| Here’s a neat little piece of derivatives structuring. Let’s say you’re an investment bank, perhaps Morgan Stanley. You might want to buy crash insurance on SpaceX’s stock: You might be willing to pay a premium for long-term, out-of-the-money put options that pay out if SpaceX’s stock goes down a lot. Why might you want that? Off the top of my head: - Your customers — institutional asset managers — might want to buy that insurance from you, and you might want to hedge. The customers might own SpaceX for its upside (rockets, artificial intelligence, etc.), but they might worry about an AI bubble, Elon Musk risk, overvaluation, etc. They might be looking to buy insurance from you, and you might be looking to source that insurance from someone else.
- You might run a big private wealth management business that recently signed up a bunch of multimillionaire SpaceX employees. You might be earning big fees for managing their millions, and you might also be lending them some money to buy houses and Lamborghinis and stuff. Were SpaceX’s stock to crash, they might stop being multimillionaires, which would be bad for you from a fee-income and credit-risk perspective. SpaceX puts might work as insurance for your wealth management business.
Where can you buy this insurance? There are long-dated listed put options on SpaceX, but I’m not sure you could buy a lot of size in that market. You could go to a hedge fund and ask it to sell you some over-the-counter options, but it might charge you a high price. Plus there is some wrong-way risk there: A world in which SpaceX stock has crashed is not necessarily a world in which the entire AI boom has collapsed, but there is probably some correlation, and you might worry about relying on a hedge fund to pay out your insurance in that state of the world. One move is to buy the insurance from your own retail and private-wealth customers, and have them pay upfront so you know they’re good for it. That is: - You buy some long-term out-of-the-money SpaceX puts from retail investors: If SpaceX ends up below some threshold price at expiration, the investors have to pay you $X for every dollar it ends up below the threshold, up to a maximum of, say, $1,000 if SpaceX goes to zero.
- You collect the $1,000 upfront, so you know that the insurance will pay out even in the worst-case scenario.
- When the puts expire, if SpaceX is flat or up, or even down but not below the crash threshold, you give the retail investors their $1,000 back, plus interest, plus some extra money to pay for the put premium.
This is a kind of structured note, and one nice thing about it is that the investors might not think about the fact that they’re selling you crash insurance against SpaceX. To the investors, this looks like a bond that pays a high interest rate (regular interest, plus the extra money for the put premium) and is (sort of) linked to SpaceX’s stock performance. The investor might even think that she is buying protection against a decline in SpaceX: After all, if SpaceX goes down a little (but not below the threshold price), she gets a positive return. She’s protected from a little decline in SpaceX, but she’s protecting you from a big one. We have talked about this general idea — investment banks buying crash insurance from structured-note customers — before. As I once put it: This solves the two problems of crash insurance supply. First, you have transformed the story from “you are selling me insurance against a market crash” (scary!) into “you are getting a high yield on your money” (everyone likes yield!). Second, you have taken the money up front, so you don’t have to rely on the credit of the put seller: If the market crashes, the money will be there.
That was about general market crash insurance, but it works for SpaceX too. Bloomberg’s Lu Wang reports: Wall Street is racing to roll out complex investment products tied to SpaceX shares that aim to shield buyers from future losses amid a sharp selloff following its market debut. At least five financial firms, including Morgan Stanley and Marex Group Ltd., are seeking to offer SpaceX-linked structured notes that limit downside — in some cases protecting against declines of as much as 50% — while capping gains over the coming months or years, according to regulatory filings. … Blending fixed-income characteristics with derivatives, structured notes function as debt-like securities that provide enhanced payouts compared to standard bonds. They are most commonly sought by high-net-worth individuals, family offices and discretionary managers looking for customized risk profiles in their investment portfolios. … Marex is offering a note due in nine months that can be redeemed, or called, automatically, with the principal returned in full should SpaceX’s stock close on predetermined dates at or above its initial value. As long as the note is active, investors are promised monthly interest of at least 1.8% regardless of how the stock performs. Upon maturity, investors will be protected against declines of as much as 35%, but will face the full downside if the stock falls beyond that point. At Morgan Stanley, a note is designed to offer a fixed payout of 40% as long as SpaceX shares are flat or up at maturity in early 2028. That payout also applies when the stock is down by less than 50%, but for any drop beyond that threshold, note holders are fully exposed to the downside. Here’s that Morgan Stanley note. One way to describe it is that it is a product “tied to SpaceX shares that aim to shield buyers from future losses … while capping gains.” That is roughly Morgan Stanley’s description: The securities are for investors who seek a return based on the performance of the underlier [i.e. SpaceX stock] and who are willing to risk their principal and forgo current income and returns above the upside payment in exchange for the upside payment feature and the limited protection against loss of principal.
