Markets close. That is not an opinion; it is the first line of every equities compliance manual. The New York Stock Exchange has a closing auction. The Nasdaq has a closing cross. Regulators, clearing houses, prime brokers and risk teams all built their settlement rails around a single assumption: at some point, the tape stops. Coinbase is now submitting a regulatory framework that would make that assumption optional. The company has filed a notice registration with the SEC while seeking CFTC approval to offer perpetual-maturity derivatives on individual US stocks, available to US customers around the clock.
Most coverage reads the filing as a headline: Coinbase wants to trade Tesla and Apple like Bitcoin. The trade, the reporting suggests, is already priced. Good. Markets are efficient until they are not. What the commentary misses is that this product sits in a jurisdictional crack that no court has fully sealed, and that the harder constraints are not legal at all. They are structural. They are about hedging windows, funding mechanics, corporate actions, and the simple fact that the underlying asset does not trade at 2 a.m. Volume screams, but liquidity whispers the truth.
I have spent enough time on both sides of this market to know that the most dangerous derivative is not the one with the highest leverage. It is the one whose reference price exists only in the imagination of its own order book. That is the problem nobody in the press conference wants to explain.
The Filing That Is Not a Product
Let me start with what is actually known. Coinbase operates a Bermuda-based derivatives venue that has already received a non-action letter from the CFTC. That letter allowed the offshore platform to serve certain non-US customers without immediate enforcement action. A non-action letter is not a regulation. It is a promise from a current enforcement staff that they will not recommend action today. It can be withdrawn tomorrow, by the same staff or by the next one. Anyone who built a business on a no-action letter has built a business on a courtesy.
Now Coinbase wants to bring the same product into the United States. The product is a perpetual futures contract, or perp, referencing the price of an individual American equity. The contract is cash-settled, margined, and typically allows leverage. It has no expiration date and no physical delivery. It is the crypto-native template that drove billions in volume on venues like Binance and Bybit, transplanted onto the most regulated equity market in the world. The application is a test. It asks a question that no agency has definitively answered: when a derivative references a security but trades like a commodity, who owns it?
The SEC regulates securities. The CFTC regulates futures. A perpetual futures contract on a single stock is, in plain language, a future on a security. Under the Commodity Exchange Act and the Securities Exchange Act, a contract of that kind is called a security future, and it has historically required joint oversight from both agencies. That joint framework was designed for products that expire, settle, and clear through traditional channels. A perpetual contract does not expire. It is marked to a reference rate and rolls indefinitely. It synchronizes long and short positions through a funding payment rather than a settlement event. The legal infrastructure does not have a clean drawer for it.
The result is a regulated entity deliberately walking into the gap between two regulators. Coinbase is not asking for a single yes. It is asking for two agencies to agree on a categorization that will define the product's existence. If the CFTC treats the contract as a commodity future, the SEC may still treat it as a security-based swap. If the SEC grants a registration notice, the CFTC can still decline to approve the venue. The filing is not a product. It is a diplomatic negotiation, submitted in writing.
What the Approval Path Actually Requires
The compliance path Coinbase has chosen is not the only one available, but it is the one that maximizes optionality. By filing a notice registration with the SEC and separately engaging the CFTC, Coinbase creates a record where neither agency can claim ignorance. That is an advantage inside Washington, where jurisdictional disagreement is often resolved by whoever moves last. It also creates a trap. If the SEC stays silent while the CFTC acts, a court may later find that silence was not consent. If the CFTC issues a no-action letter for the domestic venue, that letter will be cited by every future applicant as evidence of a new standard. If it is rescinded, Coinbase absorbs the reputational damage and the capital cost.
There is a historical precedent that every serious trader should remember. In the late 1990s, Congress created the security future category to break the SEC-CFTC logjam. The result was not a vibrant market for single-stock futures. It was a decade of regulatory dead letter. Traders avoided the instrument because every trade required the approval of two masters, two rulebooks, and two enforcement philosophies. The product that was supposed to merge the markets ended up serving neither. Coinbase is not building a new category. It is reopening an old wound and betting that the political climate has changed enough to let it heal. That bet may be correct. It is not guaranteed.
