
9. Starlink to a Trillion
The valuation argument follows from the cost, and the numbers are very large.
"So Starlink that is the internet for the entire planet. That means everywhere you are there's no dead areas." In the air, on the ocean, in the mountains, in the desert.
"And I thought that the revenues which last year were $18 billion dollars for SpaceX, they could be a trillion dollars for just Starlink in 10 years." On that trillion he models seven or eight hundred billion of EBITDA and a value of 14 to 15 trillion.
"SpaceX overall for everything now is worth less than two trillion. So two trillion to 14 trillion just for that."
The satellite plan changed while he was doing the work. "Now he's changed that. It's now 100,000 satellites." The earlier plan was 10,000 going to 20,000.
The reason is not more subscribers. He models a move from 15 million customers to 300 million over ten years — against three billion people with no internet at all and five billion whose service he called unsatisfactory — but that is only 30% to 35% of the revenue. "The rest is coming from the government, from enterprise, and from mobile. That's where you get the trillion dollars from."
What pushed the satellite count up is agents. Baron's account of Musk's view is that agentic AI will overwhelm everything else: people will have agents working 24 hours a day on their behalf, and the communications capacity has to exist for it.
The replacement cycle compounds it. Satellites are deorbited every five years and replaced with ones he says are ten times as powerful.
On margins: "I'm describing to you how demand is going to be huge relative to what anyone could have ever imagined. and they're going to make high 70% margins on that business is what's coming."
10. The Telco Cartel Claim
Asked what happens to the existing carriers, Baron made an accusation the hosts pushed back on twice.
"It's a monopoly." He put the three carriers' market capitalizations at 100 to 150 billion each and said there are only three of them.
Asked whether he was suggesting price fixing, he said the companies announced it themselves. "Not only are I suggesting it, they did a press release that they did it." He named Verizon, AT&T and T-Mobile. "They put out a press release that we that's the monopoly. That's the whole country. It's a cartel." His reading of the release is that the three are discussing how to price the satellite communications they do not have.
His explanation for why they need it is that they did not see the demand coming. "So basically they did could not possibly have foreseen agentic demand coming for communication."
"All they want to do is they want to keep their existing customers and they want to be able to retire and they want to pay a dividend and that's that's what they're doing. They're not risking anything."
The reason he watches for this is his own worst investment. Baron put $500 million into an auction house at $20 a share in 1999; the position doubled, then the chairmen of the two major auction houses were found to have conspired on pricing. One went to prison, the other stayed in England beyond the reach of extradition. "So we invested $500 million at $20 a share. It went to a billion. Then we sold it at half the price we paid for it. 250 million. The biggest loss ever."
The economics he cites for ground infrastructure is $16,000 a mile to lay wire, against $400 or $500 for an antenna pointed at the sky. He said federal incentives kept the wire programs going where satellite would have been cheaper.
The carriers' remaining asset, on his account, is spectrum — which he pointed out is a public good, awarded by auction.
His prediction is regulatory, and he flagged it as his own. "But the bottom line is that the government, they forced EchoStar to sell their business because EchoStar was just accumulating spectrum" without a plan to use it. "You can't just keep that in and bankroll and you think you're gonna get rich on that. So they forced them to sell to SpaceX." He expects the same pressure to be applied to the carriers. "That's my theory."
The hosts put the counter-argument — that carriers trade at far lower multiples than SpaceX because the business behaves like a bond, and that SpaceX would need enormous urban infrastructure for line of sight. Baron said SpaceX believes small cells solve it, or it can rent the carriers' spectrum, and that direct-to-phone satellites being tested this summer still do not reach through walls or into the middle of a building.
His aside on who wins that fight came with a biographical detour. Baron went to law school at night in Washington while working at the patent office in the day, and dropped out one semester short. "And the one thing you learn in law school is the best lawyers always win. And the government doesn't have as good a lawyers as those guys do."