But that’s not how I would put it. Notice that the investor gets no actual SpaceX exposure except in the worst case: She gets a fixed 40% return except in cases where SpaceX is down more than 50%, in which case she eats all of the losses. I would have described it as: Morgan Stanley is buying a prepaid down-and-in knock-in put option, struck at the money but with a down-50% knock-in. [1] If the stock is down a little bit, the put does not pay off, and the investor gets her money back. If the stock is down more than 50%, though, the investor pays Morgan Stanley all of the downside below today’s price. Roughly speaking, a $1,000 structured note represents about nine SpaceX shares at today’s price of around $112. If SpaceX is trading at, say, $50, when the notes mature in February 2028, those nine shares will be worth about $450. Morgan Stanley will give the investor back the $450 (the value of the underlying stock at maturity) and will keep the other $550: That is the payoff of a put option struck at today’s price. (But it’s a knock-in put, because if the stock finishes at $100 Morgan Stanley returns the whole $1,000.) Effectively, Morgan Stanley gets a big payout only if the stock drops a lot. For this put option, plus the right to keep the investor’s money for 18 months, Morgan Stanley pays a 40% premium. Morgan Stanley is buying SpaceX crash insurance from its structured note customers. Also it’s buying it cheap: Morgan Stanley estimates that the value of the structured note is $947 per note, meaning that there’s about a 5.3% profit margin for Morgan Stanley in the product. I guess the fun question is, how do you replicate the sports index? We talked last week about a set of proposed “sports indexes,” and now Bloomberg’s Katherine Doherty reports: Exchange operator CME Group Inc. is planning to offer futures and options contracts tied to professional and college sports, according to a statement seen by Bloomberg News. The firm is working with FutureSports, an upstart index provider whose products will underpin CME’s new tools. The first sports futures contracts, which will be cash-settled and expire on a monthly and quarterly basis, could begin trading as soon as this summer, pending regulatory approval, according to the statement. … By listing regulated futures contracts tied to sports, CME would bring the industry a step closer to the traditional financial markets. The exchange is also making a bet of its own that institutions will use the new futures to protect against potential losses related to sports. “We’re seeing incredible demand for these unique new contracts as hedging instruments from a broad range of potential participants, from stadium owners and operators, to sports sponsors and endorsers, insurers, sports apparel manufacturers and league broadcasting partners,” Leigh Taylforth, co-founder of FutureSports, said in the statement. Sure! Here’s the press release, which has a lot of stuff like this: Bob Fitzimmons, EVP at Wedbush Securities, said: “The FutureSports indexes take the insular world of the business of global sports and open it up to the global capital markets. What was once relegated to wealthy individuals and private entities now becomes a tradable financial asset that can be used as a risk management tool in the dynamic ecosystem that encompasses sports, media, entertainment, real estate and technology. We at Wedbush are proud to be a part of this burgeoning asset class.”
