Swap or Future? A Field Guide to CME v. CFTC

On 18 June the Chicago Mercantile Exchange filed suit against the Commodity Futures Trading Commission, the agency that licenses and oversees it. The complaint asks a federal court in Washington to undo the CFTC’s 29 May decision to let Kalshi list a bitcoin perpetual as a future, and to strike down the policy statement that invited every other exchange to do the same (complaint, Chicago Mercantile Exchange Inc. v. Selig, No. 1:26-cv-02157 (D.D.C.), Courtlistner; the order it challenges is at CFTC Filing, and the policy statement at 91 Fed. Reg. 33,160, Policy).

The whole case turns on a single question.

Is a crypto perpetual a swap or a future under the law?

One word, and everything downstream follows from it.

This edition is a guide to that case. Not a prediction of who wins, a map of what is actually being argued and why it matters well beyond one contract on one exchange.

A perpetual, in plain terms

Start with the product, because the law cannot be followed without it.

A normal future has an expiry. On a set date it settles, the position closes, and the contract is pulled to the spot price because it has to be. A perpetual has none of that. It never expires, nothing is delivered, and there is no settlement date. You can hold it indefinitely.

What keeps a perpetual tethered to spot is the funding rate. Every few hours one side of the market pays the other. When the perpetual trades above the spot price, say bitcoin is at 50,000 and the contract is at 50,500, the longs pay the shorts, which gives traders a reason to sell the contract back down toward spot. When it trades below, the shorts pay the longs. That payment is the convergence mechanism. It does the job an expiry date does for an ordinary future, continuously, with money rather than a deadline. Margin and liquidation run against a mark price, an index of the underlying spot market rather than the last traded tick, so a single print cannot set off a cascade. (The contract’s mechanics are set out in the CFTC’s own approval order, CFTC Filing; for a plain-language primer see FIA, “Explainer: Perpetual futures contracts,” FIA Explainer.) At least that is the theory.

The idea is older than crypto. Robert Shiller set out the structure in 1992 as a way for homeowners to hedge house prices without an expiry (Shiller, “Measuring Asset Values for Cash Settlement in Derivative Markets: Hedonic Repeated Measures Indices and Perpetual Futures,” Journal of Finance, 1993, Wiley). A venue called ICBIT listed the first crypto version in 2011 and coined the word funding, and BitMEX industrialised it in 2016 (on the origin and spread, Wikipedia). The perpetual is now the most traded instrument in crypto, with annual volumes past 60 trillion dollars in 2025 (Coindesk Article). The feature to carry into the rest of this piece is simple. A perpetual swaps an expiry date for a continuous payment. Every legal question in the case grows out of that one substitution.

Two boxes, and why the box is the whole game

American commodities law has two boxes that matter here, and a product sits in one or the other.

The first is a swap. Dodd-Frank added the definition to the statute after 2008, and it is deliberately broad: an agreement to exchange payments based on the value of a commodity, transferring the financial risk of a future price change, without conveying any ownership interest in the asset and without a date certain (7 U.S.C. § 1a(47), US Code).

The second is a contract of sale of a commodity for future delivery, what everyone calls a future. To be a future, an agreement has to provide for delivery, of the commodity or its cash value, at a particular time in the future (7 U.S.C. § 2(a)(1)(A), US Code). Futures are written out of the swap definition. A product cannot be both.

This sounds like a librarian’s question. It is not, because the two boxes carry completely different rulebooks. Put a product in the swap box and the venue and its dealers face swap dealer registration, swap data reporting, a separate customer collateral regime, and, at the clearing house, an initial margin sized to a five day close-out under the CFTC’s margin rule. Put the same product in the futures box and none of the swap machinery applies, the clearing margin is sized to a one day close-out (the floors are in 17 C.F.R. § 39.13(g)(2), US Code), and the contract qualifies for the favourable Section 1256 tax treatment that futures receive and swaps do not (26 U.S.C. § 1256, US Code).

So the same economic contract, depending on which word a regulator chooses, attracts a different capital cost, a different compliance burden, and a different tax bill. That is why anyone is in court. The label is not a description of the contract, it is the entire commercial difference between two regimes.

How the agency got here

For most of the past decade the CFTC’s own position was that a perpetual is a swap. It said so in court, repeatedly, when it brought enforcement actions against the offshore venues, BitMEX, Binance, Mango Markets, Deridex and KuCoin, each time describing the perpetual as a swap under the same statutory definition (the complaint collects all five, Courtlistener). In April 2025 it opened a formal request for comment that asked the question directly, swap or future, and whether a future even needs an expiry (CFTC, “Request for Comment on the Trading and Clearing of ‘Perpetual’ Style Derivatives,” Apr. 2025, CFTC Perps). More than 150 comments came in (CFTC Comments). No rule followed.

Then, on 29 May 2026, the agency moved all at once. It approved Kalshi’s bitcoin perpetual as a future (Kalshi CFTC Filing). It issued a policy statement inviting other exchanges to self certify similar perpetuals on the same basis (US Policy). It gave Coinbase a no action position to reach American customers with perpetuals through its foreign affiliate Deribit (CFTC Letter 26-17, CFTC). And a fortnight later it let two exchanges strip the expiry dates off existing contracts to turn them into true perpetuals (CFTC Letter 26-19, CFTC). The Kalshi approval came one day after Kalshi applied.

Two facts about that approval are worth stating plainly, because both sides lean on them. The Commission acted through a single sitting commissioner, its chairman, the other seats being empty. And the order approving the contract does not use the word swap once.

CME’s case

CME filed in the federal district court in Washington, represented by Jenner and Block and led by a former acting Solicitor General. It sued the CFTC and its chairman, not Kalshi. Its standing rests on competitive injury: the agency has cleared a rival to compete for CME’s retail crypto customers on terms CME says are unlawful, a swap dressed as a future and handed the lighter regime that goes with it (complaint, Courtlistener).

The complaint runs on two tracks. The first is that the approval is simply contrary to the statute. A perpetual fits the swap definition and lacks the one thing a future requires, delivery on a date certain. CME points to the courts that have said as much, to the CFTC’s own five enforcement actions, and to a neat structural tell: a real future generates a forward curve, a set of prices for delivery at successive future dates, and a perpetual has no forward curve at all, because it has no future dates. It also heads off the obvious rejoinder. Being standardised, exchange traded and centrally cleared does not make something a future, and CME notes that Kalshi self certified its own event contracts as swaps even though they too are standardised, listed and cleared.

The second track is administrative. Even if the question were close, CME argues, the agency decided it the wrong way: it reproduced Kalshi’s own analysis, never grappled with the statutory definition, never explained why it was reversing a position it had held for years, and resolved a matter it called novel and complex in a single day with no public comment.

CME asks the court to vacate the order and the policy statement and to declare that crypto perpetuals are swaps. One detail tells you how carefully the suit is drawn. CME does not attack perpetual style products that keep an expiry date. Its target is precisely the no expiry feature, nothing more. That keeps its own dated crypto futures, and a good deal else, safely outside the line of fire.

The other side of the argument

The CFTC’s case, which is mostly the reasoning of the approval order itself, does not depend on a date certain at all (order, CFTC). It applies the test the courts have actually used for newer, intangible products, asking whether there is trading in the contract: is it standardised and fungible, are the obligations guaranteed by a clearing house that sets margin, can a holder exit by offset. On that test a perpetual looks like a future. And on the expiry point the order leans on a line of cases, one of which is Chicago Mercantile Exchange against the SEC, for the proposition that futurity means value set in the future, not a fixed final date, so a contract of indefinite duration can still be a future. The agency is, in other words, using a precedent that carries CME’s own name.

The order then makes a market structure argument rather than a legal one. The funding rate can stand in for terminal convergence, it reasons, because bitcoin’s spot market is deep, continuous and broadly distributed, which keeps the reference price reliable and hard to push around. That is framed as a question of manipulation resistance, one of the core principles a contract market has to satisfy.

The agency’s procedural footing is a narrow one. The statute tells it to approve a new contract unless it would violate the law, which is a gate, not a grant of discretion (7 U.S.C. § 7a-2(c)(5)(B), US Code). And since the Supreme Court ended automatic deference to agencies last year, the court will read the words swap and future for itself, with no thumb on the scale either way (Loper Bright Enterprises v. Raimondo, 603 U.S. 369 (2024), Supreme Court). Publicly the CFTC has been blunter, calling the suit frivolous and litigation in place of competition (The Block).

Why this is bigger than one contract

It would be easy to read the case as a fight about a single bitcoin contract. It is not, on three counts.

The first is that none of this is new. A no expiry, continuously funded future is a thirty year old idea that has passed through regulated markets several times. I have my own history here. In 2013 I filed a patent, with Vijay Angelo, for a constant maturity future, an interest rate swap future with no fixed expiry that was marked each day to a constant maturity index, with a daily payment between the two sides to hold the maturity in place (US Patent Application 2015/0026028 A1, USPTO). That daily payment is the same mechanism a crypto funding rate uses, three years before BitMEX, and it was built to be centrally cleared from the start. It is not the only example. Eris swap futures have been listed as futures since 2011, and CME itself lists them today (CFTC order designating Eris Exchange, 28 Oct. 2011, CFTC; CME Group, “CME Group to List USD Eris Interest Rate Swap Futures,” 2018, CME Group). The rolling spot currency future, a daily rolled contract with no fixed settlement, was pioneered by CME in 1993, and CME even holds a live patent on perpetual futures with periodic funding (both are described in that patent, US 8,473,402, USPTO). The question of whether a no expiry product is a swap or a future is an old one, and the institutions that built the product family are the very ones now arguing about it.

The second is that the same machinery is being tested on a second front. Kalshi lists its sports event contracts the way it listed the perpetual, by self certifying that they comply with federal law and relying on the CFTC’s exclusive jurisdiction to keep state regulators out. That claim is in front of the courts right now. A divided federal appeals court has sided with Kalshi (KalshiEX v. Flaherty, No. 25-1922 (3d Cir. 6 Apr. 2026), Federal Court), a state supreme court is weighing it, and a third appeal is pending. In those cases the CFTC argues the contracts are swaps (CFTC amicus brief to the Massachusetts Supreme Judicial Court, CFTC). So the agency is telling one set of courts that Kalshi’s products are swaps while telling the market that Kalshi’s perpetual is a future. The common thread is not the label. It is the claim that whatever a registered exchange chooses to list falls under the CFTC’s umbrella.

The third is simply the map. Perpetuals are already trading onshore. Bitnomial listed the first in 2025 (Bitnomial Press Release), Kalshi has the first the CFTC formally approved, Coinbase reaches them through Deribit, and Kraken now clears through Bitnomial (Coin Desk). The one large venue not on that list is the one that brought the suit.

What the label actually changes

Strip the law away and two things are genuinely at stake, and both sit in the clearing layer, which is where this newsletter spends its time.

The first is how the position is margined and how losses are absorbed. As a future, Kalshi’s perpetual clears the way every regulated future clears. The customer faces a clearing member, the clearing member faces a registered clearing house that stands in the middle by novation, and the positions and collateral sit in a segregated futures account with the clearing house’s own default arrangements behind them. The clearing house sizes initial margin to a one day close-out, the floor the CFTC’s rules set for futures, where the same contract treated as a swap would carry a five day floor (17 C.F.R. § 39.13(g)(2), US Code). That is the practical edge of the swap or future question at the clearing house, and it is exactly what CME is pointing at.

Look at what the offshore version does instead, because it is the thing onshore clearing replaces. On the offshore venues there is no clearing member and no central counterparty in the regulated sense. The venue holds the collateral, and when a default outruns the liquidation and the insurance fund, the loss is taken by auto deleveraging: the venue reaches across the book and closes out the most profitable traders on the other side to cover the hole. The losses fall on whoever happened to be winning. There is no mutualised default fund and no waterfall in the clearing sense. The October 2025 cascade, where insurance funds were exhausted and several venues deleveraged at once, is what that looks like under stress (Coindesk).

The second thing at stake is subtler, and it is where the legal fight and the maths meet. CME’s strongest point is textual: a perpetual has no date certain, and the statute asks for one. But the finance does not quite agree that a perpetual has no future delivery. Worked through properly, and this is now in the peer reviewed literature, a perpetual prices as a future on the spot price with a random maturity, an expected settlement date set by the strength of the funding rate rather than a fixed one (Ackerer, Hugonnier and Jermann, “Perpetual Futures Pricing,” NBER Working Paper 32936, NBER). The funding rate is, in effect, a clock that runs at a variable speed. So the thing the statute wants, a date certain, is exactly what the perpetual replaces with a distribution over dates. That is the real centre of the dispute, underneath the briefs. Whether the law can treat an expected settlement as the settlement it was written around is the question the court will, in its own language, have to answer.

Where it goes from here

A few things are worth watching. The case itself, in Washington, on a pure question of statutory reading. Whether the parallel fight over sports contracts produces a split between appeal courts and pulls the jurisdiction question up to the Supreme Court with it. Whether the empty seats on the Commission get filled. And whether the policy statement’s open invitation to self certify fills the market with perpetuals before any court has ruled on whether they belong in the futures box at all.

A word on where we sit. DCN is not a party to this. We are not the plaintiff, not the defendant, and not an exchange trying to list the product. The court will settle the label. The question we actually care about is the one underneath it, and it does not change with the verdict: how do you clear a contract that never expires and is funded continuously, so that the margin is sized to real risk and the losses, when they come, do not land on the traders who happened to be right.

That is the clearing gap, and it is the same gap whichever box the perpetual ends up in.

This first appeared on LinkedIn on 22 June 2026. If you want to comment or discuss, that’s the place.