11:47 p.m. ET. A docket entry lands. O.G. Com — New Zealand-domiciled exchange, thin balance sheet, zero US footprint — has filed with the Commodity Futures Trading Commission to list perpetual futures on single US equities.
No tickers. No leverage cap. No funding interval. No dividend clause. No custodian named. No clearing member named. No contract multiplier. No settlement rail.
Four facts. That is the entire payload. A New Zealand exchange wants to bring stock leverage to "DeFi," and the CFTC is the gate.
I have covered this beat for 29 years. The interesting part of a filing is never the headline. It is the annex that isn't there. A perpetual contract on a dividend-paying equity with no dividend adjustment clause is not a derivative — it is a bet with an unpriced cash-flow leak, and the leak is the product.
Everyone is reading this as a "crypto meets TradFi" story. It isn't. It is a rulebook play. And the rulebook is the asset. Let me walk the arithmetic before the narrative eats the tape.
Context: Why Perpetuals Never Should Have Worked, and Why Equities Break Them
Perpetuals were not a crypto invention. They were a crypto improvisation that stuck.
BitMEX shipped the first viable perpetual swap in 2016. The design was a hack around a boring problem: futures expire, and rolling them costs money. So you delete the expiry. In place of convergence-at-maturity, you install a funding rate — a periodic transfer between longs and shorts that pulls the contract price back toward the underlying spot index. Hold a perp above spot, longs pay shorts, arbitrageurs short it, price compresses. Hold it below, the flow reverses.
The mechanism only works because crypto spot has no cash flows. Bitcoin pays no dividend. Ethereum does not split. There is no board, no transfer agent, no record date, no ex-date. The perp's fair value is the spot price plus a financing term and a sentiment term. That is it. Two inputs, one clock, clean math.
Equities are the opposite animal. A stock is a claim on a cash flow stream with a corporate action calendar bolted to it. Dividends hit quarterly. Splits and reverse splits re-denominate the share. Mergers convert the claim into cash, into another share, or into some blended structure. Halts freeze the tape mid-session. Borrow costs float with the locate market. None of that exists in crypto spot, and none of it is addressed in a filing that names no tickers.
Now the US legal layer, which is where this gets sharp.
Single-stock futures were illegal in America for eighteen years. The 1982 Shad-Johnson Accord — a jurisdictional truce between the CFTC and the SEC — banned them outright, largely to end a turf war over which agency owned equity-adjacent derivatives. The Commodity Futures Modernization Act of 2000 lifted the ban. Products launched in 2001 on OneChicago and NQLX. OneChicago limped along for nearly two decades and shut down in September 2020. NQLX died in 2004.
So the precedent is not a victory lap. It is a graveyard. Two venues tried regulated single-stock futures in the US and one of them survived nineteen years on institutional crumbs before folding.
That matters, because O.G. Com is not proposing a familiar instrument. It is proposing the instrument that already failed once, wrapped in an expiry-free structure that has never cleared a US equity dividend, aimed at a retail cohort the CFTC has historically treated with maximum caution.
Governance isn't a vote. It's a filing deadline. And this filing is missing the only clause that decides whether the product is a hedge or a hole.
Core: The Five Specs That Decide Everything
I pulled what was available. I checked the public dockets. I found no contract specification, no rulebook exhibit, no Federal Register notice. What I have is four claims and a jurisdiction. So here is the engineering audit of what a single-stock perpetual must solve — and what silence on each point implies.
1. The dividend leak
This is the whole game.
A perpetual has no maturity, which means no terminal convergence. Convergence must be manufactured continuously through funding. For a non-yielding asset like BTC, funding anchors cleanly to spot.
A dividend-paying equity breaks that anchor. Hold the perp through an ex-date and the spot price drops by the dividend amount. The perp, if it tracks spot, drops too. But the long perp holder never received the cash. The short perp holder never paid it. Somebody ate the dividend, and unless the contract explicitly adjusts — via a funding-rate offset, an index divisor change, or a cash credit — the long side is systematically short a known cash flow.
This is not a rounding error. On a 3% yielder held for a year, the long is down roughly 300 basis points against a spot holder before leverage. That is a structural, predictable bleed, not a market risk.
A perpetual that doesn't price dividends is a perpetual that slowly transfers money from longs to shorts every quarter. Any market maker worth their margin will price that in on day one and quote a spread wide enough to make retail pay for it.
Does O.G. Com's contract handle it? Unknown. Not disclosed. Which, given that the exchange chose to announce the product, is the single most telling omission in the entire story.
2. Splits, reverse splits, and the multiplier problem
A 4-for-1 split cuts the share price by 75%. A perpetual tracked to "the stock" must either adjust its contract multiplier or accept a 75% cliff.
Crypto has nothing analogous. Token splits exist but are social conventions, not legal events. Equity splits are corporate actions with ex-dates, transfer-agent records, and exchange-mandated adjustments.
Without a disclosed corporate-action policy, the contract either inherits the split mechanically — meaning the perp's notional resets — or it doesn't, meaning the perp decouples from spot by 75% overnight and the funding rate spirals trying to reconcile an irreconcilable gap.
No spec, no answer. And in derivatives, no answer is an answer.
3. Halts and the oracle gap
US equities halt. LULD bands stop trading when a stock moves too far too fast. Market-wide circuit breakers halt everything at 7%, 13%, and 20% on the S&P 500. Pre-market and post-market sessions are thin, gappy, and structurally different from the 9:30–16:00 window.
A perpetual trades 24/7. That is the entire selling point in crypto. In equities it becomes a liability, because the contract keeps quoting while the underlying is frozen.