The indexes, as we discussed last week, are weird. Today’s press release says: Under development since 2022 and launched in 2026, Chicago-based FutureSports has created a proprietary index methodology for measuring on-field, on-ice and on-court performance for a range of professional sporting teams and athletes. Partnering with many of the most recognizable sports leagues and financial market participants, FutureSports transforms live, play-by-play statistical data into rules-based, benchmark indexes that may be referenced by exchange-listed financial products.
That is: For each of “a range of professional sporting teams and athletes,” FutureSports will construct an “index” that somehow captures their “performance.” There will be, say, a Mets Index, which will apply a set of rules-based calculations to a set of “play-by-play statistical data” about the Mets to generate a single number. The number “starts from a standardized base value of 7,500 prior to the start of each season, and moves up or down based on officially reported statistics from every game.” So presumably if the Mets win a game, the index will go up to, I don’t know, 7,550 or something. And if they hit a lot of home runs in the game, it will go up to 7,565, maybe. And if their star player is ejected for brawling, maybe it will go down to 7,558. Et cetera. Some proprietary (but presumably published) mechanism for turning on-field data into a single number. And now CME will list futures on that number. (And pretty soon someone will launch exchange-traded funds based on those futures.) But, like: What is that number? It “can be used as a risk management tool,” I guess, but nobody actually has preexisting risk that the Mets Index will decline from 7,500 to 7,400. What do you get, if you buy (or sell) these futures? You could imagine it going one of two ways: - The Mets Index is essentially a bet on the Mets being good. If you think the Mets will be better than expected, you buy the Mets Index futures. If you think the Mets will be worse than expected, you sell the Mets Index futures. Two sides make a market, and if there are buyers who are optimistic about the Mets’ performance and sellers who are pessimistic, there will be some market-clearing price. People do like to bet on sports, and often the essential nature of that bet is “I think a certain team will be [better][worse] than expected.” Putting on that bet in the form of CME-traded futures has a certain appeal. You can do it through your broker, and it gets better tax treatment than going through a sportsbook. Also it is a long-term continuous bet: Instead of betting on the Mets to win tonight, or betting on them to win the World Series, you make a bet on generic Mets goodness, you keep it on for as long as you want, and you can exit at a profit at any point if the Mets are outperforming. I don’t think that’s the sports bet most people want — they want immediate binary action — but perhaps some people want it. Perhaps there’s some bar somewhere whose beer sales are correlated with overall Mets goodness, and would want to short this futures contract as a hedge.
- The Mets Index is actually a weighted average of other bets on the Mets: It goes up X points if the Mets win tonight, and Y points if the Mets win tomorrow, and Z points if the Mets hit a lot of home runs today, and down W points if an important player is injured, etc. You can — at a sportsbook, or on Kalshi — buy bets on each of those individual items. [2] You can put together a collection of these bets to replicate the index. [3] There is an arbitrage between the Mets Index and some set of binary sports bets on the Mets; if the Mets Index trades too high, arbitrageurs can sell the index and buy the underlying bets on Kalshi, or vice versa.
I wrote yesterday: My prediction markets white whale is something like “a sports gambling exchange-traded fund that has a systematic positive-expected-value strategy.” We talked a few weeks ago about a proposed sports gambling ETF that would be actively managed by a professional gambler, who would try to pick undervalued bets on Kalshi: Possibly a positive expected value, but not systematic or transparent. I half-jokingly proposed an alternative idea, a sports gambling ETF that just bet on the Mets every day or whatever: Very systematic, and possibly a fun gamble, but obviously negative expected value. But what I really want is something systematic and plausibly good.
