Money Stuff: Fake Trading Is Hard Work

A stylized fact of financial markets is that there are some risk-loving retail traders who like to make highly levered bets on stocks, currencies, commodities and crypto, and in the medium run almost all of them will lose almost all of their money. The bets are of the form “if Thing X goes up 2% I double my money, but if it goes down 2% I lose all my money”; this works for a couple of days and is fun, and then they lose all their money. This trade seems to have a ton of very robust negative alpha, which makes it very valuable to take the other side.

How do you take the other side? The most obvious answer is “literally take the other side of their trades”: Buy what they’re selling and sell what they’re buying. This isn’t bad, and it is in some approximate sense the business of retail market makers in the stock and options markets. It’s not perfect: A lot of the retail traders’ negative alpha comes from their leverage (it’s not that they never pick stuff that goes up, but rather that one losing trade wipes them out), so just trading against them does not make you money as quickly as they lose it.

Another promising approach is “take the other side of their leverage”: They put up 5% of the money to trade, you put up 95%, the asset goes down 2%, you liquidate them and keep the extra money. [1] In the general case this seems like a lucrative business, though it does have some risk (if the asset goes down 10% you lose money). 

You can combine the two approaches: You lend them the money to trade, and you trade against them. You are on the other side of every aspect of their leveraged trading, meaning that you are purely on the other side of their negative alpha, meaning that you capture lots of positive alpha. (You do still have a lot of risk, though: Occasionally they will win big on a levered bet, and you will owe them a lot of money.)

This business model — lending people money to trade, and taking the other side of all of their levered trades — is traditionally called a “bucket shop.” You can find, uh, aspects of this model in some aspects of some modern crypto exchanges. Here are — in increasing order of cynicism —  columns from me, Patrick McKenzie and John Hempton about those aspects. 

There are other, simpler, safer, fee-based models: You set up the exchange where the retail traders trade, or you broker their trades on a riskless basis, and you charge a fee for each trade. (Or you combine that with trading against them: We talked once about a broker that sent its customers’ trades to exchanges, but also did its own trades on the other side of “its clients who were regularly unprofitable.” Mostly a fee-based agency business, but if you know who your most inept customers are, you might be tempted to use that signal.)

None of these models quite capture the essential value of the bad retail traders. The problem with the bad retail traders is not “they make bad trades” or “they take too much risk.” The problem is more like “they keep flipping coins until they lose all their money”: Each individual trade might be fine, but in the medium term they face gambler’s ruin. If you take the other side of each trade, you have a lot of risk: They have a coin flip’s chance of making money at your expense. What you want is to take the other side of the whole sequence: They come to you with money, they keep betting the money, their balance goes up or down, you just don’t worry too much about it, and eventually the balance goes to zero and you’re like “thanks for playing, try again sometime.” If you are marking their trading to market — if you lose $1 when they make $1 — then in some sense you are not capturing the real value, which is “they might be up now but we know how this story ends.”

If you offered retail traders an explicit proposition like “give me $100 and I will let you bet $10,000 a day on foreign exchange markets, but you can’t withdraw any money until you have lost it all,” presumably nobody would take quite that offer. You can get pretty close, though! Bloomberg’s Alice Atkins reports:

The pitch is enticing: Learn to trade like a real Wall Street master of the universe. Pay a fee—typically starting below $100—and manage thousands in fictional funds in a simulated market of stocks, currencies or commodities. If you stick to the strict rules, and meet the profit target, you can become a “funded” trader for the firm. That means you could receive payouts of as much as 90% of the ersatz profits you make. If you fail, you can keep trying, by paying the fee again.

More than 400 companies offer forms of these challenges, up from 10 only five years ago, according to FPFX Technologies LLC, a Florida-based software provider for the industry. … They have about 6 million customers collectively, according to their websites. …

Some customers complain they’ve paid high fees, while the chances of winning rewards are slim. Only 4% of users get any payout based on their trading, according to FPFX. In social media posts, people describing themselves as prop-firm customers claim to have lost thousands on fees.

One of them, Samim Sharif, 28, an accountant from London, estimates he’s paid about $10,000 to simulated-trading firm PipFarm since last year. “When you have a bad streak it’s very easy to think, ‘I’ll just buy one more.’ Before you know it you’ve racked up a lot in losses,” he says.

I do not find that pitch enticing at all, but apparently 6 million people do.

