One widely known fact about Elon Musk is that he says a lot of stuff that isn’t true. Most famously, he has a long history of promising that Tesla Inc. is months away from rolling out fully autonomous cars. Here is the Wikipedia page “List of predictions for autonomous Tesla vehicles by Elon Musk,” which goes back to 2013. You could imagine a world in which:

• When Musk says “our cars will drive themselves perfectly with no human oversight in six months,” Tesla’s stock goes up, because that would be good and people believe him; and
• When, six months later, the cars do not drive themselves perfectly, Tesla’s stock falls, because the market was valuing Tesla based on optimistic expectations about autonomous driving, and those expectations were falsified.

In that world, you could make some easy money from Musk skepticism: He’d announce some optimistic thing, the stock would go up, you would short it, the thing would not happen, the stock would go down, and you’d make money on your short. You would have some risk on this trade, of course: If, this time, Musk was correct, the stock would go up more and you’d lose money. But if you were correct in disbelieving Musk’s claims, you’d make money.

But this is a childish view of stock prices, and Tesla’s stock does not actually work that way, for several reasons. Here are three

1. You’re not the only person who can read that Wikipedia page. Tesla investors do not unthinkingly believe everything Musk says, and if the stock price is high now, it reflects general bullish long-term expectations, not a huge specific bet on self-driving cars in the near term.
2. Relatedly, even if the stuff about self-driving cars turns out to be wrong and disappointing, Tesla has a lot of irons in the fire, and maybe some other good stuff will happen with, you know, robots, and the stock will go up.
3. Conversely, to the extent that Tesla investors do unthinkingly believe everything Musk says, they’ll still believe it in six months, and the stock won’t drop when a particular prediction is falsified. You’ll short Tesla, wait six months for the cars not to be self-driving, the cars won’t be self-driving, but the stock will be up. You’ll be like “people! Don’t you see that the cars are not self-driving,” but they will ignore you because Musk will be doing something else.

That Wikipedia page goes back to 2013, and the stock is up roughly a bajillionty percent since then. Also Teslas kind of drive themselves now!

A broader way to put this is that, even if you can correctly predict some important economic fact about a company, that doesn’t necessarily mean that you can profitably trade its stock, because:

1. It is not immediately obvious whether or to what extent the fact is already incorporated into the stock price: Maybe you’re right about the fact, but the stock already reflects it, so you have no valuable insight.
2. Stock prices reflect lots of facts, and if you’re right about the one fact, you might still lose money because other facts turn out against you.
3. Stock prices reflect investor psychology, and even if you’re right about the fact, the market can ignore it for longer than you can remain solvent.

If you are quite tidy-minded and also an Elon Musk hater, this will annoy you. You will think something like this: “Elon Musk keeps promising imminent self-driving cars. His fans and shareholders keep believing him. I alone know the truth, which is that he’s wrong. I wish to profit from my insight and their error, by betting against the imminent self-driving cars. But for some reason shorting Tesla’s stock doesn’t work.” You want some other product that will allow you to monetize your view more cleanly.

This product exists. It is conventionally called a “bar bet.” You can walk into a bar in the right part of Texas, say “I’ll bet anyone $1,000 that Elon Musk won’t do the thing he said he’d do on the timeline he promised,” and someone will stand up and say “them’s fightin’ words” and take your bet. And you will negotiate the terms and odds and resolution method, and in six months you’ll check back in and he won’t have done the thing and you’ll make $1,000. And you will get to say to the Musk fan on the other side of the bet, “nyah nyah nyah nyah nyah,” which is what you really wanted.

Of course there is a market structure problem, which is that you have to find the right bar and then laboriously negotiate the terms of the bet, so you can’t actually pour your life savings into Being Skeptical About Elon Musk’s Promises. Or you couldn’t until recently. But now we have prediction markets, which are centralized electronic exchanges, regulated by the US Commodities Futures Trading Commission, for coordinating bar bets. Here’s a Wall Street Journal story

Alan Cole put his life savings, all $342,195.63, into a prediction-market wager. … Until Elon Musk’s Department of Government Efficiency came roaring into the nation’s capital last year, he was largely a plain-vanilla investor or, as he puts it, a “normal, conventional Wall Street Journal-reading adult.”

But Musk’s boasts and his eager fans brought an unusual opportunity into the burgeoning U.S. prediction markets: People willing to bet that the world’s richest man would transform and shrink the federal government.