And this is, you know, almost that. You start a Mets Index ETF, it buys Mets Index futures, and its returns track the Mets Index. It has an intuitive appeal to gamblers: If you like the Mets, you can make a generic “the Mets will be good” bet in your stock portfolio. It is systematic, in that it is literally passive (it just buys and holds the index futures), and in that the index itself is mechanically calculated. It doesn’t have a positive expected value, of course: If there are 30 baseball ETFs, each based on the index for one of the 30 Major League Baseball teams, they can’t all go up; there is no external source of positive returns. Sports betting is necessarily a zero-sum game, or negative-sum after transaction costs. It is not actually a “burgeoning asset class.” But everyone keeps treating it as though it is. That said, I have no idea how the indexes are actually calculated. You could imagine a calculation in which the indexes for all 30 Major League Baseball teams go up, in expectation, over the course of a season. (Each team gets 20 points for a win and loses 18 for a loss, each home run hit is worth 2 points and each home run allowed is worth 0, etc.) Being long the futures would have a reliable positive return, and being short would have a reliable negative return. (Presumably the futures would trade at a premium to their index level to account for this?) That would be weird! But all of this is weird. One thing that I sometimes say around here is that the modern packaging for any trade is an exchange-traded fund. In the olden days, your broker or wealth adviser would pitch you some idea — “buy stocks that do AI” or “go long Tesla and short Ford” or “buy Bitcoin and gold” or “buy the stock index but sell out-of-the-money calls for income” or whatever — and, if you bit, she’d go and trade stocks or whatever to implement the idea. But in 2026 you don’t talk to a person. Instead, ETF companies come up with ideas like that and package them into ETFs, the AI Stuff ETF or the the Long Tesla/Short Ford ETF or the Bitcoin ’n’ Gold ETF, etc. And then (1) if you do have a financial adviser, she can pitch the idea to you and also implement it with one click, and (2) if you don’t have a financial adviser, you can come across the idea by scrolling on Robinhood or reading this newsletter or whatever, and implement it yourself with one click. The ETF format is a way to mass-produce and distribute what used to be bespoke handcrafted trading ideas. This leads to what people sometimes describe as a proliferation of ETFs. There are only a few thousand stocks, but there are googols of ways to combine them, so of course there are more ETFs than stocks. In the limit, every person should have her own ETF, or perhaps several ETFs. Different investors will have different ideas and preferences and interests, and will want different ETFs to implement them. A reader recently emailed to push back on this theory, arguing that the ETF is actually an obsolescent technology for packaging investment ideas. If your goal is “implement any possibly appealing investing idea with a single click,” putting the idea into an ETF is the modern standard. But big retail brokerage firms are rolling out agentic investing tools, which will more or less allow you to describe some investing idea, preference, strategy, dream, etc. in natural language for the agents to carry out. I once wrote about Robinhood’s agentic investing plans: This is really a combination of two modern technological developments. One is agentic artificial intelligence, but the other is free stock trading. An agent that regularly rebalanced your $10,000 stock portfolio, never mind one that was constantly buying oversold stocks and selling them when they reverted to the mean, would have no appeal if you had to pay $9.95 per trade. Free retail stock trading — which didn’t really exist a decade ago — enables much more trading, which is a difference in kind, not just degree. In a world of expensive trading, most people’s investment theses have to be, essentially, stocks: ”This company is good, so I will buy its stock and hold it for a while.” In a world of free trading, your thesis can be something more general and thematic — “buy stocks that went down yesterday” or “buy stocks that went up yesterday” or “buy stocks that were mentioned on television this morning” — and you don’t have to worry about the details of implementing it with stock trades. If it requires making dozens of trades per day, that’s no problem.
