Prediction market brokering | The way the stock market works is that if a share of stock trades at $50, you can buy it for $50, and then if it goes up to $55 you can sell it for $55 and make a $5 profit. The way the market for crude oil works is different. If crude oil is trading at $95 per barrel, you can't really send your broker $95 and get back a barrel of oil. (Where would you put it?) Instead, the normal way to trade crude oil is with futures contracts: You buy a one-month futures contract now at $95, and if oil is trading at $92 or $107 in a month, you lose $3 or make $12, respectively. Someone else is on the other side of the contract: She sells you the contract now at $95, and if oil is trading at $92 or $107 in a month, she makes $3 or loses $12; her losses pay for your winnings and vice versa. I will loosely call this person your "counterparty," though in fact both you and she deal with the commodities futures exchange and have no direct contact with each other. You could imagine a world in which you buy the contract now by paying $95, and at expiration you get back $92 or $107 or whatever. Or you could imagine a world in which no money changed hands at the beginning, and you and your counterparty just agreed that, in a month, you'll pay her $1 for every $1 oil is below $95, and she'll pay you $1 for every $1 oil is above $95, but no money will change hands between now and then. In fact the way commodities exchanges generally work is somewhere in between: You don't pay $95 at the beginning, but you don't pay zero either. You put down a deposit — "margin" — with the exchange, so the exchange knows you'll be good for the money, and your counterparty also puts down a deposit (even though she is selling the contract). (The deposit is around $9.) And then as the price of oil moves around your margin will go up or down: If oil goes up, your counterparty will deposit more margin and you'll take some back. So commodities futures provide leverage. If you are a big hedge fund who wants to bet on the price of oil, you can do that without tying up too much of your cash. You put up $9, you get exposure to $95 worth of crude oil, and if crude oil prices go up by 10% you roughly double your money. (If they go down 10% you lose your bet.) But leverage creates risk. The futures exchange (or, rather, the clearinghouse) has to set margin levels to cover potential market moves, and then it has to track prices and update margins — calling margin from people who have lost money, sending margin back to people who have made money — every day. There is a whole apparatus for making sure that losing bettors pay, so that winning bettors can get paid. We talk from time to time about this system, mostly when it breaks down. In recent years there have been occasional big and goofy breakdowns of the margining systems at crypto futures exchanges — FTX is the most famous, though we have discussed more recent "auto-deleveraging" problems — but it's not limited to crypto; in 2022, there was a big and goofy breakdown in the London Metals Exchange's nickel market for basically similar reasons. (Big trader got margin calls, didn't post margin, chaos.) You might dislike this. You might say: "No, this is dumb, everyone should put up all the money at the start, just like they do in the stock market. You pay $95, you get an oil contract, and then at the end you make or lose money, but nobody ever has to issue a margin call or post more money." There are two problems with that approach: - Hedge funds like leverage, this market is more efficient with leverage, blah blah blah.
- The subtler problem is that this is impossible. The long side of a futures contract can just put up 100% of the money Day One, sure, no problem. But the short side can't. A futures contract is a bet between two people, one of whom bets that oil will go up and one of whom bets that it will go down. If you are long, you can put up $95, and if oil goes to zero you lose all your money and can't lose more. (Or can you?) But if you are short, you can put up $95, and if oil goes to $5,000 you have to pay much more money. Being short a financial contract is necessarily a leveraged trade. [1]
You might still find this annoying. "Okay," you say, "I take the point that the short side of a commodities futures contract has unlimited risk. But we can fix that. Just chop off the tails. Make a new sort of oil futures contract with a price cap. I agree to buy oil at $95, and you agree to sell oil at $95, and if oil goes down to $40 I lose $55 and you make $55, and if oil goes up to $160 I make $65 and you lose $65. But: The contract is automatically capped at $190, twice the current price. [2] If oil goes to zero, I lose $95; if oil goes to $5,000, you only lose $95 (and I only make $95). We all know that going in, so I am not surprised or unhedged if oil really zooms up. And because our losses are capped, we can each put up $95 on Day One, and nobody ever needs to make margin calls or post more money or have any other sort of credit apparatus. My credit and your credit are completely taken out of the equation; we just put up all the money upfront. (Of course the exchange, or rather the clearinghouse, has to hang on to the money and give it back to the winner, but the exchange is not taking any leverage or market risk either, so as long as they're not stealing the money it's