Two popular themes around here are: - Everything is securities fraud, and
- Private markets are the new public markets.
These themes are related. “Everything is securities fraud” means mainly that, if a bad thing happens at a public company, some enterprising shareholder class action lawyers will sue it for securities fraud. “You didn’t tell us that the bad thing would happen, so we bought stock, then it happened, and our stock went down, and we were defrauded,” etc. We talk about these lawsuits all the time because there is something funny in the accumulation of them. The other day I mentioned a lawsuit against a public rocket company whose rocket exploded, “thereby injuring investors.” Not because the investors were hit by rocket debris. Because the stock went down. “Private markets are the new public markets” means mainly that lots of large fast-growing capital-intensive household-name tech companies, which 20 years ago would have gone public to raise money to fund their growth, now stay private much longer. There is a lot of capital available in private markets these days, and big companies like SpaceX and OpenAI can repeatedly raise billions of dollars at 12-digit valuations, do frequent tender offers to give employees liquidity, and otherwise get many of the financing benefits of being public without the hassles of being public. So they don’t go public. People worry about this: If the big exciting fast-growing companies stay private, then ordinary investors won’t be able to invest in the big exciting fast-growing companies. Ownership of big and fast-growing slices of the economy will be reserved for rich institutions, and ordinary individuals will be frozen out. The big cool companies stay private because (1) now they can and (2) there are hassles of being public. What are the hassles of being public? Various things: Public scrutiny, a fluctuating stock price, short sellers, activists, hostile takeovers, extensive disclosure requirements. But “everything is securities fraud” is one of the hassles, at least. Where does it rank on the list? I mean: - The actual expected cost, to most public companies, of being sued for securities fraud if they do something bad and their stock drops, is not that high.
- It is aesthetically pretty annoying.
- A lot of the expense of having a public-company accounting and disclosure apparatus does arise from the risk of being sued for securities fraud. You have to make sure that all of your financial and business disclosures are accurate, because if they aren’t, someone will sue. This is not generally true of private companies: Private companies aren’t supposed to lie to their investors either, but the investors understand that the companies have better things to do than have lawyers and auditors scrub every communication. “Hey here are our rough preliminary financials for October, could change but we wanted to give you a sense of what’s going on” is a reasonable thing for a private company to say to its venture capital investors and an insane thing for a public company to say to its shareholders. [1]
So “everything is securities fraud,” or at least “public companies get sued when their disclosures are wrong,” might not be at the top of the list of hassles keeping companies from going public, but it’s not at the bottom either. If, like a lot of people including Securities and Exchange Commission Chairman Paul Atkins, you think it is bad that companies are increasingly staying private, you might want to address that problem by removing some of the hassles of being public. If you think that getting sued by shareholders is one of the more salient hassles, you might try to remove it. You might say “from now on, if a company’s rocket explodes, or if its CEO does some sexual harassment, or for that matter if its financial statements are wrong, shareholders can’t sue.” That would make being public less of a hassle, and might lead more companies to go public. Obviously there is a downside! The downside is not so much that more companies would do exploding rockets or sexual harassment: There are other deterrents, besides securities fraud lawsuits, for those things. The downside is that more companies would fudge their financial statements. There are other deterrents for that too — if you fudge your financial statements, the SEC might come after you, or prosecutors for that matter — but, unlike in the sillier “everything is securities fraud” cases, the main victims of financial misstatements really are shareholders, and if they can’t sue then probably no one will. [2] The SEC can’t actually say “from now on shareholders can’t sue their companies,” but last month it did what it could. From a DLA Piper memo: The US Securities and Exchange Commission (SEC) recently issued a policy statement reversing its long-standing practice concerning mandatory arbitration clauses in connection with initial public offerings. Released on September 17, 2025, this policy approved by the SEC Commissioners along party lines states that the SEC will no longer object to the acceleration of a company's registration statement [3] solely because it contains a provision requiring mandatory arbitration of investor claims arising under federal securities laws. The SEC will instead focus on the adequacy of the disclosures regarding the arbitration provision. SEC Chair Paul