| A good corporate finance rule of thumb is, "if you run a company, and people are paying much more for your stock than it is worth, you should sell them as much of it as you can." This rule can be hard to implement, though, for a few reasons: - Companies are complicated and you might genuinely not be sure how much yours is worth, making it hard to know if people are overpaying.
- You run the company and so are naturally optimistic about it; it would be odd — not unheard-of, but unusual — for you to think that the stock is overpriced.
- It just doesn't come up that much? Like, public shareholders are not usually way more optimistic about a company's prospects than its managers are; it is rarely obvious to managers that their stock is overpriced.
- Even if it does happen — even if investors want to pay way too much for your stock — that might change if you take advantage of it. If you go around selling loads of stock to meet investor demand, the investors might say "huh the company is selling, maybe management thinks it's overvalued, and they probably know better than we do," and the stock might collapse.
These problems are significant enough that I never really used to think about this. Like, "if people want to overpay you for your stock, sell it to them" is nice if you can do it, but it just doesn't seem relevant to most companies' lives. But the rise of meme stocks in recent years has changed that, and now I constantly find myself writing "well if people want to pay crazy prices for your stock you should sell them the stock." What distinguishes a meme stock is probably something like "retail investors enthusiastically buy it at prices that its own executives think are nuts." When the meme-stock wave first happened, a lot of companies didn't sell stock, because the whole thing was too novel and nuts, but over time they adapted, and now GameStop is ready with an at-the-market stock offering whenever Keith Gill tweets. Or a corollary of this rule is something like "if you are in bankruptcy and your stock is worthless, but people still want to buy it, sell it to them." I don't think this ever occurred to anyone in history before 2020, but now I talk about it with some frequency. Or we talked earlier this year about the Destiny Tech100 fund, a publicly traded closed-end fund that (1) owned shares of hot private companies worth about $52.6 million and (2) traded, for a while, at a market capitalization of about $1.1 billion, almost a 2,000% premium to its net asset value. What is nice about an investment fund, as opposed to an operating company, is that everyone can see its "actual value"; nobody (I hope) thought that Destiny's $52.6 million pot of shares was actually worth $1.1 billion. But Destiny was the way that retail investors could get access to those private shares, so they overpaid. "Sell stock," I screamed through my computer at Destiny: selling more stock at a 2,000% premium, and using the money to buy more private shares, would be enormously accretive and also help push the stock to more reasonable levels. On the other hand, Destiny was a little limited by its ability to source more private shares; raising more money and not having anywhere to invest it would be kind of bad. Whereas: MicroStrategy Inc. has hired banks to help it raise $42 billion through the sale of new shares and fixed income to buy more Bitcoin after a flurry of deals over the past year. The enterprise software maker, co-founded by Chairman Michael Saylor, hired a handful of banks to sell stock through a so-called at-the-market offering that could net it $21 billion while lining up potential sales of fixed-income securities that could bring in the same amount. The company has been raising billions over the past year through the sale of convertible senior notes as well as shares that its bankers can sell into the market to add to its cryptocurrency stockpile. ... The magnitude of a $21 billion at-the-market program trumps recent offerings from the company and would make sales in recent months for meme stock GameStop Corp. look tame.
Yeah I mean Boeing Co. raised $21 billion of equity this week and that was "one of the largest [stock offerings] ever by a public company." Twenty-one billion dollars is a lot. But it's not like it would be hard to spend: Bitcoin is pretty liquid and MicroStrategy can easily source as much of it as it wants. I don't have the faintest idea how to explain this. MicroStrategy is primarily a pot of Bitcoins, though it also has a software business that is grudgingly acknowledged in its earnings release. The pot of Bitcoins is worth about $18 billion, according to that earnings release, and is funded partly with debt; MicroStrategy reports about $4.2 billion of long-term debt. This suggests a net asset value of, you know, $14 billion? Plus or minus 50%? Its market capitalization is about $50 billion. Would you say that MicroStrategy is a "listed closed-end investment fund that trades at a 300% premium to its net asset value"? Well, no, no, I would never say that. But on the other hand yesterday it did announce "a strategic goal of raising $42 billion of capital over the next 3 years, comprised of $21 billion of equity and $21 billion of fixed income securities," to invest in Bitcoin. And, yes, if I ran a company in … these circumstances … my main strategic goal would also be "sell tons of stock." That's a good strategy! Broadly speaking the way the modern world works is: - People get together and agree on some rule for how something should work, and
- Then somebody types this rule into a computer that will implement the thing.