11. Inflation Doubles It
With the Fed meeting that day, one of the hosts asked how worried he is about inflation.
His answer is that inflation has run at four or five percent a year for his entire life, which on his arithmetic doubles prices every 14 or 15 years.
"You got to think about how you preserve your buying power." It is the argument he has always made on the show for being invested rather than holding cash.
12. Data Centers in Space
Baron insisted on adding one more subject before the segment ended, and it is the part of his case that is furthest from consensus.
"So basically all around the country people are upset about data centers. They don't want them built in their community and they take up energy. They take up water supply." They are also noisy — and the intelligence cannot be produced without them.
His cost comparison: about $10 of capital and $2 a year of maintenance on Earth. "When you put it in space, it costs $2 for us." Once, with no annual figure after it.
The satellites are replaced every five years with ones ten times as powerful. "That means that in 10 years you have 100 times the capacity of what you started with. 100 times."
On build time and rent: "Instead of taking somewhere around two and a half years, we do it in four months. Four months, two and a half years." And "Instead of having our centers rent for $10 billion a year, we're renting them for 25 billion a year." He put the cost of a gigawatt at $50 billion for everyone else and $25 billion for SpaceX.
"That shows you how short we are at compute. Other people wanted to focus on the large language models. We focused on compute."
He calls the project Star Mind, and says it will be many times the size of Starlink.
On timing he was specific: not the ten years other people say, but 2027, with a data center in space next year. He noted Musk's own record — he does what he says, a little late.
One host raised the engineering objection: heat does not dissipate in a vacuum. Baron agreed it is the problem and said the answer is radiators. "You're absolutely right. That's that's a problem. And people point that out say hey man you can't do that."
On power he said the United States has 550 gigawatts and that Musk alone expects to add 100 gigawatts a year from the satellites, using a small fraction of the sunlight now going unused.
His closing claim, when a host recapped the $71 billion: "We're going to make hundreds of billions in the next 10 years from Elon. Hundreds."
Bonus Insights
SpaceX is making its own chips. Baron said the intention is to make as many chips as the entire world uses, for Tesla and SpaceX alone.
The 100 million square foot figure was the one number in the interview the hosts kept returning to, along with the $25 billion per gigawatt claim, which they made him repeat twice.
The segment's running joke was the DeLorean, which one host brought up because the flux capacitor needed 1.21 gigawatts. Baron said his wine-colored Model S is the coolest car he has ever seen, and that every other car on the road is grey.
He closed with a joke a ski instructor told him, about a man asking directions from a porch in Vermont and being told "You can't get there from here." It was his answer to whether the hosts could reach East Hampton to try the car.
Baron's bottom line is that SpaceX is not a rocket company but a compute and communications company whose cost of getting mass into orbit has fallen by two orders of magnitude, and that the demand for both bandwidth and data centers created by AI agents is what turns that cost advantage into a valuation several times the company's current one.
Products, Companies & Tools Mentioned
SpaceX and Starlink (The $25B position and the core of the thesis: cost per kilogram to orbit down from $18,700 to about $1,500, a 100,000-satellite plan, and a trillion dollars of modeled revenue in ten years)
Tesla (He says now is the time to buy the stock, on self-driving adoption and rising market share rather than on the humanoid robots)
Baron Capital (His firm: $69B under management in June and $71B of cumulative client profits on his account)
Intel (A partner in the new 100 million square foot facility, and the source of his four-and-a-half-hour Musk anecdote)
Verizon, AT&T and T-Mobile (The three carriers he accuses of announcing a pricing cartel, and whose spectrum he expects the government to pry loose)
EchoStar (Forced to sell spectrum to SpaceX because it had no plan to use it, which is the precedent behind his prediction)
Sotheby's and Christie's (The price-fixing scandal behind the worst loss of his career: $500M in at $20 a share, out at half of what he paid)
Watch the full episode:
If this was worth your time, send it to someone who follows the name.
Get the latest market chatter as it happens: