2,500 Filings, Zero Denials — Notes on Odd Lots with CME's Terry Duffy

Contents

1. 1980: An Agency for Farm Products
2. 2008 to 2010: Principles Rewritten as Text
3. 2026: Hot Dogs, a Congressman, and Core Principle Three
4. The Line Being Fought Over: Are Perps Futures or Swaps?
5. Watch the Banks, Not the Exchanges
6. If the Rules Loosen, How Should a Retail Reader Read It?
7. An Expiration Date Is a Device That Forces You to Check Your Answer
8. Worth a Look
9. One Thing to Take With You

A state has no permanent strength and no permanent weakness. Where the enforcers of law are strong, the state is strong; where the enforcers of law are weak, the state is weak.

—— Han Fei, Han Feizi, “On Standards” (Warring States period; my own translation)

On October 20, 2026, Bloomberg Odd Lots taped a live episode in Chicago, where Tracy Alloway and Joe Weisenthal hosted Terry Duffy, CEO and chairman of CME Group. Duffy offered one number: since the second Trump administration took office, 2,500 self-certifications have been filed with the Commodity Futures Trading Commission, and not one has been denied. He called the posture “ready, fire, aim.” One limit on that number before we build anything on it: he gave the count, not the detail on filings that were sent back for revision before clearing, so whether zero denials equals zero scrutiny is a blank the listener has to leave open.

1980: An Agency for Farm Products

Duffy joined CME in 1980 at 22, in a job that paid $56 a week. He became chairman in 2002, the same year he took CME public, and he has now spent 46 years at the exchange. Next year Lynn Fitzpatrick takes over and he moves to executive chairman for a stretch. He called it a transition, not a retirement.

He traced the CFTC back to its origin around 1972 and 1973, when it was a principles-based oversight agency for agricultural products. Then financial futures arrived, then equity indexes, then crypto, each new instrument layered onto the same frame. So the starting point for any argument about tightening or loosening isn’t a finished code of law. It’s a set of principles written for soybeans and corn, stretched to cover things that look nothing like soybeans.

2008 to 2010: Principles Rewritten as Text

The housing leverage blew up in 2008, and Dodd-Frank followed. Duffy says he testified more than thirty times during that process on one narrow-looking question: what is a swap and what is a future. Sixteen years later, he says the industry is still working out parts of Dodd-Frank.

What struck me is that those thirty-odd appearances looked at the time like a long technical argument. The line they drew is the line the back half of this episode is fighting over. A definition treated as plumbing in 2010 is the subject of litigation today.

2026: Hot Dogs, a Congressman, and Core Principle Three

Joe asked it plainly: are there any laws anymore? He said it feels like there aren’t.

Duffy answered on two levels. First, the examples he keeps coming back to despite people asking him to drop it. The CFTC approved a contract on Nathan’s hot dog eating contest — how many hot dogs Joey Chestnut would eat. He asks what the economic impact of that contract is, and how it isn’t a violation of core principle three, which bars contracts readily susceptible to manipulation. Then the sharper one: someone listed a market on whether former New York congressman George Santos would appear at the State of the Union. Santos bet against himself and didn’t show. That one isn’t hypothetical. It’s a completed sample of the manipulation.

The second level is his warning to market participants, and it’s the most valuable sentence in the hour. Rules and regs written by an agency are not acts of Congress, and the next administration can wipe them away with a stroke of a pen. He points to Trump’s own executive orders undoing Biden policies: same mechanism, available to whoever comes next. Markets grown on agency forbearance are living on borrowed time. Good regulation is regulation that survives any administration.

His position is finer than the headline. He opposes the contracts that violate the Commodity Exchange Act and core principle three; he doesn’t reject prediction markets across the board. CME is involved with Future Sports, an Australian firm that turns teams into an index: every team starts at 7,500 and moves with performance, which lets owners monetize the value of a team without selling it. His example was the White Sox, whom CME sponsored this year, in a season better than anyone expected. Had there been a vehicle to lay off the risk on a sleeve patch, he says his friend Jerry Reinsdorf could have charged him more. Traditionally a contract needs a natural buyer, a natural seller, a commercial, and a producer. That’s his dividing line: if somebody out there needs it to hedge, it’s a market; if the only content is your yes or no, it’s a bet.

He also has no patience for the argument that insider information is fine because it carries informational value. After twenty-five years of trading and running markets, he says the whole point of the technology build-out was giving everyone the same information at the same time: he doesn’t have as much money as his friend Don Wilson and can’t deploy it the same way, but he should see the same information. Then he aimed higher — the President of the United States charging $100,000 on Truth Social for faster information. A president, he says, is supposed to give information to everyone at once.

The Line Being Fought Over: Are Perps Futures or Swaps?

This is the technical core, and there’s litigation attached.

The CFTC approved Kalshi to list bitcoin perpetual futures. A perp holds its price to spot through a funding rate: longs and shorts pay each other. Duffy’s argument is clean — any time two parties pay each other, the definition says swap, so perps are swaps. The label isn’t cosmetic. Futures get Section 1256 treatment, 60% long-term and 40% short-term. Swaps are taxed at ordinary income. Swaps also carry five days of margin to clear, ten if uncleared, dealer registration with the government, and a compliance stack behind it. He calls the approval a way around Congress.

The second problem is the absence of an expiration date. People focus on how convenient it is never to roll, but corporate users can’t manage risk that way. There’s $40 trillion of debt out there, and corporations lend against futures and cash positions with their customers; those arrangements need a date certain. Meanwhile the funding rate can mark once a day, twice a day, or hourly — who decides? Institutions look at that and say it’s untradeable.