There are actively managed ETFs, of course, but if you want a frenetically managed buy-the-dip-and-do-10-other-things-at-a-time strategy, now you can roll your own. I found this reader’s objection persuasive and a little sad: Just as we perfect the ETF as a wrapper for everything, it is becoming obsolete. But for now the ETF-for-everything boom is going great. The Financial Times checked in on it: As of mid-July, 1,084 new ETFs had been listed, including funds offering leveraged bets on equity indices or individual stocks. That compares with 2025’s record full-year tally of 1,161 and is well above the total for any previous year, according to data firm Morningstar. ... “The spaghetti cannon [of launches] is firing wildly right now and not even hitting the wall in some instances, let alone sticking,” said Bryan Armour, director of passive strategies research for North America at Morningstar. … Venture capital-backed start-up Corgi Funds began launching in December and has now debuted 188 funds, including the Coffee and Energy Drinks, Buy Now Pay Later and War Machine ETFs, as well as vehicles offering twice-leveraged exposure to Chinese internet, Taiwanese and Korean stocks. It has filed to launch 360 more. That puts it on course to overtake BlackRock — the world’s largest asset manager, which has 488 US ETFs, according to StockAnalysis — as the biggest issuer by number of funds in a matter of months. However, BlackRock’s assets in US ETFs total $4.5tn, compared with less than $1bn for Corgi. “We recognise that not every fund is going to be a winner,” said Anthony Crinieri, a portfolio manager at Corgi, adding that he anticipates roughly 20 per cent of the funds the company launches will gather 80 per cent of its assets. No, right, some ideas are bad, but you have to offer every possible idea because you never know what someone might want. I guess the bull case for ETFs as a universal wrapper of investment ideas is tax. ETFs, unlike you and your Robinhood agent, can trade stocks without paying taxes. Stock trading is commission-free now, but it is not tax-free, except in an ETF. Sure okay: PwC published reports on AI and electric vehicles riddled with fake footnotes, misattributed claims and unverifiable information, the latest example of a Big Four firm’s slapdash use of AI-generated content. The AI hallucinations were contained in “thought leadership” reports designed to drum up consulting work for partners in the Middle East, according to an investigation by researchers at GPTZero verified by the FT. I do feel like the essential problem here is that “‘thought leadership’ reports designed to drum up consulting work” are not actually intended to be read, which means that nobody cares enough to write them, which means that they are written by AI, which means that they are full of hallucinations and nonsense. Which is fine, because their purpose is to be sent to potential clients who will think “ah yes, PwC exists and is thinking deep thoughts about AI,” or whatever, but who will not actually read the reports and spot the nonsense. And then occasionally a journalist will spot the nonsense and go to PwC to be like “do you know you’re publishing gibberish?” And PwC will be like “ehhhh”: PwC Middle East told the FT it “takes the accuracy of our published research seriously and is updating a limited number of supporting citations” in the identified reports. “Consistent with our approach to responsible AI, we have quality control processes for research and content development we expect all our people to adhere to,” it added, without addressing how the errors were included in the reports.
“If it’s not worth doing at all, it’s not worth doing well,” as Charlie Munger said, and AI creates a lot of opportunities to do poorly things that are not worth doing at all. BlackRock Sets Out for Private Credit Glory After Year of Upheaval. NextEra, Brookfield to Build $100 Billion Kentucky Data Campus. The Price to Finance the AI Data Center Boom Is Rising, Just Ask Meta. ING Nears SRTs on $10 Billion of Loans Including AI-Linked Debt. Moonshot AI Surpasses Funding Goal to Hit $35 Billion Value. Trump Administration Bans New Humanoid Robots From China. Google DeepMind dismantles Nobel-winning AlphaFold team in strategy shift. AI Boom Spurs Insider Selling by Nvidia, CoreWeave Billionaires. A.I. Companies Are Recruiting Electricians and Carpenters by the Thousands. FTSE 100 hits all-time high as ‘anti-tech’ index shines in global chip rout. Ares $29 Billion Private Credit Fund Sees Uptick in Non-Accruals. UBS and Deutsche Bank join Wall St rivals in trading-led profit surge. Apple Launches Product Leasing Program Through Klarna. Anglo Said to Be in Talks to Sell De Beers for About $1 Billion. Aston Martin defends contentious £550mn debt deal. Corporate boardroom diversity sinks to lowest in more than a decade. Fraudster Gets Seven Years for Scamming Victims Looking for Love. Ebay pays couple $56mn for its role in cockroach harassment campaign. Ohio State, JPMorganChase reach lucrative jersey patch deal. What Is It Like to Be a Cucumber? 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! |