fine.)" The advantage of this system is that it slims down the apparatus, and the risk, of the futures exchange. You don't need a big clearinghouse and complex margin rules; you don't need to vet exchange members to make sure they're well capitalized, or require other traders to access the exchange through vetted members. The exchange can let anyone trade as long as they pay for their contracts up front. The exchange just needs a website and a bank account. The disadvantage is that hedge funds like leverage, blah blah blah, so this sort of entirely prepaid market will be less appealing to large levered professional investors. But it might be nice for retail traders who want to bet on oil. And you can solve the problem for the hedge funds. After all, the stock market loosely works this way — if you want to buy stock, you pay for it — and in fact hedge funds do lots of levered stock trades. They just borrow the money from their prime brokers: The stock exchange doesn't give you $50 of stock exposure for $5, but if you're a big hedge fund a bank will probably lend you $45 to buy $50 worth of stock. I realize that all of this sounds very eccentric, and I don't think I have ever seen anyone propose exactly this market structure, "we should have futures exchanges but with 100% margins and capped moves in either direction." Except that this describes prediction markets. A prediction market is, in the US, a regulated commodities futures exchange, in which every futures contract pays out either exactly $1 or exactly $0. [3] If you are short a contract, the most you might have to pay is $1. If you are long, the most you might have to pay is $1. Therefore, prediction markets never need to think about margin: They can just sell a package of "one long contract plus one short contract" for $1, and then they'll have the dollar, and then when the contracts mature they'll either give the dollar to the long or they'll give it to the short. No credit checks, no margin calls, no carefully vetted exchange membership, just a website and a bank account. We have talked a few times about my aesthetic annoyance with prediction markets trying to shoehorn all sorts of bets into binary yes/no contracts: A trade like " will house prices go up" makes more sense as a linear bet on house prices than as a yes/no bet on some house-price level. But prediction markets make everything a binary option. I have explained this, in (large) part, as appealing to the gambling desires of retail bettors: A binary all-or-nothing bet is simpler and more fun than making $1 for each $1 that a thing goes up. But I have also pointed out that making everything a binary solves the credit problem: If you get all the money upfront, you don't need to check anyone's credit or make any margin calls or take any credit risk. And so prediction markets can be open to anyone, through their own websites, in a way that traditional commodities futures exchanges can't be. You don't have to be a well-capitalized member firm to trade on prediction markets; you can just set up an account and send them $20 and start trading. It is a good structure for selling directly to retail traders. The disadvantage is that hedge funds like leverage, blah blah blah, but they can get that from brokers. Bloomberg's Bernard Goyder and Isis Almeida report: Brokers to hedge funds and other institutional investors are moving to give clients access to Kalshi Inc.'s event bets, a sign of the growing push to open the nascent prediction market industry to Wall Street players. Clear Street, a broker to hedge funds and sophisticated traders, is expecting to clear its first trade on Kalshi later this month and begin offering it more broadly later this year, the company's chief executive officer, Ed Tilly, said in an interview on the sidelines of the Futures Industry Association's annual conference in Boca Raton, Florida. The London-based broker Marex Group Plc is also planning to give its clients access to Kalshi in the next few months, Ram Vittal, the chief executive officer of the company's North American operation, said in a separate interview at the event. "Over the last few weeks we've seen very large hedge funds coming to us and saying 'Can you give us access to these markets?'" Thomas Texier, Marex's head of clearing, said. ... Clear Street and Marex are the latest firms that are trying to bring in professionals by developing more sophisticated market plumbing. They join other futures brokers getting involved in the space, including Wedbush and Plus500. You don't really need an institutional broker to get access to Kalshi; you can just go sign up on the website. But if you are an institutional investor, there are reasons to prefer going through a broker. [4] We talked yesterday about private credit marks. A lot of big private credit firms' private business development companies are getting large redemption requests from investors, and I wrote: One way to think about it is that those redemption requests are a form of NAV [net asset value] arbitrage. If a BDC says that its loans are worth $100, but you think their market value is only $98, you should try to redeem out of that BDC: You'll get cashed out at $100, which is more than you think it's worth. … The fact that private credit BDCs are getting historically high levels of redemptions might suggest that their investors don't believe their marks.
