Six times leverage. Two hundred and fifty times leverage. The same word printed on both tickets.
On May 29 the CFTC approved Kalshi's bitcoin perpetual futures for listing as a futures contract. On June 18 CME Group sued its own regulator. On September 2 the CFTC moved to dismiss — not on what a perpetual is, but on whether CME has standing to ask the question. No ruling has issued.
That sequence carries more information than any argument inside it. A definitional fight that never reaches a definition has a specific shape: it resolves nothing, it bills the plaintiff, and it leaves the competitor free to keep listing. A lawsuit can be a product decision executed by other means. I have watched enough settlement schedules to stop reading the complaint and start reading the filing calendar.
The variable worth tracking is not whether a perpetual is a swap. It is 6x against 250x, and which of those two numbers the word "perpetual" is supposed to describe.
No expiry, no terminal anchor
Perpetuals came out of BitMEX in 2016. No delivery date. No terminal settlement. Price is anchored to spot by a funding rate — longs and shorts exchange periodic payments, and the sign of the payment drags the contract back toward index. Offshore venues have run the structure at volume for years. The United States has not had a compliant retail perpetual of consequence until this year.
CME's position, as stated by Terry Duffy, is that the funding stream makes the instrument a swap. Periodic exchange of cash flows between two counterparties, anchored to an external reference price — formally, that is the swap definition doing its job. Swaps carry the post-2008 rulebook: central clearing, margin requirements, a heavier compliance load.
The CFTC's answer is procedural and narrower. CME can list the same contract itself. A party that can enter the market it complains about has not been injured by exclusion from it. No injury, no standing, no merits.
Kalshi's CEO Tarek Mansour answers on leverage instead: roughly 6x, comparable to traditional futures. Not 250x. The venue is not arguing that perpetuals are riskless. It is arguing that its perpetuals are not the perpetuals in the headline.
The split runs inside the agency, not between agencies. The CFTC approved the contract as a future on May 29 while CME, a registrant of that same agency, argues the identical structure is a swap. One regulator, two outputs, no adjudication between them. That is not a gap in the rulebook. It is the rulebook working as written, which is to say badly.
The offshore layer sits outside all of this and sets the terms of the debate anyway. No US rule constrains a 250x book in a jurisdiction that has not written one, and no cross-border coordination mechanism currently exists to write one jointly. The 250x number enters the American argument not as a data point about the contract, but as an externality the domestic rulebook is being asked to internalize.
Three variables, one of which moves money
Funding as definition. Structurally, Duffy is right that a funding stream is a cash flow exchange. But daily variation margin on a standard future is also a cash flow exchange. The real distinction is convergence. A future converges at expiry. A perpetual converges forever through funding and never settles — there is no final anchor, no delivery, no terminal reference print. That is a genuine design difference and it is the strongest part of the swap case. It is also a design difference, not a risk difference. The 6x product and the 250x product share it identically.
The leverage spread. If a legal category is doing protective work, the category should map onto risk. Here it does not. Both venues sell "perpetuals." The 250x figure is a property of a margin schedule, not of a contract type. Offshore venues set 250x because nothing stops them. Kalshi sets 6x because it wants the license. Duffy's number is a public-policy argument, and a good one, but it is not a structural argument, and the two should not be blended in a complaint.
A note on the mechanics, because the comparison is usually made sloppily. Kalshi's margin schedule is built to sit alongside CME's own futures margin — same order of magnitude, same liquidation logic, same clearing assumptions. The offshore 250x schedule is built for a venue where no clearing member stands between the trader and the book. Those are different products wearing the same noun, and the noun is what the lawsuit is about.
Collateral and tax. This is the variable that actually moves capital. Futures status changes how customer collateral is held and protected, and it supports futures-style tax treatment — broadly, marked-to-market with a blended rate, against ordinary treatment on the other side. Swap status routes the same collateral through a different regime with different segregation mechanics. The tax treatment is separately unresolved, which means the definitional question has a second-order question stacked behind it. At institutional size, that stack is worth more than the label.
I have run this reconciliation before. In late 2022 I spent three weeks matching FTX's public wallet addresses against its stated reserves and found $1.8 billion that was not where the ledger claimed it was. The failure mode was not a bad definition. It was collateral reclassified by practice rather than by rule, with nobody checking the practice. Classification is not a label. It is a collateral schedule. Every definitional fight in derivatives is a collateral fight wearing a nomenclature costume.
Standing. If the court grants the CFTC's motion, perpetuals remain listed under a classification no court has adjudicated. Any registered exchange can do the same, and the precedent is durable because it is an omission — nothing gets overruled, nothing gets affirmed. CME would lose the rule it came for and pay for the privilege. If the court denies, the merits get heard, and CME has to live under whatever it wins, including swap treatment for its own eventual listing. Duffy is arguing for a definition he would then have to operate under. That is a double-edged position, and it is the main reason I read this suit as competitive defense rather than product ambition.
What the other side got right
The 6x product is not a loophole. It is the conservative end of a spectrum that already exists, and declining to price it as if it were 250x is correct. More than that: the CFTC has already defined the instrument, in practice, by approving one. Whatever a court eventually says, the operative definition in the US market is currently "whatever the commission approves." That is a weaker definition than a statute, and a stronger one than a comment letter.
Kalshi is also right that the classification fight is being argued on the wrong axis. A venue that self-limits to 6x has already done most of the risk work the rule is supposed to do; what remains is paperwork and a jurisdiction label.
CME's own disclosures concede the shape of the dispute. Ninety-four percent institutional volume in the first half. Clients not asking for perpetuals. If your book does not want the product, your lawsuit is not about the product. It is about who writes the listing rules for retail flow you are not currently serving. Duffy's comments came after a record Q3 — a fine moment to signal rule-shaping capability to shareholders, and a strange one to discover a retail-protection emergency.
Trust is a variable I refuse to define. So I price the incentives instead. The offshore 250x number is the strongest public-interest argument in this case, and it is being advanced by a party with no retail perpetual book. That does not make it wrong. It makes it a position, not a finding.
What to watch
Volatility is just liquidity leaving the room, and definitions are how the room gets locked. The docket is the slowest signal available. Product filings are faster. If CME files a perpetual of its own, the suit was theater. If it does not, the suit was the product. Everything else — the briefing, the standing motion, the swap-versus-futures framing — is downstream of that one filing, and the market is currently watching the wrong document. The label will arrive after the market has already picked one.