| One theory of meme stocks is what you might call “technical.” This theory is: If a lot of people on social media decide to buy the same stock, it will go up. If it goes up, the people who bought it will make money. Therefore, they should buy it, to make money. There are flaws in this theory (eventually you run out of buyers), and more complex variations on it (short squeezes, gamma squeezes, etc.), but the important point here is that this theory is essentially self-referential. You buy meme stocks for the meme; you make money because the meme is popular, because people buy the stock. The actual business of the company is not particularly relevant. A certain sort of consumer-facing, nostalgia-driven business might be helpful for the meme: If stock investors are more likely to buy stocks with nostalgic associations, those stocks will be more meme-friendly. But the meme-stock effect will not work through, like, increasing quarterly profit margins. It will work through memes. Another theory of meme stocks is what you might call “fundamental.” This theory is: If a lot of people on social media decide to buy the same stock, it will go up. If it goes up, the company will improve. It will make more money and become more valuable, retroactively justifying the people who bought the stock. The meme attention will cause improvement in the underlying business. This strikes me as a weird theory on first principles. Usually the causation is reversed: Usually the stock goes up because the business has improved, rather than the business improving because the stock went up. But it is not impossible. In January 2021, during the great meme stock rally around GameStop Corp. and AMC Entertainment Holdings Inc., I noted that the meme-stock rally really did improve AMC’s fundamentals by helping it pay down debt: “A week ago it was not crazy to think this company was doomed; now it is entirely possible that it will survive and thrive and show movies in movie theaters for decades to come because everyone went nuts and bought meme stocks this week.” In June 2021, a few months into GameStop’s rally, we talked about the possibility that the rally had sparked fundamental improvements. I wrote: By paying a ton of money for GameStop stock, its shareholders have inspired its executives to work harder (and hire better executives). They have inspired those executives to think bigger; if you run a $20 billion company you will naturally make bigger plans than if you run a $300 million company. They have certainly ushered in a whole new era at GameStop: the era in which it is a meme stock with an enormous valuation. Perhaps also an era in which it justifies that valuation.
GameStop is now an $11 billion company so, you know, mixed results, but not nothing. Opendoor Technologies Inc. has been a meme stock for a few months now, and we have discussed that at a “technical” level. (I’ve speculated, for instance, about the role of large language models in coordinating meme-stock investor attention, and about Opendoor booster Eric Jackson’s schtick of trying to get Drake to buy the stock.) But arguably there is also a fundamental story. Bloomberg’s Norah Mulinda reports: Opendoor Technologies Inc. shares surged by as much as 69% Thursday after announcing the return of its co-founders to the board and a new chief executive officer. If the advance holds it will be the biggest one-day gain for the home-flipper turned meme stock on record — continuing its blistering 450% run up since mid-July — after the real estate company said co-founders Keith Rabois and Eric Wu will rejoin the board and named Shopify Inc.’s Kaz Nejatian CEO. “It’s everything I could’ve asked for and more. It’s a dream team,” said Toronto-based hedge fund manager Eric Jackson of EMJ Capital Ltd., who had bought Opendoor shares at around 70 cents, in an interview. The stock traded to as high as $9.94 as of 11:50 a.m. New York time on Thursday. … The management change comes just weeks after former CEO Carrie Wheeler stepped down following increasing pressure from shareholders and online chatter doubting her ability to lead the company. … “The share price is the quantitative measurement of sentiment around this business,” Anthony Pompliano, CEO of ProCap Acquisition Corp. said in an interview. Pompliano, who is a prominent cryptocurrency investor and podcaster, announced that he bought shares of the firm last month and has since been continually touting the stock on X. “Sixty days ago, it was left for dead,” he said. “Today, there are thousands of investors who are very excited about the prospect of this business, and they believe that it is in a much better position than it was 60 days ago.”
“They believe it is in a much better position than it was 60 days ago” because it has made management changes to respond to retail-investor pressure. “Retail Investors Just Pulled Off One Of The Best Activism Campaigns In Recent History,” Pompliano wrote in his newsletter: Retail wasn’t just buying the stock. They began bombarding the company with pressure to improve the business. Within weeks, the existing CEO had stepped down. The remaining management team committed to not sell any of their stock. And the interim leader of Opendoor personally purchased equity in the stock market. ... The pressure continued. They wanted a new CEO who understood artificial intelligence. They wanted Opendoor co-founders Keith Rabois and Eric Wu back on the board of directors. And the retail investors were not going to rest until they got what they wanted. Hundreds of tweets per day. Some nice, some not so nice. Just a relentless campaign to effect change at a business that these retail investors saw potential opportunity in.