We talked yesterday about how South Korean securities regulators now require retail traders to watch a one-hour training video before trading leveraged exchange-traded funds. It’s a great idea that you could extend to all sorts of retail trading products. It would be fun to do the video for this one.

A basic theme around here in recent years is that banks are re-tranching. In the not-too-distant past, there were businesses that needed money to do business stuff, and often they would borrow the money from banks. The advantage of borrowing the money from banks is that banks have cheap funding: The banks get a lot of their money from depositors, and they can pay the depositors a very low rate of interest. The disadvantage of borrowing the money from banks is that banks have risky funding: The banks get a lot of their money from depositors, and the depositors can withdraw that money at any time. 

And so in recent years, because of regulatory changes and general caution, banks have at the margin retreated from lending money to businesses to do business stuff. Instead, other lenders — “private credit” — increasingly lend money to businesses to do business stuff. What do the banks do? Well, one thing that they do is lend money to the private credit firms. Instead of lending money directly to businesses, and taking the risk of losing money if the businesses fail, the banks “move up the capital structure”: They have a senior claim on a senior claim; if the businesses fail, the private credit lenders take the first loss and the banks take the second. The banks have more seniority, possibly at the cost of less control.

This stuff is variously controversial: Is it bad that more loans are being made by less-regulated private credit firms, instead of by more-regulated banks? Is it bad that the banks are increasingly intertwined with the private credit firms? To me it seems like this system is broadly safer than the old system — more credit risk is borne by funds with long-term locked-up equity investors who knowingly take the risk, and less by run-prone regulated banks — but obviously there are ways for it to go wrong. 

In any case, it’s happening. Bloomberg’s  Laura Noonan reports today:

Global assets in the sprawling shadow banking sector have crossed the $250 trillion mark for the first time, new data from the Financial Stability Board shows, fueling fears of mounting systemic risks from less regulated corners of the financial sector.

The FSB’s annual global financial monitor shows non bank financial institutions — a group that spans hedge funds, insurers, investment funds and others — had a record $256.8 trillion of assets at the end of 2024, up 9.4% year-on-year. The group now accounts for 51% of total financial assets, similar to its pre-pandemic share. …

FSB chair and Bank of England Governor Andrew Bailey has previously called out the risks in non banks and said understanding their evolution would be an “important focus” as global watchdogs assess the resilience of the financial system.

The FSB lamented the lack of data around the growth of the multi trillion dollar private credit industry, an area that regulators are keenly scrutinizing for signs of weakness amid warnings of vulnerabilities from bank bosses including JPMorgan Chase head Jamie Dimon and UBS chair Colm Kelleher.

And Bloomberg’s Rene Ismail reported yesterday:

US banks are lending more to private credit firms, private equity shops and hedge funds, with loan volume to these non-bank financial institutions up 26% this year through November, according to Fitch Ratings.

Domestic banks made about $363 billion of new non-bank loans through Nov. 26, Fitch analysts wrote in a report on Monday, citing Federal Reserve data. Banks added $291 billion across all other loan types, according to Fitch.

Regulatory capital requirements and strong demand from borrowers contributed to the uptick in lending to non-banks, Fitch said. But the added exposure brings risk for banks, which are increasingly intertwined with the private credit and private equity industries.

“Asset quality remains benign,” notes Fitch, “with NDFI [non-depository financial institution, i.e. private credit et al.] delinquencies at 0.22% on average at Sept. 30, versus 1.3% for C&I [commercial and industrial, i.e. business] lending,” which really is the purpose of all of this.

A lot of the shares of most US public companies are owned by the same handful of giant diversified institutional investment managers, particularly the “Big Three” managers (BlackRock, Vanguard, State Street) that run big index funds. We talk sometimes about the effect that this might have on competition: There is a controversial but delightful theory that, if the same shareholders own all the companies, those shareholders will not want the companies to compete with each other too vigorously. If Company A cuts the price of its product to win market share from Company B, that might be good for Company A’s shareholders and bad for Company B’s shareholders, but it will be worse for the shareholders overall, and if Company A’s shareholders and Company B’s shareholders are the same firms, then they will prefer to keep prices high and competition gentle.