What does an equity perp do at 3:12 a.m. ET when the underlying venue is closed and the next print is nine hours away? It needs an oracle. Where does that oracle come from? A consolidated tape that isn't publishing? A last-sale snapshot that is nine hours stale?
I have audited oracle designs. In 2021 I mapped slippage mechanics inside an NFT marketplace integration by running high-frequency trades against the pool itself, and the finding that mattered was not the arb — it was that pricing inputs were structurally lagged. Same failure mode, same shape, larger stakes. An oracle that quotes a halted stock is not an oracle. It is a liquidation engine with a delay fuse.
O.G. Com has disclosed no oracle design, no halt policy, no overnight pricing methodology. That is not a detail. That is the mechanism.
4. The funding rate under equity borrow
Crypto funding rates anchor to spot. Equity financing anchors to the stock borrow market — the locate fee, which can run from a few basis points to triple-digit annualized on hard-to-borrow names.
A perpetual on a hard-to-borrow equity has no clean way to discover the borrow cost. If funding stays low, shorts get a free ride on an expensive-to-borrow name and the venue bleeds arbitrage. If funding tracks borrow, the venue needs a borrow data feed that doesn't exist as a public 24/7 index.
So the exchange either eats the basis or misprices it. Both are survivable. Neither is disclosed.
5. Settlement, custody, and the DeFi contradiction
Now the part everyone is skipping.
A CFTC-approved product listed on a registered entity runs through a designated contract market or a foreign board of trade path. It has margin, it has a clearing member, it has KYC, it has surveillance, it has a rulebook. It is, by construction, permissioned, intermediated, and custodial.
That is the opposite of DeFi, which is permissionless, non-custodial, and anonymous by default.
"Bringing stock leverage to DeFi" and "subject to CFTC oversight" are not two features of one product. They are two mutually exclusive architectures, and the filing has to pick one. It cannot pick both, because a permissionless front end cannot know who is trading, and a CFTC-regulated DCM cannot pretend it doesn't.
The likeliest reality — and I want to be explicit that this is inference, not disclosure — is that this is a centralized, KYC-gated, US-facing derivatives product wearing a DeFi label as a marketing layer. That is not a crime. It is a pitch. But the pitch and the plumbing are in conflict, and regulators read plumbing.
The jurisdictional wrinkle nobody flagged
Single-stock futures in the US are jointly regulated. The Shad-Johnson framework, as amended by CFMA, put SFPs under CFTC jurisdiction for trading and SEC jurisdiction for the underlying securities and the margin-setting role. A filing that only engages the CFTC is either an oversimplification in the announcement or a structural misreading.
Also worth noting: DCMs in the US typically list new products via self-certification under the Part 40 rules, not via case-by-case "approval." A foreign exchange serving US customers would more likely pursue the Part 48 foreign board of trade route. "Applied for CFTC approval" is the language of press releases, not the language of the Federal Register.
That gap between the vocabulary and the procedure is the first thing I'd verify.
Contrarian: The Moat Isn't the Product. It's the Clause.
The consensus read is that this is a small exchange swinging above its weight, and that the winner of this race will be whoever ships the best UI or the deepest liquidity.
Wrong axis.
The prize here is not order flow. It is the settlement standard.
Whoever gets the first CFTC-blessed single-stock perpetual through the gate writes the template for how every tokenized equity in the world handles dividends, splits, halts, and borrow. That template becomes the reference implementation. Every subsequent applicant copies it, because copying an approved structure is cheaper than inventing a new one and praying.
Under that frame, the missing dividend clause is not a drafting oversight. It is the negotiating surface. The clause is where the value sits, and it is being held back because it is the part competitors would lift.
I've seen this pattern before. In 2020, watching governance proposals on a lending protocol spike before the public announcement, the tell wasn't the vote — it was the emergency parameter hidden inside it. The mechanism was boring. The parameter was the whole story. Same shape here. The signal is in the silence.
Second contrarian point: the "DeFi" label is a liability, not an asset.
Regulators do not read "DeFi" as "innovative." They read it as "unhosted wallets, no KYC, no travel rule, no surveillance." In a filing, that word raises the approval cost. The pitch that clears a US gate is the one that never says DeFi out loud. If O.G. Com is leading with it, either their legal team isn't in the room, or the DeFi framing exists purely to capture retail narrative attention while the actual application is dressed in CeFi clothing.
And a third thing: any depth this product shows at launch will not be organic. It will be market maker inventory, funded by the venue, quoted wide because the venue hasn't published its corporate-action policy and no desk will price a book against an unknown cash-flow leak. Liquidity doesn't lie. Paperwork does.
Takeaway: Watch the Federal Register, Not the Press Release
Five things to track, in order of information value.
One — whether a Federal Register notice or a Parts 40/48 filing actually appears. Until then, this is a press release with a jurisdiction attached.
Two — which agency leads. If SFPs are genuinely involved, SEC involvement should surface. Silence from the SEC is itself a data point.
Three — the first disclosed corporate-action policy. The dividend clause is the product.
Four — leverage caps. The CFTC's historical posture on retail derivatives says the cap will be lower than anything an offshore venue offers, which means the "leverage for the people" pitch deflates on contact.
Five — follow-on filings. One exchange and it's noise. Three exchanges and it's a regulated category forming, and that is when the tape actually moves.
Approval is a stage, not an outcome. Filing, acceptance, public comment, certification, listing — five doors, and the industry habitually treats door one as door five.
I've watched a lot of "submitted" turn into nothing. I've also watched one filing become a decade of rulebook. The difference is always in the annex.
So here's the question I'll be holding until the docket updates: if the dividend clause is the most valuable clause in the document — and it is — whose name ends up on the patent for it?
Because that, not the ticker list, is what this filing is actually competing for.