Atkins characterized this move as an effort to make IPOs more attractive to companies. Previously, the SEC viewed any provision that waived or impaired the protections of the federal securities laws as contrary to public policy and would not accelerate the effectiveness of a registration statement if the issuer required investors to arbitrate such claims. The SEC and other commentators believed that mandatory arbitration of shareholder claims would effectively put an end to securities class actions because they would require each investor to arbitrate their individual claims, which would be impractical for retail investors that suffered relatively small individual losses. That is: In theory, a company’s charter or bylaws could contain a provision saying “if the shareholders want to sue us, they have to bring the claim in private arbitration, not in a federal court.” An arbitration might be more friendly to the company, and since it wouldn’t be public it wouldn’t set a precedent. More importantly, the company could require individual arbitration, making shareholder class actions impossible: If the stock goes down 10%, instead of a law firm bringing a billion-dollar claim on behalf of every shareholder, each shareholder would have to bring her own arbitration case saying “I lost $18” or whatever. Few shareholders would. In the past, the SEC thought this was bad — because it could prevent shareholders from suing for fraud — and would refuse to let companies go public with mandatory arbitration provisions. Now the SEC thinks this is good — because it could prevent companies from being sued for fraud — and will let them go ahead. Here is Atkins’s statement, saying that allowing mandatory arbitration is “among the first steps of my goal to make IPOs great again,” to “make being a public company an attractive proposition for more firms by eliminating compliance requirements that yield no meaningful investor protections.” Here is a dissenting statement from SEC Commissioner Caroline Crenshaw, saying that actually allowing shareholders to sue for fraud is kind of a meaningful investor protection. Honestly they both make some good points. Does this matter? It is not clear that those provisions would be legal under state law, at least in Delaware, though probably it’s fine for Texas corporations. But if it really is allowed now, will companies take advantage of it? Will big companies start going public with provisions in their charters saying “hey if we do fraud you can’t sue us”? I don’t know. But … yes? In theory you could imagine that big institutional shareholders might push back on that sort of provision: Big institutions that care about governance might be suspicious of companies that go public saying “hey if we defraud you you can’t sue.” But I’m not sure. For one thing, my impression is that big institutional investors will complain about bad governance but ultimately put up with it. For another thing, I’m not sure that securities class actions are net beneficial to institutional investors. “If this practice became widespread, it would effectively end shareholder lawsuits,” writes one insurance blogger, and that does seem right. It’s possible that within a few years nothing will be securities fraud. Over the years, I have spent a lot of time making fun of everything being securities fraud; I think it is a bad way to regulate public companies. I sympathize with the notion that, if fewer things were securities fraud, it would be easier and more pleasant for companies to be public. But I do think that, like, securities fraud is securities fraud. You shouldn’t fake your numbers, and if you do you should expect to get sued. I’m not sure that nothing being securities fraud is an improvement. Two other popular themes around here are: - Lol Citigroup Inc. sent someone an extremely wrong amount of money, and
- Blockchain blockchain blockchain.
The first theme came up, for instance, when Citi meant to pay $7.8 million of interest on a Revlon Inc. loan that it administered, and accidentally paid $900 million instead. It came up again when Citi meant to sell $58 million of European stocks and instead sold $444 billion of those stocks. And then again when Citi “credited a client’s account with $81tn when it meant to send only $280.” And then also when Citi “almost shifted about $6 billion to a customer’s account by accident after a staffer handling the transfer copied and pasted the account number into a field for the dollar figure.” How we all laughed. “The best job in tech,” I wrote, “is designing the user interface for the payments department of Citigroup Inc. As far as I can tell the job is just doing pranks.” The second theme comes up when banks, or people looking to disrupt banks, talk about how with an immutable distributed public ledger of transactions, finance will be more secure and transparent and modern. Maybe! Here’s a stablecoin fat finger: Fintech giant PayPal made an interesting move [yesterday], minting $300 trillion of its PYUSD stablecoin on the Ethereum blockchain, briefly making it the largest stablecoin in the world by a landslide. Paxos, which issues PYUSD, minted the supply on Wednesday afternoon and burned it shortly after, confirming it was an accidental mint. PYUSD, which is backed by U.S. dollars, is the eighth-largest stablecoin in DeFi with a $2.6 