Sometimes a public company has a controlling shareholder who wants to take it private by buying out all of the other shareholders, and that’s always messy. [1]  The controlling shareholder will to some extent be negotiating with herself: She will want to buy the company for a low price, but the company’s shareholders will want to get a high price, but she’s the controlling shareholder and can vote for the low price. There are standard solutions to the problem, but they are only partial solutions:

1. The company’s board of directors negotiates the deal with the controlling shareholder, and has fiduciary duties to the minority shareholders. It will generally appoint a special committee of independent directors to negotiate with her, and empower that committee to reject a deal that isn’t in the best interests of shareholders. But of course if the controlling shareholder is unhappy with the board’s negotiating position, she can fire the directors and elect new ones, which undermines their negotiating leverage.
2. The company can try to shop itself to other potential buyers, but any sale would have to be approved by shareholders, which means that the controlling shareholder can block any deal she doesn’t like. That undermines any other potential buyer’s incentive to make a bid.
3. If the minority shareholders are unhappy with the deal, they can always sue, but that’s sort of a roll of the dice.

In the past few months, I have written a few times about 23andMe Holding Co. as an illustration of these problems. 23andMe is a publicly traded genetic testing company that was once worth about $6 billion, but it has now fallen on hard times. Its founder, Anne Wojcicki, owns about 49% of the voting power of the stock, making her effectively a controlling shareholder. She offered to buy all the stock she didn’t own, to take the company private and fix its problems “outside of the short term pressures of the public markets.” But the board of directors, whose job was to find an “actionable proposal that is in the best interests of the non-affiliated shareholders,” didn’t think her offer was good enough.

But they were in a tough situation: They couldn’t force Wojcicki to offer more, and she could always fire them. So they all quit, leaving Wojcicki as the sole director. That was awkward! And she went and appointed a new board of directors, and the new board created a new special committee to negotiate with her, and she put in a new bid that was lower than her first proposal — just 41 cents per share, down from a previous bid of $2.53 — and the new board once again said no.

But they were still in a tough situation. What could they do? Well they came up with a pretty clever answer. Or perhaps I should say: They were forced into choosing a terrible answer that actually turned out to be pretty clever. But in any case, what they did was file for bankruptcy in March.

Filing for bankruptcy is generally a bad sign, but it has some advantages. One thing it does is buy breathing room from creditors. But another thing it does is put the company under court supervision: The company can still go out and try to find someone to buy its assets, but now a judge is involved and gets to supervise the bidding process and approve the final deal. If you are the board of directors, the controlling shareholder is effectively your boss, and you can’t really overrule her. But once you’re in bankruptcy, there is a new, higher boss — the judge — and maybe he will be on your side. “We have determined that a court-supervised sale process is the best path forward to maximize the value of the business,” said the board chairman.

We talked about this at the time, and what I said was: “Whoops!” Because one interpretation of this outcome was that Wojcicki was offering 41 cents per share, the board was holding out for more, but in fact the company was a melting ice cube and, by holding out for more, the shareholders ended up with zero. Bankruptcy is usually bad! Shareholders normally do end up with zero in bankruptcy. Wojcicki said that she would still try to buy the company’s assets, and I wrote:

I assume that her next bid will come out to $0.00 per share, but that is not investment advice and the stock was trading at about $0.89 at noon today so who knows.

Well, I was wrong and the shareholders were right. 23andMe found a buyer, and now:

23andMe has a path to a higher purchase price than the $256 million offered by biotech giant Regeneron after the genetic-testing company’s former chief executive pushed a bankruptcy court to reopen its sale process. 

23andMe was set to sell itself in bankruptcy to Regeneron before former CEO Anne Wojcicki bid $305 million after the auction ended through her recently founded nonprofit TTAM Research Institute. Wojcicki’s bid dwarfs her previous offer to acquire the company for $40 million just ahead of its March bankruptcy filing. …

Because 23andMe has a relatively small amount of debt on its balance sheet, its equity holders could walk away with a recovery, a relative rarity after bankruptcy. Its stock has rallied to a market capitalization of over $100 million.

There will be a new auction that might get a higher price, but the $305 million seems like a floor. The stock — again, in bankruptcy — traded at $4.91 per share at noon today, a roughly 1,100% premium to Wojcicki’s last pre-bankruptcy bid. As Julian Klymochko tweeted:

A phenomenal job done by 23andMe's board of directors that will become a case study on how to deal with a coercive and shareholder-unfriendly bid from a controlling shareholder.

A masterful bankruptcy filing led to a 663% increased offer from the company's founder.

You never really want to be on a board of directors where the clever strategic move is to file for bankruptcy, but sometimes it really is the right move.