Cole took the opposite position, one he didn’t see as a gamble at all. If federal spending in each quarter of 2025 exceeded federal spending in the fourth quarter of 2024, he would win big. …

From Cole’s perspective, even if Musk cut government contracts and shrank the federal workforce—which he did—he couldn’t meaningfully dent Social Security and Medicare benefits. And that left no plausible path for cutting overall federal spending. ...

The key feature of the prediction market offered on the Kalshi website was that it measured federal spending in annualized, seasonally adjusted nominal dollars. To win, Cole didn’t need spending to stay above a past projection. He just needed federal spending to go up, as it almost always does.

He made like $128,000, or 37%. Notice the features of this product:

1. If you have a view like “federal spending will go up,” it is immediately obvious whether and to what extent the market incorporates that view: The market price just is the market’s expected probability of that happening.
2. This product reflects only one fact, whether federal spending goes up or not.
3. Investor psychology will affect the price at which you get into the trade — if Musk’s “eager fans” think he will cut federal spending, then you will pay a relatively low price for the spending-will-go-up contract — but not the price at which you exit, because the trade resolves: Eventually, spending either goes up or it doesn’t, there’s some resolution mechanism to establish the fact, and if it went up you get paid. Tom Gara wrote on Threads: “He knew that it’s basically mathematically impossible to reduce federal spending, but he also knew Elon fanboys are often morons.” With Tesla stock, that is a problem; with prediction markets, it’s an opportunity.

It is the opposite of a stock investment. 

I guess the point I would make here is that this is still somewhat childish. The stock market exists to allocate capital to productive businesses, and those businesses are irreducibly messy. You do not decide how to allocate capital by understanding one isolated fact about a company’s business; you have to figure out which facts are important and then do your best to understand all of them. Financial markets are not bets on individual facts; they are bets on the economic consequences of those facts. 

Prediction markets are a “truth machine📰

We have talked

times

the usual modern way

That is, if you think about it, a bit weird: If a bunch of competing firms get together and sign an agreement to coordinate their actions and not cut prices, isn’t that an antitrust problem? One person who thought about it was David Nemecek, a lawyer at Kirkland & Ellis who has led a lot of big liability management exercises, and who said at a Bloomberg conference

That was just some public musing about an untested theory, but in December Optimum actually tried it out, suing its lenders and alleging that “the cooperative is a classic illegal cartel.” Kirkland & Ellis represented Optimum in its liability management, but it did not file the lawsuit. Nonetheless📰

aggressive LME work

raised the issue

And then (1) Kirkland stopped representing Optimum in January and (2) Nemecek left Kirkland for Simpson Thacher & Bartlett this month. 

Which is, if you think about it, even weirder. When Kirkland dropped Optimum, I wrote: “I don’t want to say ‘they all got together and agreed to boycott everyone involved in the lawsuit,’ but,” and then quoted the Financial Times story about the client pushback. I added: “I feel like Optimum’s (remaining) lawyers might want to add this to the lawsuit.” I don’t think it proves any sort of antitrust conspiracy, but Optimum’s story here is something like “our lenders have joined together in a powerful cartel that prevents anyone from trading except on their terms,” and the lenders getting Optimum’s lawyers to quit does kind of fit with that story.

Optimum’s lawyers did add it to the lawsuit. Bloomberg’s Chris Dolmetsch, Reshmi Basu and Irene Garcia Perez report today:

Optimum Communications Inc. said a top law firm’s withdrawal from representing it under pressure from Apollo Global Management, Ares Management, Oaktree Capital Management and others showed that they were participating in a cartel against it.

Optimum, previously known as Altice USA, sued the credit arms of those private equity giants and other creditors in November, saying they had formed an “illegal cartel” to freeze it out of the US credit market. In a revised complaint filed Wednesday, the telecommunications firm said “the antitrust conspiracy has only intensified” in the last three months.

“When Optimum sued, Defendants began searching for a way to retaliate,” said Optimum, which is controlled by billionaire Patrick Drahi. “They soon found a target: Optimum’s transaction counsel, Kirkland & Ellis LLP.” …

“Defendants’ success in bullying Kirkland served the Cooperative’s core aim – obstructing Optimum’s access to the credit markets – while showing the lengths to which Defendants will go to protect their scheme,” Optimum said in the amended complaint.

Here is the revised complaint, which includes this passage:

As one leading commentator put it, Defendants were angry that Optimum sued them for their illegal group boycott, so they “all got together and agreed to boycott everyone involved in the lawsuit.”