The Technical Architecture Is the Real Filing
Let me move past the lawyers and look at the machine. Crypto exchanges have spent a decade building matching engines that run twenty-four hours a day, seven days a week, without a closing auction, without a circuit breaker, and without a designated market maker stepping in to restore order. Those exchanges process funding payments every one, four, or eight hours. They liquidate underwater positions in milliseconds. They monitor margin in real time because there is no end-of-day mark to hide behind. That technological base is real. It is the reason Coinbase can make this filing at all. Traditional brokers do not run their equity risk systems at 3 a.m. The crypto-native stack does.
In DeFi Summer 2020, I ran automated yield farming strategies across Aave and Compound. My execution logic was a set of Python scripts, standardized and deployed on a schedule. The strategy worked because the system acted faster than I could think. When gas prices spiked and the network congested, the scripts still fired. They executed exits before the retail crowd could click a button. That experience taught me a permanent lesson: in a 24/7 market, the speed of your code is your risk management. Coinbase is applying the same logic to equities, and the logic is sound—until the underlying market closes.
Here is the structural problem the press releases do not address. A perpetual contract on Bitcoin does not have a counterpart in the traditional financial system that closes. Bitcoin trades somewhere at all times, and the perpetual references an index that is continuously observable. A perpetual on Apple is different. The corporate equity market closes at 4 p.m. Eastern. There is no official Apple price at 2 a.m. There is no continuous auction, no SEC-regulated tapes, no official close. The reference price for the derivative must be built from something else. That something is typically the venue's own order book or a composite of third-party feeds from offshore venues. Once the reference price becomes endogenous, the contract stops being a derivative of the stock and starts being a derivative of its own liquidity. That is the moment where manipulation enters.
In 2021 I analyzed on-chain data for a thousand NFT projects and found that eighty percent of apparent floor price strength was generated by wash trading. The metric everyone watched was volume. The metric that mattered was unique holder count. The same discipline applies here. The first question I ask about any new derivative venue is not whether it has volume. It is whether the price can be verified against an independent, continuous, honest market. For single-stock perps listed in the United States, that market closes every night. In the void of 2017, only structure survived. In the void of 2 a.m., only the venue's own matching engine is speaking.
Hedging and the Hidden Basis
There is another technical issue that separates a real market from a casino: hedging. A market maker who sells a perpetual on Apple at a price the customer is willing to pay does not want to hold that inventory. He wants to buy Apple in the cash market to offset the exposure. That is how a derivative venue keeps its quoted spread tight. But the US cash equity market closes. From 4 p.m. to 9:30 a.m. Eastern, the market maker cannot hedge a US equity position through the traditional broker-dealer system. He can access futures on an index, but not the single stock. He can use a dark pool, but liquidity there is thin. He can go to an offshore venue, but that introduces basis risk between the offshore price and the official US price that will appear at the open.
The result is that the market maker will charge for the unhedgeable hours. The spread will widen. The funding rate will become the pressure relief valve. During the four hours when the underlying exchange is closed, whoever holds the wrong side of the contract will pay for the privilege of being wrong. That is not a design flaw. It is the market correctly pricing the fact that risk cannot be laid off. The question is whether retail users understand the mechanics. In my experience, they do not. They see 24/7 access as a feature. What they are actually receiving is a continuously running margin call machine with no opening bell to save them from their own leverage.
I have audited enough smart contracts to know that complexity is not a sign of sophistication. In 2017, I reviewed over forty ERC-20 token contracts during the ICO mania. I found reentrancy vulnerabilities in three high-profile projects and refused to invest until the code was patched. The same rigor applies to financial products. The complexity of a single-stock perpetual is substantially higher than a standard crypto perp because the product must handle corporate actions. Dividends, stock splits, mergers, and delistings happen on a schedule that has nothing to do with the perpetual's funding cycle. A crypto asset does not pay dividends. It does not split into two crypto assets under a legal scheme. An equity does. The software that tracks the reference index and calculates funding must be able to adjust for a 4-for-1 split without liquidating every long position in the process. That is manageable. It is also the kind of edge case that only appears after launch, when real money is on the line.
The Revenue Model and the Real Customer
The commercial case is easier to understand. Perpetual futures are the most profitable product in crypto. They generate trading fees, funding payments, and liquidation revenues. They attract professional traders who transact at high frequency and high size. The unit economics are better than spot trading because the product has built-in churn: leverage forces exits, funding forces flows, and volatility creates constant activity. Coinbase, which has seen its spot trading volumes compress in bear markets, needs a product that generates revenue even when the market is flat. A perpetual on GameStop or Coinbase's own stock would generate revenue from boredom. That is a real business model.