The great arbitrage is that nobody cares about a hot startup’s profits, but everybody cares about its revenue. If you sell $5 billion of artificial intelligence widgets, that’s amazing: It means that you have found product-market fit, that people love your AI widgets, that you will be able to scale and achieve a leading position in the AI widget space. If you have $6 billion of research and marketing costs, that’s fine, that’s great, venture capitalists will love you, they will understand that you have to invest a lot now to build the business; the profits can wait.

Whereas if you sell $10 million of AI widgets and have $5 million of costs, that is a higher net income, but not as exciting to investors looking for a home run.

And so if you are in the business of selling AI widgets, you might go to your mom and say “hey mom would you buy $1 billion of AI widgets from me,” and she would say “I’d love to honey but I don’t have $1 billion,” and you would say “that’s okay I’ll just buy $1 billion of cookies from you, which will give you the money to buy the widgets,” and she would say “but I’d have to bake a lot of cookies,” and you would say “no just bake one, the cookies are so good that I’ll pay a billion dollars for one,” and she would say “okay but do you have $1 billion,” and you would say “no but it’s fine we’ll just net the transactions.” Or variations on that theme. We have discussed some of those variations, which play a crucial role in a lot of tech market bubbles. As long as investors are investing on revenue rather than profitability, someone will find a way to pay for revenue.

It would be very weird if there was none of this in the AI boom. Here’s the Financial Times on Builder.ai:

London-based Builder.ai collapsed last month after an internal investigation found evidence of potentially bogus sales, with revenues previously reported under [founder Sachin Dev] Duggal’s watch restated to just a quarter of previous estimates. ...

Now, interviews with former employees and documents seen by the FT suggest that Builder.ai is suspected of using a broader range of methods to inflate some of its revenues, including improperly booked discounts, tiny upfront deposits and seemingly circular transactions with key customers. …

Several ex-employees said it was commonplace for staff at Builder.ai to offer large discounts to customers while booking the full value of the contract under its sales.

One said they saw revenues based on these inflated prices presented in a company-wide meeting and a slide deck given to shareholders, who were seemingly unaware that the figures did not account for discounts. …

Former employees said that Builder.ai sometimes booked discounts as a marketing or sales cost.

There are also claims of circular transactions with customers. If you can find someone willing to pay you $1 for an AI widget, and you bill them $1 million and then pay them $999,999 out of your marketing budget, that’s better than just billing them $1. But also worse.

As I often write around here, right now the US stock market will pay $2 for $1 worth of Bitcoin. This is most famously demonstrated by MicroStrategy Inc. (or just Strategy), a company with a $60 billion pot of Bitcoins and a $118 billion equity market capitalization, [2] but we have talked about lots of other “Bitcoin treasury companies” (or other crypto treasury companies) that also trade at premiums to their Bitcoin stashes. [3]

I talk about this a lot, because it is weird and I have no good explanation for it. But if you think “this is weird and there is no good explanation for it,” you can do more than just talk about it. You can do something about it: You can bet that, if it’s weird and there is no good explanation for it, it has to stop. Specifically, you can buy $2 worth of actual Bitcoin and short $2 worth of Bitcoin treasury company stock. [4] If this anomaly stops, the Bitcoin treasury company stock will go down to $1, and you will make money. This is not investment advice! Not at all! This is a famous widowmaker trade — Strategy’s premium has been really persistent — but here’s Jim Chanos doing it:

“I haven’t seen an arbitrage like this in this size in years, years and years,” Chanos said during a recent interview on the Risk and Return podcast, referencing the trade he disclosed on CNBC in early May. …

Chanos, known for his prescient bet against Enron Corp. 20 years ago, emphasized that his current favored trade isn’t a wager against Bitcoin or Strategy, but rather a mirror of Saylor’s own playbook: selling equity and raising capital to buy more digital assets. The core opportunity, he said, lies in the divergence between Strategy’s market price and the company’s Bitcoin-adjusted book value.

“The fact of the matter is, this is a Bitcoin holding company,” Chanos said on the podcast explaining the premium dislocation. Buying Strategy shares at their current price of around $400, he said, is effectively equivalent to buying Bitcoin at about two times its value — “paying around $220,000 for Bitcoin that trades at $110,000. But the company is doing everything it can to close that spread, which is great — there’s a catalyst.”

Ahahaha he’s kind of right about that: The very biggest player in this trade, the trader who buys the most Bitcoin while selling the most Strategy stock, is Strategy itself. In theory, as Strategy sells stock to buy Bitcoin, that should bring down the premium (bringing up the price of Bitcoin and bringing down the price of Strategy). In practice, as I have written, even “selling as fast as it possibly can, MicroStrategy still can’t sell enough stock to close the premium.”