You could have slightly different models — I have previously analyzed the redemptions in terms of liquidity premia — but you certainly could interpret the wave of redemptions as evidence that investors are suspicious about how private credit firms are valuing their assets. There is more direct evidence of that. Today Bloomberg's Esteban Duarte, Davide Scigliuzzo and Silas Brown report: Blue Owl Capital Inc. defended its recent sale of $1.4 billion of loans from three of its funds, arguing the transaction contained no backstops or hidden incentives, as the asset manager remains a primary target of bets on a private-credit reckoning. In a private conference call last week with investors, Blue Owl Co-President Craig Packer said the four institutions that struck a deal with the firm did so on an arm's length basis, conducted their own due diligence, bought the debt on the same terms and received no special guarantees. Each buyer purchased roughly the same amount at the same asset prices and at the same time with "no hidden economics or discounts," Packer said, according to people familiar with the matter. We talked about that loan sale a few weeks ago, and I basically took Blue Owl at its word that selling loans at 99.7% of par to big outside investors suggested those loans were worth about 99.7% of par. I got a lot of emails from people who were more suspicious! Apparently so did Blue Owl. We talked last week about the betting markets on whether Ayatollah Ali Khamenei would "leave office" as Supreme Leader of Iran before the end of March. He, uh, did. If you were going to Kalshi at the end of February to bet that Khamenei would leave office by the end of March, you probably assumed that he was going to leave office via bombs, which is in fact what happened. But I pointed out that Kalshi, as a futures exchange regulated by the US Commodity Futures Trading Commission, is not technically allowed to list betting markets on assassination. And so, technically, the Khamenei contract was not a market on assassination — not only in the sense that you could win if he left office peacefully, but also in the sense that you could not win if he left office via bombs. The official rules of Kalshi's contract specified that, "If [Khamenei] leaves solely because they have died, the associated market will resolve and the Exchange will determine the payouts to the holders of long and short positions based upon the last traded price (prior to the death)." That is, the contract pays off $0 if he stays in office, $1 if he leaves office due to normal political mechanisms, but it refunds everyone at the last traded price if he dies. See, not an assassination contract! (Polymarket, the other big prediction market, doesn't care about this and just paid $1 for his death.) I wrote: One possible interpretation here is "you were trying to bet on war and assassination, which is not allowed, so you don't get to complain that you didn't get paid." If you think that you are getting a wink-wink bet on a prohibited subject, most of what you are getting is not the clean bet you wanted (Khamenei "out") but rather a fuzzy and unpredictable bet on legality and resolution mechanisms. Another possible interpretation is that most of Kalshi's customers don't know about Rule 40.11 [which prohibits assassination markets], or didn't until Saturday, and just figured betting on death was the whole and legitimate point of this market. Everyone's intuitions about prediction markets are still developing.
Obviously you know what happened next: The death bettors are suing Kalshi for their winnings. Bloomberg Law reports: Kalshi Inc. should've paid out on a $54 million market speculating when Iran's Supreme Leader Ali Khamenei would vacate his role after he was killed during US and Israeli airstrikes, a proposed class action said. Kalshi and its exchange's plain-language rules made traders who predicted his departure expect payouts, and they didn't adequately disclose a "death carveout" until after news of Khamenei's death spread, said the complaint filed in the US District Court for the Central District of California. Then rather than halt trading as reports of military strikes accumulated Feb. 28, Kalshi lured more traders into "yes" positions on Khamenei's exit while knowing it wouldn't deliver payouts, Thursday's suit said. Here is the complaint. It seems pretty clear, even from the complaint, that (1) Kalshi's rules clearly said that you don't get paid if he dies, (2) Kalshi followed its rules but (3) most of the customers didn't read them: Upon information and belief, the death carveout was buried in the fine print of the contract's formal terms, which were not disclosed in a manner that would inform a reasonable consumer of their existence or effect.
See, I think that if the members of a commodities derivatives exchange make bad trades because they didn't read "the fine print of the contract's formal terms," and then they complain, the correct answer is "you are a commodities derivatives trader, it's your job to read the fine print, buzz off." But of course if the users of a consumer gambling site have the same complaint, it is a bit more sympathetic. You should tell your consumer gamblers what they're betting on! The problem is that the consumer gambling site is also a registered commodities derivatives exchange. Elsewhere, here's Polymarket founder Shayne Coplan arguing that war bets are good: Speaking at the MIT Sloan Sports Analytics Conference 2026, Coplan argued prediction markets still serve a genuine informational function but described Iran as "complicated" and said the fog of war breeds misunderstanding. "There's still a lot of resistance to innovation that kind of also seems jarring to begin with," Coplan said on Saturday. But "that's what makes it innovative and disruptive." … "When I get hit up by people in the Middle East who are saying, 'Hey, we're looking at Polymarket to decide whether we sleep near the bomb shelter; we look at it every day' and I'm like, 'Oh, it's really that popular over there?'" he added. "That's very powerful. That's an undeniable value proposition that did not exist before." I guess? There is something particularly dystopian about the idea that: - Some countries will bomb other countries.
- The people doing the bombing will profit from the bombing by insider trading the bombing contracts on prediction markets.
- This will cause the prediction markets to correctly reflect the probability of bombing, allowing the people getting bombed to avoid being bombed.