And they got what they wanted: Rabois, Wu and Nejatian. Opendoor also announced a $41 million private placement of stock to Khosla Ventures and other buyers at $6.65 per share (Friday’s closing price), providing some institutional validation to the otherwise pretty meme-y stock price. (The stock was at $0.51 in June.) Nejatian’s pay package includes $30 million (in cash and restricted stock) as a make-whole payment “in respect of compensation awarded by his former employer that he is forfeiting,” as well as a performance-based package of up to 81.8 million shares of restricted stock, which would be worth more than $800 million at that $9.94 peak price. (And more than a billion dollars if he actually improves the business beyond current market expectations.) In June, Opendoor’s market capitalization was under $400 million; now it’s over $6 billion. If you can pay your new CEO $800 million, you can probably get a pretty good new CEO. Because Opendoor is a meme stock, it can.  | | | If you are an artificial intelligence researcher at a big tech company, it is plausible that your market value a year ago was $1 million a year, and your market value today is $100 million a year. Which is more. You have a large gain in theoretical value, but you do not want theoretical value; you want money. The simplest way to turn your market value into money is to go to your boss and say “hey, I think I am underpaid, I would like a 9,900% raise,” but that probably won’t work. Your boss is not accustomed to giving 9,900% raises. To mark your compensation to market, you need to test the market. This is particularly true if you work at Meta Platforms Inc., which has for years employed AI workers at regular-ish big tech salaries, but which recently went on a spree of hiring outside AI workers at nine-digit salaries. You might reasonably assume that the rate Meta will pay for outside talent is roughly 100x the rate it will pay for inside talent, which means that if you are inside talent you had better get outside. The Wall Street Journal reports: Some Meta employees have had success leveraging competing offers to get a share of the money and prestige Zuckerberg has been lavishing on outside recruits. In July, a handful of employees from one of Meta’s AI infrastructure teams secured offers from former OpenAI executive Mira Murati’s new AI startup, Thinking Machines Lab. After the employees took the offers to Meta, their compensation was increased and they were moved to the TBD Lab team. A spokesman for Meta said the company had already been planning to move those staffers to the new team and adjust their compensation “regardless of any recruitment offers they may have received” and that the company hasn’t made counteroffers to employees threatening to leave.
And given the volatility in this market, it pays to be a high-frequency trader. If you get hired at $20 million this week, you might be worth $60 million next week, in which case you’ll just have to quit and get hired again: Shengjia Zhao, a co-creator of ChatGPT, joined Meta in June, but decided within a week to return to OpenAI, resigning from Meta and signing paperwork to rejoin his former employer, according to people familiar with the matter. Meta managed to retain Zhao by offering him the title of chief scientist—and tripling his compensation.
Only fair, honestly. I suppose if the market price for AI researchers goes down, they’ll fire him and hire him back at a lower rate. Two bits of conventional wisdom in modern finance are: - The rise of artificial intelligence will be bad for junior employees. For a wide range of roles, financial firms will still need senior employees with market experience, good judgment and client connections. But a lot of the grunt work that used to be done by junior analysts can now be done by AI; the senior people can be more productive supervising an AI than they can supervising three analysts. In the long run that might be bad — you can’t train new senior people without hiring them as junior people and giving them the grunt work — but in the short run it’s efficient, and in the long run I suppose the AIs will replace the senior people too.
- All the senior traders go on vacation in August, so markets are thin and weird in August because every trading book is managed by a nervous first-year analyst keeping the seat warm.
The synthesis of these two ideas is “now, in August, the AIs manage the trading books,” and that is roughly true. Bloomberg’s Isabelle Lee and Caleb Mutua report: In August, as US credit traders go to the beach, algorithms are increasingly stepping in for them, allowing transaction volume to stay relatively high even during a traditionally slow period. Algorithmic trading accounted for more than 40% of trading in the US high-grade market in August, a percentage that has climbed steadily since that month in 2020, when it was less than 10%, according to data from MarketAxess Holdings Inc., an electronic trading platform. In the last four years, algo activity usually dipped in September and picked up again in the last three months of the year, underscoring how automated trading helps to sustain volume — particularly in block sizes — during the summer and year-end holiday periods. Credit traders are increasingly relying on algorithms, decades after they began dominating equity markets, and they have become reliable stand-ins when activity would normally sag. The result has been smoother markets with less volatility and lower trading costs.