Sort of? That’s me, I’m the leading commentator, but what I actually said was “I don’t want to say ‘they all got together and agreed to boycott everyone involved in the lawsuit,’” which is not quite the same thing as saying “they all got together and agreed to boycott everyone involved in the lawsuit.” In some technical sense it’s the opposite, though in a more colloquial sense I take their point. In any case I did say they should add it to the lawsuit so this serves me right.

Incidentally I remain pretty sympathetic to the creditors here. As I wrote when the case was filed, “there is something off-putting about this whole [LME] process, a broad norm of ‘distressed companies should have to treat all of their creditors the same’ seems fine, and if all the creditors want to get together and agree on that then who am I to complain.”

One way to think about it is that, when they make lending or trading decisions, these creditors are competitors with each other, but in their capacity as existing lenders to Optimum, they are something different. Ares and Apollo and Oaktree are not, now, competing with each other to make loans to Optimum; they are already Optimum lenders, already part of the company’s capital structure, already in a loose sense co-owners of Optimum.

As the lenders put it in their motion to dismiss the lawsuit (citations omitted):

Antitrust law protects competition — not a borrower’s desire for leverage in renegotiating its liabilities in times of distress. That is why every court that has addressed antitrust challenges to creditor cooperation has held that the antitrust laws do not reach the challenged conduct. “[T]hin to the point of invisibility,” “border[ing] on the frivolous,” “the very opposite of price-fixing.” That is how courts in the past 50 years have characterized such antitrust claims—and they are right. …

Leveraged financing exists because risk is shared across many sophisticated lenders and bondholders, and the governing agreements hardwire collective action through voting thresholds, class votes, and consent rights. Plaintiffs acknowledge that so-called “liability management exercises” (“LMEs”) are “a creature of contract,” often pitting creditors against one another to “create individual winners and losers.” The creditor cooperation alleged here is a permissible response to that dynamic. It prevents opportunistic maneuvers that reward certain creditors at the expense of others; it reduces transaction costs and the destructiveness of brinkmanship; and it promotes the very mutual forbearance that can keep a borrower operating while preserving the contractual expectations that made the original financing possible. In short, the cooperation Plaintiffs attack is not merely lawful—it is procompetitive, and it benefits borrowers and consumers alike. 

Still weird to make their lawyers quit though.

One model you could have is: 

1. The optimal amount of fraud, in venture-backed startups, is not zero, and is in fact rather high. If you are a venture capitalist and you make 10 investments, then if two of them are huge successes with 1,000% returns, while the other eight are total failures, you’ve done very well. Economically you don’t care very much if the total failures are good-faith failures or frauds. Maximizing variance is good, backing wild-eyed visionaries
   💡

   The Bloomberg opinion article "Silicon Valley Is on Drugs," published June 27, 2023, argues that high drug use—particularly psychedelics, ketamine, and stimulants—is prevalent among tech founders, venture capitalists, and executives in Silicon Valley, driven by a culture of risk-taking, innovation, and performance enhancement. It highlights surveys showing elevated rates of psychedelic use in startup hubs compared to the general population, linking this to the personality traits of entrepreneurs who experiment with mind-altering substances to fuel ambition and creativity.

   [more]

   Commentators on Hacker News note this aligns with expectations for risk-tolerant innovators, though some express concern over VCs mimicking founders without adding unique value.

   [read]
   is the way to get there, and if one or two of the wild-eyed visionaries are just lying to you about everything, ehh, that’s the cost of doing business
   💡

   The Bloomberg opinion article "Slaying the Blood Unicorn," published January 4, 2022, portrays Theranos as a fraudulent "blood unicorn"—a private startup valued over $1 billion that misled investors and customers with unproven blood-testing technology. It highlights how Theranos, founded by Elizabeth Holmes, conducted nearly one million inaccurate tests before collapsing amid scandals, including voiding results and legal battles, while staying private to avoid scrutiny that public markets would impose.

   [more]

   The piece argues that unicorn scandals like Theranos thrive due to limited transparency in private markets, referencing over 900 unicorns by early 2022 and calling for mechanisms like tradability to expose misconduct faster.

   [read]
   . 
2. Venture capitalists know Point 1, and will have a sense of humor about fraud.
3. People who are not venture capitalists — who are not in this business of buying lottery tickets from wild-eyed visionaries — do not know Point 1, and have less of a sense of humor about fraud. If you’re just investing your hard-earned money in a local business expecting to earn a decent return, and it turns out to be a fraud, you’ll be angry and call the police.
4. Entrepreneurs know Points 1, 2 and 3, and everyone responds to incentives.