So a bank will sign a derivative contract with a client, and someone at the bank will type the terms of the contract into the bank's computer systems, and those systems will send out cash to the client (or demand cash from the client) based on the terms of the derivative. And every so often the bank will type the terms in wrong and the client will get less money than it is supposed to. Sometimes it will notice and the error will get fixed, and sometimes it won't. And I suppose you could ask the metaphysical question "is the actual trade the contract they signed, or the representation that the bank put into its computers?" Or we talked last month about a guy who exploited Spotify's streaming revenue calculations to extract millions of dollars in ways that Spotify did not intend. Spotify's deal with holders of music rights consists of (1) a long written document containing the terms of service, which no one reads, plus (2) the actual algorithm that it uses to allocate payments to those holders. Spotify, and US prosecutors, will tell you that both of those things are essential parts of the deal. But you can sort of understand why the guy might have thought "no, the real deal with Spotify is what its computers do; this other stuff doesn't matter unless it is implemented in the code." Or: There are lots of off-by-one errors in the world. Some things have to be at least X, while other things have to be more than X. Some things have to happen on or after the one-year anniversary of some starting date, while other things have to happen after the one-year anniversary. If a contract or law specifies "after ___," the computer program implementing it should also specify "after ___." Do some of those programs instead say "on or after ____"? Why yes they do! Here's a story about 79 fake divorces in the UK: Dozens of divorce applications were wrongly approved due to a computer error, the high court has heard. A hearing in London on Wednesday was told that 79 divorces were incorrectly approved after the applications were submitted a day before the law allows. An online system failed to detect that the submissions came a year after marriage, when the law only permits divorces from a year and a day. Lawyers for the justice secretary have asked the court to rule that the divorces are "voidable" as opposed to "void", meaning they would still stand, claiming that voiding the divorces would have "highly unfortunate and highly unwelcome" consequences for the couples concerned. The court was told that between April 2022, when the validation system of the online portal was introduced, and April 2024, a total of 96 divorce applications were made on the first anniversary of marriage. Of these, a final divorce order was granted in 79 cases.
The law says that you can get divorced after the one-year anniversary of your marriage, but the computer program said that you can get divorced on the one-year anniversary. People who were in a hurry submitted applications early, and rather than bouncing them according to the law, the computer granted them according to its own logic. But code is not actually law in the UK, so those people were not actually divorced, even though the computer said they were. Which is very inconvenient, because "some of the divorcees had either since remarried or were planning to." They thought they were divorced! The computer said they were! Why would the computer be wrong? Not that this is all the computer programmers' fault. Sometimes the actual contract doesn't reflect the deal that you thought you had, in ways that seemed minor and technical at the time and turned out not to be. If a company makes money, the stockholders can benefit in one of two ways. The company can give the stockholders the money, in the form of a dividend: Then the stockholders have cash, plus their stock. Or the company can keep the money and invest it itself, hopefully in something productive: Then the stockholders have no cash, but they have stock, which ought to be worth more by, in expectation, roughly the amount of the dividend. And so in theory a company could choose between, say, paying a $10 dividend and being worth $20 per share, or paying no dividend and being worth $30 per share. Shareholders ought to be roughly indifferent between these two choices, but holders of stock options will not be. A call option of the form "you can buy this stock at $15 per share" is worth $15, if the stock is worth $30, but only $5, if the stock is worth $20. And the holder of the call option doesn't get the dividend, so she is not indifferent: She would much rather the company keep the money (and have a higher stock price) than pay out a dividend (that she doesn't get). And so it is common for options contracts to contain dividend adjustments of roughly the form "if the company pays an $X dividend, the strike price of the option is reduced by $X," or some more complicated calculation. Common but not universal: Not every holder of every call option bargains for this protection, and in particular normal US listed options do not generally adjust for regular dividends. But sometimes you just forget? Here's a strange story from the Wall Street Journal: When Unit Corp. was in bankruptcy, the oil-and-gas driller made a deal with employees who held its stock in their 401(k) retirement accounts. In return for supporting the company's restructuring plan, they would receive stock warrants to share in its future growth. The warrants, designed like call options, were modified just as Unit's bankruptcy plan went into effect in September 2020 to remove a key protection for holders. The subtle modification wasn't made public until a week later. It went unnoticed for years among warrant holders, some of them said. Unit has since recovered from its Covid-era doldrums and returned some $400 million in capital to investors, including roughly $145 million to a hedge fund affiliated with its chief executive, according to court records and securities filings. The warrants, however, are now nearly worthless after the last-minute modification, touching off an ongoing legal battle for the company.