But how will the junior traders learn how to manage a book, if they don’t get to do it in August? Just as a general comment: With the rise and normalization of prediction markets, you’re going to see a lot more papers like this: We study how financial markets price threats to Federal Reserve independence by analyzing one of the most direct challenges to Chair Powell’s tenure on July 16, 2025. Following media reports of escalating efforts to dismiss him, prediction markets sharply repriced dismissal risk within a two-hour window, creating a unique high-frequency setting to estimate market responses. Regression estimates imply that the certainty of Powell’s dismissal could erase $0.88–1.51 trillion in total market capitalization. Losses were larger for financial and credit-dependent firms, consistent with the credit channel of monetary policy transmission, and for high-beta firms. Treasury yields declined at short maturities but increased at longer horizons, steepening the yield curve, while the dollar depreciated against major currencies. These results show that markets price not only near-term policy-rate expectations but also the Federal Reserve’s ability to act independently.
That’s the abstract to “The Market Value of Fed Independence: High-Frequency Evidence from a Natural Experiment,” by Linghang Zeng and Jérôme Taillard. The point is that historically financial academics could study the effect of some event on some asset price — “when companies announce spin-offs, their stocks go up by ___% on average,” etc. — but you needed the event to occur. It was harder to study counterfactuals; harder to say things like “a ___% increase in the probability of X causes a ___% increase in the price of Y.” Asset prices are influenced by lots of different uncertain things, and it is generally hard to back out probabilities from prices. And even when the event occurred, other events also occurred; it was not always easy to isolate the effect of some event on asset prices. Whereas prediction markets just give you high-frequency, ready-made probabilities. (More or less.) If you want to answer a question like “what effect would firing Jerome Powell have on asset prices,” you can look at the day-to-day prediction-market-implied probability that he will get fired, regress it against asset prices, and get an answer. From the paper: We leverage intraday data from highly liquid modern platforms, namely Polymarket and Kalshi, to quantify the evolving probability of Powell’s dismissal before the end of 2025, effectively capturing the risk of removal prior to the scheduled end of his term in May 2026. These probabilities
are derived from the prices of binary contracts. Intuitively, under risk neutrality the price of such a contract can be interpreted as the market’s assessment of the likelihood of that outcome. This interpretation motivates our empirical strategy: we trace changes in intraday probabilities to changes in asset prices across three major markets, equities, Treasuries, and foreign exchange, to quantify how markets price threats to Federal Reserve independence.
And: At the onset of the event, dismissal probabilities implied by traded contracts on Polymarket and Kalshi rose sharply, while equity markets fell; when dismissal probabilities declined, equities rebounded. This inverse relationship is evident throughout the episode. To formally assess it, we regress 2-, 3-, and 5-minute SPY returns on contemporaneous changes in dismissal probabilities. Coefficients are highly significant and range from –0.013 to –0.024. The strength of the estimated relationship is further underscored by R2 values reaching up to 0.39 in the tight event window. Scaling these estimates by the observed increase in dismissal risk from 23 to 40 percent (Δp = 0.17) implies aggregate equity losses of $125–214 billion for the S&P 500 and $150–256 billion for the total U.S. market. Extrapolating to the counterfactual case in which markets move from assigning zero probability to full certainty of Powell’s dismissal (Δp = 1) implies declines of –1.3 to –2.4 percent, corresponding to losses of roughly $0.88 to $1.51 trillion in total market capitalization.
Obviously (?) you need some prior reason to think that the one thing is related to the other. “Stocks will go down if Jerome Powell is fired” seems like a reasonable hypothesis to start with. But there are a lot of prediction markets these days, and they are not necessarily concentrated on the most macroeconomically relevant events. If you run a regression of bond yields against, like, the Bills’ probability of beating the Ravens, you might find something, but it’s probably spurious. You can run a Ponzi scheme at different levels of (fake) financial sophistication, depending on your own financial sophistication and that of your target victims. If you are a real hedge fund manager and you turn Ponzi, and you raise money from real hedge fund investors, you will want to make a big show of compliance and audits and nicely formatted presentations and monthly statements and talking like a hedge fund manager. But there are also pretty down-home Ponzi schemes that appeal to less professional investors, where you do not need to do all that stuff. Here is a Securities and Exchange Commission enforcement action against “Arsalan A. Rawjani and the business enterprise he operated, Trade with Ayasa, LLC, … for allegedly conducing an affinity fraud and Ponzi scheme centered in the North Texas Ismaili community.” He was not trading on hedge-fund sophistication and a ton of formalities. From the complaint: Throughout the Relevant Period, Rawjani did not follow regular and accepted financial practices with respect to managing a pooled-investment program. For example, one of the main methods that Rawjani used to pay dividends to investors was to provide his clients a stack of hand-written checks at the time of their initial investment. Rawjani post-dated these checks and signed them, but the other check fields were often left blank. Rawjani wrote memos like “[principal amount] * 0.05,” “monthly,” or “dividend” on some checks, but left the memo line blank on other checks. Rawjani wrote in the dollar amount on some checks but left that field blank on others. Rawjani sometimes filled out the payee (i.e., “To”) line on the checks, but he frequently left that line blank. Rawjani then directed investors to cash their dividend checks each month. At various points during the scheme, Rawjani learned that certain checks he provided investors were being written to and apparently cashed by persons who were not the original investor. Further, Trade with Ayasa did not provide monthly or regular account statements, tax paperwork, or even basic marketing materials.