Will the marginal effect of the prediction-market trade-on-bombing profit motive be to encourage more bombings, or to save people from bombings? How are you enjoying the war | Disclosure, I used to work at Goldman Sachs Group Inc., selling derivatives to corporate clients, including oil and gas companies. This was in the years right after the 2008 crisis, when Goldman was famously "a great vampire squid wrapped around the face of humanity, relentlessly jamming its blood funnel into anything that smells like money." The oil and gas guys loved us. "Everyone used to hate us for the global warming stuff," they'd cackle when we walked into their offices, "but now they hate you instead! Thanks for the distraction!" I didn't sell a lot of derivatives but that was a me issue; the conditions were good. Similarly, I realize that there are a lot of nuances, but if you run a private credit fund and your investors ask to take back 10% of their money and you only give them back 5%, that has to feel bad. They will complain, your stock will go down, the headlines will be negative, you will go home to your family and say "I had a bad day at work." And then you will turn on the news and perhaps be cheered up, a bit, by news of war and devastation: At least someone is having a worse day than you. No? I don't know. I am not personally cheered up by news of war and devastation, but I am kind of tired of writing about private credit gates, so I see both sides here: A top Goldman Sachs executive has said that the bank's clients in the private capital industry are "glad" that the Iran war is providing a "distraction" from questions over the sector's exposure to software. Kunal Shah, the co-chief executive of Goldman Sachs International and global co-head of fixed income, currencies and commodities, made the remarks on a call with clients on Wednesday, which also featured Sir Alex Younger, the former chief of Britain's Secret Intelligence Service. According to three people familiar with the call, which was titled "Strikes in Iran — End of the Beginning . . .?", Shah remarked that some of the bank's "private markets clients" were "just glad there's something to talk about that isn't software exposures and private credit". "And this is at least a distraction from that," Shah added, in response to a question about the views of Goldman's clients on the conflict. You probably don't want to be quoted saying "our clients like war because it distracts from their own problems" but that doesn't mean it's wrong. My basic theory is that professional investors sometimes meet with executives of companies, they ask those executives questions about the business and finances, the executives answer the questions and the investors get useful information that they then use to make investing decisions. In US public markets, there is a rule that the executives can't give the investors "material nonpublic information" in these meetings, but the meetings still happen. My theory is not troubled by this — the executives don't tell them "we're doing a merger next week," but they help them understand aspects of the business that are not quite "material nonpublic information" but are nonetheless of interest to sophisticated institutional investors — but other people are, and the standard way of reconciling the facts is by pretending that the executives don't tell the investors anything new. Instead, the investors get to observe the executives' "tone and body language" to inform their investing decisions. I think that this is very silly, but I am also a collector of counterexamples, cases where an investor really did observe some unconscious and nonverbal signal from an executive and then made a correct investing decision based on that signal. My all-time favorite is the UK asset manager that sold its position in a soon-to-be-bankrupt company after meeting with its chairman who was "looking unfeasibly tanned for this time of year." Perfect fact, perfect analysis. Yesterday the Financial Times had another really good one, about how Cairn Capital avoided investing in Market Financial Solutions, the now-insolvent UK bridge lender: "On the individual loan details, there was incomplete information," said Asif Godall, then the co-chief investment officer at credit fund Cairn Capital. "That's not good enough." He was also put off by what he called the "watch-to-house ratio". Paresh Raja, MFS's director, sported a Richard Mille timepiece that can cost upwards of £200,000. A contact of Godall estimated the watch was about half the value of Raja's north London home at the time. Godall decided to pass up on the opportunity to invest in MFS. Obviously I want to see an academic finance paper collecting the data, computing executives' watch-to-house ratios and checking to see if it has alpha. And obviously if you work at a quant fund and you use the watch-to-house ratio as a signal, I want to hear from you. And if you don't, why not? Anthropic in Talks With Blackstone, Other PE Firms to Form AI Consulting Venture. Deutsche Bank Flags $30 Billion Exposure to Private Credit. Cliffwater $33 Billion Private Credit Fund Redemptions Reach 14%. Janus Henderson Rejects Victory Capital Takeover Proposal, Backs Trian Deal. Tilman Fertitta in Talks to Buy Caesars for $7 Billion After Topping Bid From Icahn. Papa John's Draws Fresh Takeover Interest From Qatari-Backed Fund. Big-Budget Fish Tanks Are Taking Over America's Most Expensive Homes. Gene-Tweaked Banana Startup Tropic Secures $105 Million to Boost Output. Dishwashing Home Robot Maker Sunday Hits $1.15 Billion Valuation. "David Wick, an English and Latin teacher at Lake Forest, said [AI startup Aaru co-founder Ned] Koh seemed 'more excited about his business ventures than about Latin.'" "[A Live Nation Entertainment Inc. director of ticketing] bragged about charging '$50 to park in the grass' and ' $60 for closer grass.' 'Robbing them blind, baby,' he said, 'that's how we do it.' The company said in a statement that the messages did not reflect its values." 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! | 