Basically the warrants in the original bankruptcy plan had an anti-dilution adjustment for dividends, but the final official version "eliminated the 17-word anti-dilution provision." Starting in 2023, Unit has paid big cash dividends to shareholders, making the warrants almost worthless. That is rough on the warrant holders. Also, though, I said above that in general shareholders should be indifferent between a $10 cash dividend and $10 re-invested in the company, but that does change if 12.5% of the company belongs to warrant holders who don't have an antidilution adjustment. With the warrants, $10 invested back in the company is worth about $8.75 to shareholders and about $1.25 to warrant holders; $10 paid out as dividends goes entirely to the shareholders. So that's where the money went. "There is no place in Trafigura for wrongdoing," Trafigura Group Chief Executive Officer Jeremy Weir said in a statement yesterday, but there was, and that place was Mongolia: Trafigura Group is preparing to take a $1.1 billion hit after discovering what it suspects was fraud involving employees in its Mongolian oil business. ... The company found that some of its employees in Mongolia had manipulated data and documents to inflate the amount of money it paid out, as well as deliberately concealing overdue debts over a period of about five years, Trafigura said in a statement on Wednesday, confirming an earlier report by Bloomberg News.
If you are a global commodity trading firm, you will have a lot of employees in global trading hubs, and you will have reasonably good incentives and ability to keep an eye on them and make sure they're not just stealing your money. And then you will have tiny handfuls of employees in a lot of far-flung commodity-rich locations, and their incentives and monitoring will be different. Happy Halloween, here's a great sentence: Wealthy Americans worried about who will win the White House are preparing to accelerate a huge transfer of wealth to their heirs by using large estate-tax breaks set to expire at the end of next year.
You know what that means, right? If you're a wealthy American, you've got 14 months to die cheaply; after that dying is more expensive. Presumably you're a wealthy American because you are good at optimizing your financial returns so, you know, something to think about. I am kidding, I guess; it's also a gift tax, so you can pass on some of your estate while you're alive: Those breaks came in the 2017 Republican tax law, which doubled the amount the wealthy could pass on without paying the estate and gift tax. Rich Americans can move about $14 million out of their estate by next year to avoid a 40% tax when they die. The limit reverts to about $7 million in 2026, creating a use-it-or-lose-it exemption as the "tax cliff" deadline nears.
Still. I like to quote Michael Graetz's two laws of tax, which are (1) it's better to make more money than less money and (2) it's better to die later than sooner. These laws are good rules of thumb, for taxes and life, but they are not always true. (For taxes.) Sometimes there are tax incentives to die sooner, and people respond to incentives. The classic 2001 paper is "Dying to Save Taxes: Evidence from Estate Tax Returns on the Death Elasticity," by Wojciech Kopczuk and Joel Slemrod, examining "the temporal pattern of deaths around the time of changes in the estate tax system periods when living longer, or dying sooner, could significantly affect estate tax liability" and finding "some evidence that there is a small death elasticity, although we cannot rule out that what we have uncovered is ex post doctoring of the reported date of death." Here though I guess you just give your heirs $14 million next year to minimize taxes when you die many, many years from now. Did Elliott Management commission hackers in its bond fight against Argentina? No, says Elliott, but on the other hand Jay Newman, who led that fight for Elliott, did write a novel featuring a hedge fund manager who deals with hackers, so who knows: Prosecutors want to know what is fiction and what is reality. Their globe-spanning probe is scrutinizing, among other things, the relationship between [Israeli private investigator Amit] Forlit and Newman. At the heart of that part of the case: Did an alleged hacking operation set up by Forlit aid Elliott's victory over Argentina?