I suppose you could tell a simpler antitrust story about the concentration of stock ownership in a small group of big institutional managers. The simpler story might be: “If there are three firms that own all the stocks, then there will not be enough competition in stock ownership. Those three firms will be a cozy oligopoly, [2] and they will not bid against each other too aggressively to buy stocks. When companies need to sell stock, they will have to sell it to those three firms, who will keep stock prices low to capture more value for themselves at the expense of the companies.”

I have never really had this thought before? The US stock market is quite notably competitive. The barriers to entry are low; you can start a Robinhood account for free and buy some stocks in like five minutes. “All the stocks are owned by three firms” is an occasionally useful overstatement, but it is not true; in fact the Big Three own something like 22% of the average big US public company. And their effect on prices is ambiguous. They manage a lot of passive index funds, which buy stocks at whatever the market price is; the market price tends to be set by active hedge funds viciously competing with each other for edge. “BlackRock and Vanguard suppress stock prices” is not a thing that people usually think. (Sometimes they think the opposite: “Too much money in index funds artificially inflates stock market bubbles.”) 

Still, maybe a little? There are some corners of the stock market that have higher barriers to entry. For instance, you generally can’t open a Robinhood account and buy shares in a hot initial public offering. [3] Here’s “The Price of Power: The Big Three and IPO Underpricing,” by Adi Libson, Danielle Chaim and Yevgeny Mugerman:

We present empirical evidence that concentrated market power in the hands of a core group of giant asset managers has exacerbated IPO underpricing — defined as the difference between the offer price and the stock's closing price on the first day of trading. Our analysis indicates that from 2002 to 2022, the simultaneous participation of the three largest asset managers — BlackRock, Vanguard, and Fidelity — in IPOs increased underpricing levels by an average of 16.7 percentage points. Even after controlling for IPO size, bookrunner, industry, and year fixed effects, this impact remains substantial at 9.7 percentage points. The participation of such market-moving institutional investors can drive up underpricing through various mechanisms. Our analysis pinpoints several channels through which these investors signal their bidding intentions, share information, and even coordinate their positions during the IPO process. Some of these mechanisms warrant closer scrutiny, as they may constitute collusive behavior by institutional investors in their role as competing bidders in IPOs-potentially violating antitrust laws. Our novel analysis of underpricing through the lens of institutional-investor market power adds a crucial piece to the IPO underpricing puzzle and illuminates the marked correlation between rising underpricing levels and the ascendancy of asset manager capitalism. Notably, over the past decade, underpricing has soared to extraordinary levels, resulting in an unprecedented $90 billion left 'on the table' by issuers.

Note that their Big Three are BlackRock, Vanguard and Fidelity: Index funds mostly don’t invest in initial public offerings, so they focus on big managers of active funds. Is it possible that IPO bankers tell companies “if you want Fidelity and BlackRock in your IPO, you have to price it to appeal to them,” and some companies say “yes we do want that” and accept a lower price? Maybe?

During its brief glorious flourishing, we talked a lot around here about the strategy of selling $1 worth of crypto for $2 on the stock market. Obviously, if you can do that, you should do it all day long. For a while, Strategy Inc. (formerly MicroStrategy), the original crypto treasury company, could, and there were many copycats, some of which could sell $1 worth of crypto for like $10. Definitely do that all day long.

But, you know, come on. That can’t last forever. For one thing, eventually people will notice and stop paying $2 for $1 of crypto. For another thing, though, you’re closing the arbitrage. If you are a Bitcoin treasury company and people are buying your stock at a 100% premium to the value of your underlying Bitcoin, and you are merrily selling them as much stock as you can and using the proceeds to buy Bitcoin, then your sales are pushing down the price of your stock and your purchases are pushing up the price of Bitcoin. If you do this without limit, eventually the premium will close.

And then, if you’re unlucky, your stock trades down to 90 or 80 or 50 cents on the dollar, and you face hard choices like “should I run the trade the other way and sell my Bitcoin to buy back stock?” (“Yes” is the good financial trader answer, but “no” is the good crypto true believer marketing answer; I don’t know what you should do.)

If you’re lucky, though, your stock trades down to 101 or 105 or 110 cents on the dollar, and you face hard choices like “well this isn’t nearly as fun at it used to be, but should I keep selling $1 worth of crypto for $1.10 on the stock exchange?” Actually that’s not that hard a choice. You keep doing that, it turns out:

Michael Saylor’s Strategy Inc. acquired almost $1 billion in Bitcoin for a second consecutive week, as the original digital asset treasury company continues to ramp up purchases following the recent pullback in the price of the largest cryptocurrency.