I don’t know that this model is quite right; I think it probably overstates the visionary/lottery-ticket nature of venture capital, and it is too glib about not distinguishing between risky ambitious companies and pure frauds. But I think there’s something to it. And the point is that the model would predict more fraud at companies backed by venture capital firms than at other companies. The VCs will tolerate some fraud, and entrepreneurs who want to do fraud will know to target VCs.

Here’s an NBER working paper on “Venture Fraud,” by Alexander Dyck, Freda Fang, Camille Hebert and Ting Xu: 

We assemble the first comprehensive sample of venture fraud cases involving 614 U.S. venture capital (VC)-backed startups founded since 2000. We find that VC-backed firms are 54% more likely to face fraud charges than comparable non-VC-backed firms within a subsample of newly public firms where detection likelihood is high and homogeneous. … We find that fraud is more likely in startups with stronger founder control rights, more convex founder cash flow rights, more investors, and greater participation of non-traditional investors. Founder-controlled boards are 88% more likely to commit fraud than VC-controlled or shared-control boards, even within the same firm. Governance variables matter much more than founder characteristics in predicting fraud. Hot funding conditions at the initial round, which weaken governance incentives, predict future fraud.

Seems right. As they put it:

Startups’ high growth pressure generates incentives to exaggerate or fabricate performance. The use of novel and complex technologies in markets with unclear demand can raise monitoring costs as outcomes stemming from deception are often difficult to disentangle from those driven by technological or market risk. Moreover, the strength of VC governance may have been overestimated. With fat-tailed return distributions, VCs have incentives to prioritize upside capturing over downside mitigation.

A VC who cares a lot about avoiding fraud is going to miss out on a lot of good deals, and missing out on good deals is worse than getting defrauded a little bit.

Insider trading, I often say around here

This is all too cute and nobody quite thinks it. Way back in 2013, the US Securities and Exchange Commission reached a $600 million settlement with a unit of SAC Capital Advisors over some insider trading. Mathew Martoma, an SAC portfolio manager, got inside information from a doctor who worked on some drug trials, and used that information to trade the stocks of Elan Corp. and Wyeth. The SEC caught this, Martoma went to jail

talked

times

Eventually — in 2024 — the SEC went back to court to be like “I think we’re done here, we’re gonna give the $75 million to the Treasury,” but Pfizer Inc. (which now owns Wyeth) objected that it should get the money, because after all Wyeth was the real victim of the insider trading: Its secrets were stolen by Martoma, so it should get compensated. In November 2024, a district judge disagreed:

Pfizer, in conclusory fashion, argues that Wyeth was harmed when its fiduciary misappropriated the results from a clinical trial for insider trading and suffered reputational harm from the proceeding scandal. The Court certainly agrees that corporations whose secrets are misappropriated for insider trading purposes are generally victims of wrongdoing, but Pfizer has failed to allege how the insider trading scheme and Wyeth’s subsequent reputational harm qualifies as pecuniary harm for purposes of distributing the disgorged funds. Pfizer does not point to any facts indicating a monetary loss Wyeth suffered as a result of the insider trading scheme.

Well, I mean, right. It’s a weird theory! I just kind of think that it is the theory.

Ben Miller reports

Pfizer Inc. reached a deal with the Securities and Exchange Commission in which the drugmaker would collect $29 million to end a long-running dispute over an insider trading action involving Steven A. Cohen’s former hedge fund.

Cohen’s SAC Capital Management paid around $602 million in 2013 to settle the SEC’s claims over an ex-employee who sold shares of Pfizer’s Wyeth LLC and Elan Corp.—a deal largely paid out to harmed investors and the US Treasury, leaving $75 million remaining in the SEC’s fair fund.

Pfizer was appealing a 2024 ruling from Judge Victor Marrero in the US District Court for the Southern District of New York that it would receive none of the fair fund total. The drugmaker argued the nonpublic information used for insider trading came from an employee of its subsidiary Wyeth, whose fiduciary breach entitled the parent company to the leftover settlement amount.

Makes sense, I mean, to the extent that any of this makes sense. SAC stole some drug-trial information from Wyeth, so it should have to pay back Wyeth.

How Blue Owl’s battles📰

First Brands Collapse

Janus Bidding War

dispersion trades

New Benchmarks

Markets Bond

Private Credit Worries

Warner Bros.

Red Lobster Clawback

If you'd like to get Money Stuff in handy email form, right in your inbox, please subscribe at this link. Or you can subscribe to Money Stuff and other great Bloomberg newsletters here 📰. Thanks!