billion valuation. It’s unclear whether Paxos intended to mint a specific amount, but any substantial increase in supply would see PYUSD surpass BlackRock’s BUIDL USD and potentially World Liberty Financial’s USD1. Like some (not all!) of Citi’s oopsies, this is firmly in the no-harm-no-foul, accidental-internal-crediting category: Paxos, which issues PYUSD stablecoins on the Ethereum blockchain, created 300 trillion new PYUSD by accident, but it didn’t send them anywhere, and it uncreated them 22 minutes later. [4] In general, and by law, each PYUSD is supposed to be backed one-to-one by US dollar cash equivalents, but that is a fact about Paxos’s corporate standards and US stablecoin regulation, not a fact about blockchains. Paxos doesn’t need to have $300 trillion in the bank to push the button that mints 300 trillion of PYUSD. For 22 minutes yesterday, Paxos was just holding $300 trillion of assets that it had created out of thin air by accident. The price of PYUSD did not budge, varying between $0.9995 and $1.00015 while this was happening. Presumably if Paxos had announced “hey we created $300 trillion of naked PYUSD, want some?” then the price would have collapsed: The 2.6 billion actually existing PYUSD are worth a dollar each because they are backed by $2.6 billion of dollar assets; 300.0026 trillion PYUSD backed by $2.6 billion would be worth considerably less. But by the time anyone cared it was fixed. In a good financial thriller, the person who accidentally created that $300 trillion would have sent it all to herself, sold it as quickly as possible and cleared, you know, at least enough to buy a first-class ticket to a non-extradition country. (The market price for $300 trillion of fat-fingered PYUSD is nowhere close to $300 trillion, but it’s not zero either, and you don’t need to get particularly close to $300 trillion to have a very good day.) But in the real world they just deleted the extra $300 trillion and went about their day. “Hedge funds are a compensation scheme masquerading as an asset class,” is the famous line, but that isn’t really true anymore. The compensation scheme, famously, is (or was) 2 and 20: Hedge funds traditionally charge their investors 2% of assets each year as a management fee and 20% of returns as a performance fee. The “asset class” was, you know, there were like a dozen important hedge fund strategies, so institutional allocators could think “I want 5% of my money in global macro so let me call some global macro funds.” Each hedge fund manager was trying very hard to get high and uncorrelated returns, but there was some empirical correlation among managers in any category. The modern multistrategy multimanager “pod shop” hedge funds are different. For one thing, they are a different asset class: Each of those funds combine many different “pods” — essentially individual hedge funds with their own portfolio managers, operating quasi-independently with a central management allocating capital to them and overseeing their risks — into one return stream, with the result that a pod shop’s returns do not look like “long/short equity” or “global macro” or “managed futures” or whatever, but rather like their own aggregated thing, “pod shop hedge fund returns.” What are those returns? What is the asset class? Well, I think of pod shops of being in sort of a service business, providing liquidity and price discovery to capital markets. (I think of this business as being analogous to some of the traditional trading businesses of investment banks.) That service is valuable, and so it has some expected return, and the pod shops broadly earn that return. There is risk and sometimes they lose money, but they offer a lot of different somewhat uncorrelated services so most of the time they get paid. It is a kind of business income. Investing in pod shops is not quite the same as investing in bank stocks, but it’s kind of related. Meanwhile the compensation scheme is also quite different. The compensation scheme is: - Investors give the pod shop their money.
- It uses the money to do its business and generate returns.
- Out of those returns, it pays its expenses, which include rent and photocopying and data feeds and taking investors out to dinner to solicit their money, but which mostly include paying its portfolio managers for generating returns.
- Those paychecks will generally be commensurate with the returns — a portfolio manager who generates $1 billion of trading profits will get paid more than one who generates $10 million — but not in some universal mechanical way. It’s not like every portfolio manager in the industry gets paid exactly 20% of the money she generates. A PM who loses money might still get a paycheck if the firm wants to keep her. [5] A PM who was recently hired in a competitive job market might initially get some sort of elevated payout to incentivize her to switch jobs. A PM who generates $100 million in her own pod but who makes money for the firm in other ways might get paid more than the general rate. Etc. The pod shop is a big sophisticated company in a competitive market for labor, and it will pay its star employees what it needs to pay them, not the mechanical result of a generic formula.
- The management company and senior executives of the hedge fund also get a cut.
- The investors get back what is left over.