The former MicroStrategy bought $980.3 million of the digital asset between Dec. 8 and Dec. 14, according to a regulatory filing on Monday. This marks the company’s largest amount of Bitcoin acquired since July and its second consecutive week of acquiring over 10,000 Bitcoin on its books - the first time this has happened since January. 

The majority of the most recent acquisitions were made using proceeds from at-the-market sales of its Class A common stock. Critics of Saylor’s model have raised concern that selling the shares is diluting existing equity of shareholders and eroding the formerly high premium the stock garnered over its now roughly $59 billion Bitcoin holdings. ... The common share price premium over the enterprise value of the company is about 1.1.

Yeah look if you had asked me two years ago “is it a good idea to sell $1 worth of stuff for $1.10” I would have said “of course what are you even talking about.” Compared to, you know, any other possible business, selling stock at a 10% premium to buy Bitcoin is still pretty good. Not what it used to be, but still pretty good.

One mystery about Jeffrey Epstein is that he was extremely wealthy, but he didn’t seem to do anything for a living other than pal around with other wealthy people. Where did the money come from? I feel like there are three main theories:

1. He would pal around with other wealthy people, do sex crimes with them, record them on video and then blackmail them.
2. He would pal around with other wealthy people, give them self-taught but extremely acute tax advice, and charge them a lot of money for it.
3. He would pal around with other wealth people, get them to trust him with their money, and then steal it.

Theory 1 is the most salacious, but there’s never really been any evidence for it. Theory 2 is kind of blah, but there is evidence for it: He seems to have had at least one client who found his tax advice valuable, and a tax lawyer once emailed Epstein a column I wrote about a tax trick to be like “great trick, was this you?” Theory 3 is in some sense morally satisfying — “he was a sex criminal and a financial criminal” — and embarrassing for his associates — “their association with him tainted them morally and got them robbed” — and there is some evidence for it too: In 2019, Leslie Wexner, one of Epstein’s biggest benefactors, accused him of misappropriating money he managed.

I suppose they are not mutually exclusive. Epstein could have managed his friends’ money, saved them a lot of taxes, and stolen money for himself. My own instinct is that geniuses of tax minimization and geniuses of theft are two mostly disjoint categories, but I can see why someone might think otherwise.

Anyway here’s a new investigation in the New York Times that comes down solidly for Theory 3:

In his first two decades of business, we found that Epstein was less a financial genius than a prodigious manipulator and liar. Abundant conspiracy theories hold that Epstein worked for spy services or ran a lucrative blackmail operation, but we found a more prosaic explanation for how he built a fortune. A relentless scammer, he abused expense accounts, engineered inside deals and demonstrated a remarkable knack for separating seemingly sophisticated investors and businessmen from their money. He started small, testing his tactics and seeing what he could get away with. His early successes laid the foundation for more ambitious ploys down the road. Again and again, he proved willing to operate on the edge of criminality and burn bridges in his pursuit of wealth and power.

Okay but later he was really good at saving taxes? It is difficult to end up owning a Manhattan townhouse, a private jet and a Caribbean island by abusing expense accounts, but I suppose that, with a big enough expense account and thorough enough abuse, it’s possible.

Dutch Pension Shift Is About to Get Real for Worried Bond Market. CoreWeave’s Staggering Fall From Market Grace Highlights AI Bubble Fears. Morgan Stanley Climbs Debt Rankings as Go-To Bank for AI Bonanza. Kalshi and Other Prediction Markets Should Scare Sports Leagues. High-Speed Traders Are Feuding Over a Way to Save 3.2 Billionths of a Second. OpenAI Deal to License Disney Characters Is Entirely in Stock. Shell mergers chief departed after CEO blocked bid for BP. Bitcoin Fatigue Sets In as Token Heads for Fourth Annual Loss.  Databricks Is Raising Over $4 Billion at $134 Billion Valuation. Juventus Shares Climb After Agnelli Holding Company Rejects Bid From Tether. Global brands seek private equity partners to save their China businesses. Pimco Gathers $7 Billion for New Asset-Based Finance Strategy. Family of Late Zappos CEO Declares Mysterious Will a Forgery. Morrisons loses £17mn VAT fight as court rules rotisserie chickens are hot food.

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!