To be fair, I wrote that in December. Since then, Strategy has stopped selling much common stock; now it is selling weird convertible preferred shares (“perpetual strike” and “perpetual strife” preferred stock, which are less equity-sensitive), and its latest thing is a non-convertible 10% preferred stock (“perpetual stride,” why not). Strategy is not doing the simple arbitrage — sell MSTR, buy Bitcoin — anymore. On the other hand tons of other companies really are doing a related trade — selling their stock at a premium to buy Bitcoin — and maybe that will bring down the premium for everyone. Or maybe not. This whole thing is weird and I have no good explanation for it, which means that I have no good intuition for when it will stop.

Here’s a fascinating post by Eric Neyman from March that has been making the rounds recently. Polymarket, the prediction market, lists a contract on “Will Jesus Christ return in 2025?” It trades today at what is conventionally called a “3% chance”: You can bet about 3 cents on “Yes” and get $1.00 on Dec. 31 if Christ returns, or you can bet about 97 cents on “No” and get $1.00 on Dec. 31 if he doesn’t. 

Nothing in this newsletter is ever eschatological advice, but that 3% chance seems high. For one thing, over the last few centuries, the realized probability of Jesus Christ returning is considerably below 3% per year. Also, if Jesus Christ does return this year, are you really going to want to go to the online crypto prediction market to collect your gambling winnings?

So let’s say you think the actual probability is lower. Say you are completely confident that the actual probability of Christ returning this year is exactly 0%. How should you bet on that? What is the correct price of the No contract?

Well, the bet is that you put some money in now, and on Dec. 31, 2025, if you’re right, you get back $1. There’s a US Treasury bill that matures on Dec. 26, 2025, that has a yield of about 4.07%, meaning you can buy it for 97.7 cents today and get back $1 at the end of December. Surely you should not be willing to put more into a Polymarket bet today to get back $1 at the end of December. So 97 cents on the dollar — about a 5.2% yield — feels like a plausible price for this contract, if you’re sure you’re right. 

That is a function of interest rates, though. In a world of zero or near-zero interest rates, you might pay 99 or 100 cents on the dollar; a dollar in seven months would be worth as much as a dollar now. In a world of very high interest rates, you might pay only 93 or 95 cents today for a dollar in seven months.

If 97 is the correct price to pay for the No contract, what is the correct price to pay for the Yes? Again, let’s assume that you are completely confident that the correct probability is zero. An intuitive answer would be “zero dollars”: There’s no reason to pay 3 cents today to get back 0 cents in seven months. But there is an important no-arbitrage condition, which is that the price of the No contract and the price of the Yes contract sort of have to add up to $1 (ignoring fees and bid/ask spreads). From Polymarket’s documentation: “Holding 1 ‘Yes’ share and 1 ‘No’ share is equivalent to having $1.00 (minus trading fees). As a result, Polymarket allows you to merge these shares back into USDC.” USDC is the Circle dollar stablecoin, and it seems fair to assume that it’s worth exactly a dollar.

And so if the No contract traded at $0.97 and the Yes contract traded at zero, then you could buy one No and one Yes for $0.97, combine them, and get back $1 immediately (not in seven months), which is 3 cents of free money. You’d do this until you arbitraged the Yes contract up to its correct price, which is $0.03.

One (dumb, but funny) way to put this is that the “Yes, Jesus Christ will return in 2025” contract is a bet on interest rates going up: If short-term interest rates went up to 8%, the No contract would be worth about $0.95, so the Yes contract would be worth about $0.05 and you’d make a 40% profit.

Another (somewhat less dumb) way to put this is that the Yes contract is a way to bet on Polymarket discount rates going up. Those discount rates are related to short-term risk-free US dollar interest rates, but they’re not the same thing. Neyman writes:

The Yes traders are betting that the time value of Polymarket cash will go up unexpectedly: that other traders will be short on cash to place bets with, and will at some point be willing to pay a premium to free up the cash that they spent betting against Jesus.

Has this galaxy-brained trade ever gone well? Yes! In late October of last year — a week before the election — Kamala Harris was trading around 0.3% in safe red states like Kentucky, while Donald Trump was trading around 0.3% in safe blue states like Massachusetts. On election day, these prices skyrocketed to about 1.5%, because “No” bettors desperately needed cash to place other bets on the election. Traders who bought “Yes” for 0.3% in late October and sold at 1.5% on election day made a 5x profit! 