Well, if you retire from working at a hedge fund with a tough reputation, and then you write a global espionage thriller about a hedge fund manager, you are taking certain risks. I've read Newman's novel, Undermoney, and I have to say that getting suspected of hacking is not the biggest risk he was taking in writing it: Newman has consistently denied any link between [Elliott manager Paul] Singer and the psychopathic hedge-fund manager in the novel, and told the New Yorker in an interview that the book further departed from reality when his editor suggested it needed to be much shorter—but include more sex. He said he reluctantly agreed on both counts. … The book sparked anger at Elliott, up to Singer himself, employees told people outside the firm. He made it clear: Staffers were not to bring the book into the office or discuss it with clients.
Yeah "the psychopathic hedge fund manager in my novel has no relation to my actual former boss at a hedge fund" is not an easy sell. It seems implausible to me that Albert Einstein ever said "the most powerful force in the universe is compound interest," but "ooh the magic of exponential compounding ooh ooh" is a staple of a certain sort of financial-literacy education, using examples both relatable ("if you save $100 a month you'll have enough to retire") and not (doubling grains of rice on a chessboard). Here is a new classic of the genre: A Russian court has slapped a $20 decillion fine on Google for blocking 17 YouTube channels belonging to local TV networks, RBC reports. The issue dates back to 2020 when YouTube blocked Tsargrad TV after the US imposed sanctions against its owner. At the time, the court had ordered a daily fine of 100,000 rubles ($1,025) per day, with the total amount set to double every week the fine was unpaid. Things escalated after YouTube blocked more channels following Russia's invasion of Ukraine in 2022. A total of 17 channels sued Google, and the court's latest order was passed after calculating the fines for each.
For fun I plugged "$1,025 doubling every week for four years" into Excel and the answer is 4 x 10^65, which is way way more than $20 decillion (2 x 10^34), so I guess they showed some mercy. Some of the headlines say that the fine is "more than the entire world's GDP," which is a weirdly inadequate comparison; here's a better one: In a post on X, Nigel Gould-Davies, senior fellow for Russia & Eurasia at the International Institute for Strategic Studies, described the sum as "an insane number," explaining it was equivalent to "1.9 x 10 to the 15 times greater than current global GDP." "About 5 x 10 to the 12 days have passed since the start of the universe," he wrote. "So even if Google gave Russia everything the world produced this year, every day since the universe began, it would only have paid about 3% of this fine."
Yes but see Google's artificial intelligence research might be worth … no I'm kidding this fine is purely for comic effect, just "get out of Russia and stay out." Apparently whatever assets Google has in Russia have been seized, but they come to something like $100 million, so I guess Google will end up paying 0.000000000000000000000000005 cents on the dollar. TripAdvisor Corporate Relocation Case Hinges on Delaware's Value. "They were sort of the beta of alts." KKR, Capital Group Partner to Debut Funds for Wealthy Investors. KKR's $20 Billion Goal Tests Investor Appetite for Buyout Funds. Sotheby's Gets a $1 Billion Financial Lifeline in Deal With Abu Dhabi. Billionaire Wes Edens' One-of-a-Kind Gas Export Plan Hits Turbulence. Kalshi's daring bet on election betting is paying off — to the tune of $100 million. $2 billion marina development aims to turn Fort Lauderdale into 'mini Monaco.' Private Equity Hipsters Are Coming for Your Favorite Apps. "One of Silicon Valley's youngest—and poorest—angel investors." Starbucks Is Bringing Back the Condiment Bar. How 4 People Are Spending Their Juul Settlement Money. 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! | 
No comments:
Post a Comment