See, I read this and think “if your hedge fund manager hands you a stack of handwritten postdated checks, that probably means he’s not a real hedge fund manager,” but not everyone is looking for a real hedge fund manager. Perhaps his alleged victims thought the handwritten checks were a nice personal touch. One financial market that I used to follow is the market for Pokémon cards at my children’s elementary school. It is a somewhat illicit market — they’re not supposed to trade the cards at school — and there is a wide range of financial sophistication among market participants, because they are children. For instance, I was intrigued by the widespread concept of “fake Pokémon cards”: Older kids will look at a younger kid’s rare fancy card and confidently pronounce that it is fake and worthless. As far as I can tell this is rarely true, and never based on deep connoisseurship and wide market knowledge, but it has a lot of credibility coming from an older kid, and it sometimes results in below-market trades. Also, there is some connection between the local market and the broader, internet-based, global market for Pokémon cards, but not that much. Most of the kids are not making markets locally and laying off risk on eBay, and the law of one price rarely obtains. So if a kid were to have a parent who is a financial columnist, some arbitrages might be available. Anyway here’s a Wall Street Journal article about the broader, internet-based, adult market for Pokémon cards, which is hot: Pokémon cards, which pay no dividends and aren’t subject to financial regulation, have seen a roughly 3,821% monthly cumulative return since 2004, according to an index by analytics firm Card Ladder tracking trading-card values through August. That trounces the S&P 500’s 483% jump over the same period. Meta Platforms, one of the Magnificent Seven, has climbed around 1,844% since the company went public in 2012. … While financial advisers generally caution against betting retirement savings on fictional battling critters, the cards caught fire among amateur investors during the pandemic. As some investors banded together to spark the GameStop meme stock mania, a more fringe group of traders, also stuck at home and armed with cash from government stimulus, began scooping up Pokémon cards. … “I like diversifying my investments. So I’ve got some stocks, I’ve got some crypto, and then, I figured I’d try starting a bit of Pokémon investing as well,” said [Charlie] Pryds, a mason apprentice. “If you like risk in your portfolio, I think it’s a good way to go.”
Obviously Pokémon trading predates crypto and meme stocks, and baseball card trading — its precursor — is decades older. Still I feel like 20 years ago the Wall Street Journal would not have bothered to point out that trading cards “aren’t subject to financial regulation.” Crypto has popularized the intuition that there is a market to trade all sorts of tokens, whether or not they have any cash flow or intrinsic value, and that is good even for non-crypto tokens. But, right, they do not have any cash flow or intrinsic value: Critics say Pokémon card prices are inconsistent and subjective. There isn’t a standard price for the cards, and it’s unknown how many of each are in circulation. The market’s monster gains have also raised concerns about a potential bubble.
Meh, critics. There is an argument, which Cliff Asness made on the Money Stuff podcast, that the lack of an intrinsic underlying asset can be good: A meme stock can trade above its intrinsic value for months or years, but eventually there will probably be some connection between the stock price and the cash flows. A crypto token has no intrinsic value, so it can trade as high as it wants for as long as it wants. (Asness: “I think the more absolutely unsubstantiated by anything something is, the longer the craziness can go on.”) Relatedly: While it’s hard to predict if the cards will retain their value, Pokémon buffs say they are a safer investment than another alternative asset that also took off during the pandemic: Baseball cards. … Even some baseball players agree. “Pikachu’s not going to tear his ACL and miss the whole season. Charizard is not going to get a DUI driving home,” Philadelphia Phillies pitcher Matt Strahm said in a recent interview on freelance journalist Tyler Boronski’s YouTube channel.
Well, I would pay a lot of money for a Charizard DUI mugshot photo card. But that would probably be fake. Subprime Auto Lender Collapse Delivers Blow to Risky Debt Market. SEC chief threatens ban on European accounting rules over sustainability. The CEO Who Wants to Double the Size of His Bank to $1 Trillion. ‘The only company that controls a country.’ JPMorgan Sees Record US Buybacks Jumping by Another $600 Billion. Trump’s CFTC pick claims Winklevoss twins meddled with confirmation. Barings Poaching Suit Turns to £6.3 Million Paid Out Bonuses. JPMorgan Draws Up New Plans for London’s Biggest Office Building. Colleges Are About to See a Big Decline in Applications. AI can’t write good analyst research yet, says analyst. 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! | 