The point here is that it is intuitive and conventional to say that prediction-market prices directly represent probabilities. If there is some contract where Yes trades at 80 and No at 20, you say that (the market says that) Yes has an 80% probability. But to make that interpretation plausible, the market needs to make some design choices, particularly about the time value of money. If Polymarket didn’t allow you to combine a No and a Yes and cash out $1, then long-term contracts would trade at a discount to their implied probability: If an event had an 80% probability, the Yes contract might trade at 77.6 cents today (and the No at 19.4), [5] the prices would creep up over time as the contract approached expiration, and it would be sort of annoying to convert the prices back into probabilities. So Polymarket’s design creates a no-arbitrage condition that forces the total prices to add up to 100, for ease of converting into probabilities, but the result is that the No contract is not a pure bet on the probability but also an interest-rates arbitrage.

Elsewhere here’s a fun post from Dave White at Paradigm about “Multiverse Finance”:

This paper introduces Multiverse Finance, which splits the financial system into parallel universes so you can short the market today only if your favorite candidate is going to lose the next election.

A “verse” is a parallel universe where some event we care about has happened or will happen at some point in the future. ...

Today's universe, where the regular financial system exists, corresponds to the event containing the whole sample space.

We call a verse a "parent" to another verse if the child verse is a subset of the parent, meaning every outcome in the child verse is also in the parent verse. …

We call a set of child verses a partition of their parent verse if the child verses are disjoint (non-overlapping) and their union makes up the entire parent verse. ...

The high-level intuition when working with verses in Multiverse Finance is that owners can choose to push ownership down to a partition, or pull it up from a partition.

In other words, let’s say we have a verse V and some partition P1, P2, P3 of V.

If I own 1 unit of a given token T in V, I can chose to “push down” my ownership to the partition, meaning I give up my ownership of T in V and instead now own 1 unit of T in each of P1, P2 and P3. This is equivalent to how you can always take $1 and mint one YES and one NO token for a given prediction market. …

Conversely, if I own 1 of token T in each of P1, P2 and P3, I can choose to “pull up” my ownership of T to V, losing my ownership of it in each of those partition verses. This is equivalent to how you can take a YES and a NO token for a given prediction market and get back $1. 

Verses, like prediction markets, rely on some underlying oracle for resolution. … At the time of resolution, any verse that does not contain the observed outcome reported by the oracle immediately disappears, together with all of its state.

One thing you could imagine doing with this is solving the time value problems of prediction markets by paying conditional interest. You have 100 dollars, and you split them into 100 Jesus-Returns-Yes-Dollars and 100 Jesus-Returns-No-Dollars, you sell the Yes-Dollars to someone else for (a little) actual cash, and you deposit the No-Dollars at a crypto bank that pays you interest in No-Dollars. The contract resolves, and if Jesus does not return your No-Dollars — principal and interest — convert back into, say, 103 actual dollars (your original 100, plus interest). If he does return, the No-Dollars go poof, the bank owes you nothing and presumably whoever bought the Yes-Dollars is happy.

Elsewhere in prediction markets, here is a weird story about a contract on Manifold, a play-money prediction market, on “Who will be the prime minister of Canada after the next election?” “By the beginning of May, 2025, after the Canadian voters had handed the Liberal party a mandate to form a government, and Carney was all-but-formally the prime minister, the market settled at 99% odds for Carney.” And then shenanigans:

On May 4, 2025, the market creator, Peter Njeim, decided to embroil himself in these shenanigans. He swapped human-readable labels for the two options "Pierre Poilievre" and "Mark Carney"; the shares were still attached to their original, unique outcome identifiers so that traders who bought NO on the former could see "NO | Mark Carney" in their portfolios which was not in fact how they had traded! Almost immediately after swapping the labels, Njeim resolved the market to the outcome originally representing Pierre Poilievre, but which now displayed the label "Mark Carney.”

This kind of resolution meant that traders who chose the correct answer did not get their payouts.

Honestly swapping the labels is kind of elegant nonsense. It was eventually fixed, but a very levered (in play money!) bettor on the wrong side (called “Tumbles”) was “nuked,” leaving him with negative play-money balances and “potentially making a few select election markets less liquid and more directionally accurate.”

Anyway there is also a separate Manifold market about whether this story will be “covered by a ‘large’ podcast or news outlet,” with Money Stuff specifically mentioned as a qualifying outlet, so several readers sent it to me and disclosed that they had bets on that market. Incentives!

Bill Ackman Is Tweeting Through It. UK fintech Wise to switch main listing to New York. Citigroup lays off 3,500 tech staff in China. Allianz in Talks to Buy Private Credit Manager Capital Four. Boeing Agrees to Pay $1.1 Billion to Avoid Prosecution for 737 MAX Crashes. Reddit Sues Anthropic, Alleges Unauthorized Use of Site’s Data. Santander scraps plan to appoint executive under criminal investigation. Powerful new cannabis outstrips dangers of ‘Woodstock pot.’ “Bbq Into Bankruptcy.” 

If