So CME didn’t list a 24/7 perp. It filed for a 24/7 oil contract on 10 barrels, a small spec, to compete with overseas platforms that are supposed to be off limits to U.S. participants — except, as he puts it, a phone with a VPN gets you in, and nobody is policing it.

On Hyperliquid, he says he’s talked with founder Jeff Yan and thinks he’s brilliant, but Yan’s position on the show — that the platform is plumbing and the deployers and users are what should be regulated — doesn’t comply with U.S. law. His list of problems: no KYC, so nobody knows who is on the platform; auto-liquidation on chain; high leverage products, running from the low teens up to 250% on some. At 250%, a tick or two takes someone out, and taking him out takes out the next guy, cascading in either direction. He cites core principle nine, fair and ready market access, to argue the institution itself carries a policing obligation that can’t be pushed onto users.

Tracy played devil’s advocate: Kalshi caps leverage around six times, and the CFTC chairman has said a U.S. launch would come with leverage limits. Isn’t a cap enough? Duffy says he doesn’t know what those limits actually are. Then he inverts the question: perps are being brought to the U.S. for leverage. If that isn’t the motive, why are firms seeking leverage on prediction markets and sports contracts, with single-stock perps next on the list? Retail participants, himself included, don’t stay out of a trade because they’re worried about expiration or a roll date — they’re in and out. And the funding rate is tied to spot, not to the futures price.

A contrast in the same stretch is worth keeping. GPU futures are being held back. He says that market is already over a trillion dollars, crypto was deemed an asset class at under three trillion, and GPUs will be multiples of crypto by 2030. The administration cites national security and he says he has no idea what that refers to — if the country wants to lead in artificial intelligence and data centers, the compute and components need a risk offset. The irony: Kalshi runs a prediction market on compute and wasn’t made to take it down.

Watch the Banks, Not the Exchanges

Tracy asked who wants 24/7 trading besides the people placing orders at 2 a.m. and the exchange staff who have to stay awake for them.

Duffy has been saying 24/7 is inevitable for a while. You can dislike it, you can say you don’t want to sit at a desk at two in the morning, but risk doesn’t know what time of day it is. The real bottleneck he points to is banking. U.S. cash equities are a $76 trillion market, and with banks closed over a weekend, nobody on the planet can pre-fund $76 trillion on a Friday. Crypto is small enough to pre-fund, which is why CME trades it around the clock. Rates and equities wait on a banking system willing to run 24/7. He mentions real-time payments and other Fed work, and an hour and a half with Chair Atkins on Tuesday, and still calls it not enough.

He did leave a near-term checkpoint: cash equities going to 23.5 hours by year end, with Nasdaq close to launching. A first. He expects pockets of illiquidity, then adds that there are pockets of illiquidity intraday today, which is nothing new for the industry.

If the Rules Loosen, How Should a Retail Reader Read It?

The question in most people’s heads here is whether having more things to trade is good for them. It sounds good — more choice.

Duffy’s test travels well: ask who is on the other side of the contract. If somebody needs it to hedge, there’s external demand and the price has something to anchor to. If the only content is your yes or no, the spread and the fees are the entire economics. You don’t need derivatives pricing to use that test. You need one question: who naturally needs this?

The second lesson hides in the Santos example. Whether a market can be manipulated turns on whether the people who determine the outcome can also take the other side. That’s visible in sports and politics contracts, and the same problem lives in financial news. The good news you just read — how many people could have known it before you? Duffy made it concrete with the $100,000 early feed. My own habit is to back out how long a piece of news could have been knowable, and the answer often changes what I’d pay for it.

An Expiration Date Is a Device That Forces You to Check Your Answer

The more everyday pain is the other one: I’m underwater, do I cut or hold? Entering comes with reasons. Exiting comes with feelings.

Duffy told his own story without padding. He went broke in 1981, lost $150,000, didn’t have a dollar, and the $150,000 was borrowed. He says it taught him discipline. Speaking at Georgetown, he pointed at the EXIT sign on the wall: anybody can get in, but before you get into something you should know how to get out of it. Then the most honest line in the hour — it’s hard to accept being wrong, and human instinct doesn’t like being wrong.

Put that next to the perp discussion and the two halves meet. The convenience of a contract with no expiration is that you never have to settle, which means you never have to admit you were wrong. An expiration date is uncomfortable because it forces you to check your answer. That discomfort is the function.

Worth a Look

• Bloomberg Odd Lots, October 20, 2026, taped live in Chicago, with Terry Duffy of CME Group
• The earlier episode in the same thread: Hyperliquid founder Jeff Yan, whose position is cited repeatedly here
• Core principle three (not readily susceptible to manipulation) and core principle nine (fair and ready market access) under the Commodity Exchange Act
• The Dodd-Frank definitions of swap and future, and the 60/40 treatment under Section 1256
• The CFTC self-certification process, and how far review goes before a contract lists

One Thing to Take With You

One idea: an expiration date is a device that forces you to check your answer. Anything without a settlement date never turns into experience.

One thing I’ve tried: pick something you’re currently dragging out, and it doesn’t have to be an investment — a relationship that isn’t working, a course you bought and never started, a job you keep hesitating over, a project parked in the corner. Give it a date and an exit condition, two sentences: “By this date, if this specific thing hasn’t happened, I stop.” Specific enough that when you open the notebook that day, you don’t have to reason it out again. I’ve written a few of these. The hard part was never the writing. It was looking at it on the day.

Raw Markdown

This article is an educational discussion of investment method. It is not advice to buy or sell any individual security, offers no target prices, and does not analyze any current holding. Investing carries risk; make your own decisions or consult a qualified professional.