That is, the compensation scheme is not “you give us money, we make more money, and we give you 80% of the profits.” It’s “you give us money, we make more money, we take out whatever we need to keep sustainably making that more money, and we give you the rest.” There’s no guarantee, no fixed cut of the return. You sign up for this deal because you think that the hedge fund will make more money for you (after whatever fees it feels like charging) than you’d get elsewhere, and you stay signed up as long as that keeps happening most of the time. If one year the hedge fund tells you “your returns are 2% because all our PMs had to buy yachts,” or for that matter “your returns are 2% because all our PMs did a bad job of picking stocks,” you might say “hmm I guess but this is not what I wanted.” If the next year it tells you the same thing, you will pull your money. The compensation scheme is “we made $X of trading profits, we need to pay the PMs at least $Y to attract and retain good PMs, and we need to return at least $Z to the investors to attract and retain the investors.” If X - Y > Z, everyone is happy. If X - Y < Z, somebody is unhappy, which is not good for the long-term health of the hedge fund. In this, too, the pod shops are like banks: They have some cost of capital, and if they do not earn their cost of capital over the medium term they will not be able to keep their capital, but it is all a bit subjective and negotiable. And they have some cost of labor, and if they do not pay their employees their market value they will not be able to keep their employees, but that too is subjective and negotiable. The Financial Times reports that the cost of labor is 25% now: Top traders at some of the biggest hedge funds are taking home almost a quarter of the profits they make for investors, according to a Goldman Sachs report, as the likes of Citadel and Millennium extend their hold over the industry. A war for talent between multi-manager hedge funds has pushed up pay in the industry, with firms luring traders with packages that can be worth more than $100mn. Goldman said that managers at the highest paying firms had this year secured payouts worth 24.5 per cent of the profits they made for investors, compared with 22 per cent in 2022. … Whereas traditional hedge funds would usually charge a 2 per cent management fee and 20 per cent on profits made for investors multi-manager hedge funds pass their expenses through to investors, enabling them to invest more in technology and traders to extend their advantage over rivals. Under the model, costs including client entertainment, portfolio manager bonuses, and technology are charged directly to investors rather than incurred by the fund’s managers. Also “the total managed assets across the 57 hedge funds surveyed by Goldman jumped by 16.1 per cent in 2025 from the year before, driven both by performance and new capital,” so apparently they are earning their cost of capital too. We talked yesterday about the possibility that a trader on Polymarket might have successfully predicted that María Corina Machado would win the Nobel Peace Prize, hours before it was announced, by scraping the Nobel website and seeing it update with information about Machado. I was pleased, writing: The point here is not so much “perhaps this trader did not have inside information.” The point is: Perhaps someone was incentivized to build, if not quite “a statistical model using only public information,” at least, like, an automated feed of public alternative data on the Nobel Peace Prize winner. The point is that doing informed but non-insider trading on the Nobel Peace Prize winner is (1) possible and (2) possibly lucrative enough that people will do it. In a dumb, degenerate, not-especially-fundamental way, but still. The point is that the Nobel Peace Prize prediction market is working the way it is supposed to: It incentivizes people to find out information and incorporate it into prices. This prompted Serge Ravitch to email me to reminisce (lightly edited): Back in 2008, there was a brief splash of interest around how [once-popular prediction market] Intrade picked up on the fact that Sarah Palin would be the [Republican vice presidential nominee] six hours or so ahead of the actual announcement. [6] The story has been told several times on random prediction market podcasts listened to by fifty people but I doubt it’s ever hit anyone outside that circle, so, in short: Domer (@domahhhh) and I were trying to figure out the pick and going back and forth with one another looking for clues. At 3 AM EST, I hit upon a random Free Republic post from some guy who had seen a private flight go from Alaska to AZ and back on the 2008 equivalent of flighttracker. We spent the next half hour frantically reading her Wikipedia bio and failing to find any other flights from would be contenders (Tim Pawlenty, for example, had not budged from Minnesota). By 4 AM, we had bought out every possible share of Sarah Palin available online at every political betting market and the rest is history. “I think we cleared, like, $10,000 each or something,” he adds, so this is not quite institutional-level incentives, but the point is, yes, prediction markets do have some track record of incentivizing people to incorporate quasi-public but not especially widely disseminated information into prices. The Little-Known Crypto Powerhouse Behind Billions in Trading. Griffin Says GenAI Fails to Help Hedge Funds Beat Markets. Inside the Credit Card Battle to Win America’s Richest Shoppers. ‘ |