The target customer is not the average Coinbase retail user. It is the sophisticated trader who wants 24/7 exposure to US equities without opening a margin account at an online broker. It is also the crypto-native trader who no longer wants to trade only crypto. The customer acquisition cost is high, but the lifetime value is higher. The network effect is genuine: more traders attract more liquidity, tighter spreads, more institutional flow, and better pricing. The fragile part of the model is that the liquidity must arrive before the users do. If the first week of trading produces thin books and wide spreads, the venue will be marked as a poor execution venue and the professional traders will not return. In a market where reputation is formed in the first hundred trades, the launch is not an event. It is a verdict.
Some analysts will compare this product to Robinhood, arguing that Coinbase is trying to steal the retail equity trader. That comparison is lazy. Robinhood offers the underlying asset with no leverage and no expiration. Coinbase would offer a synthetic derivative with leverage and perpetual funding. The user who wants to own Apple stock for ten years will not buy a cash-settled perpetual, because it does not pay dividends and does not deliver the share. The user who wants to speculate on Apple's price over the next three days is a different animal. That user is a trader, not an investor. The competition is not Robinhood. It is offshore venues like Hyperliquid and Kalshi, which are already positioning themselves to serve US clients with event contracts and perpetual products. Coinbase is not entering an empty market. It is entering a market where the infrastructure has been tested outside the regulatory perimeter.
The Contrarian Read: The SEC May Be the Better Friend
The instinctive read in the crypto community is that the CFTC is the sympathetic regulator and the SEC is the opponent. That framing is too simple. The CFTC has been friendly to crypto derivatives, but it has also shown a willingness to bring enforcement actions. The SEC, for all its hostility, has a predictable process. A notice registration creates a clock. Silence is not a denial, but it is also not an approval. The CFTC, by contrast, has discretion that it can exercise without public explanation. If the regulator wants to kill the product, it does not need to sue. It can simply decline to issue the no-action letter for the domestic venue. The product dies without a trial, without a headline, and without a legal record.
There is a deeper counter-intuitive lesson. If this product is classified as a security future, it would face requirements designed to protect retail investors. Those requirements are burdensome, but they are transparent. If the product is classified as a swap, it would face requirements designed for professional counterparties. That may actually be safer for Coinbase, because it would restrict the product to accredited investors and institutions. The worst outcome for the market is a product classified as a commodity future but marketed to retail without an investor education regime. That is the path to the same disaster we saw with Terra and LUNA, where the legal wrapper was missing exactly when the mechanism failed.
In May 2022 I watched TerraUSD depeg in real time. I had an emergency protocol that had been written in 2020, and I executed it within minutes. I sold my stablecoin positions into Bitcoin and fiat without hesitation. Other traders waited, hoping the price would recover. That protocol saved me close to two hundred thousand dollars. The lesson was not that I was brave. The lesson was that I had pre-planned the reaction to chaos. Coinbase has pre-planned a product for a market that does not yet exist. The danger is not that the product fails. The danger is that it succeeds too quickly, attracting retail capital before the mechanism has been tested through a full cycle of corporate actions, a flash crash, and a funding rate that goes violently negative.
What Smart Money Should Track
I do not trade applications. I trade the signals that applications create. The first signal is the calendar. If the CFTC schedules a public comment period, the product is being taken seriously. If the desk goes quiet, the product is being killed with silence. The second signal is the venue's collateral standard. If Coinbase accepts USDC as margin, it has solved a major operational problem but created a new one: the custodian risk of stablecoin reserves. Tether dominates seventy percent of the stablecoin market and has never received a truly independent audit. The entire industry pretends this problem does not exist. If Coinbase uses USDC, which is more transparent, that is a positive signal. If it accepts a broader basket of crypto collateral, the liquidation complexity rises exponentially.
The third signal is the first weekend. The US equity market closes at 4 p.m. on Friday. It does not reopen until Monday at 9:30 a.m. That is the true stress test for a 24/7 stock perpetual. The funding rate during that stretch will reveal whether the venue can attract market makers willing to leave inventory open over the weekend. The spread will reveal whether the market is real or synthetic. A real market holds a tight spread because the market maker is hedged. A synthetic market holds a wide spread because the market maker is exposed. Volume screams, but liquidity whispers the truth.
I have no position in Coinbase stock and no opinion about its future valuation. My opinion is about the product structure. The technology is ready. The legal path is unclear. The liquidity question is open. The historical answer to that question, from the CFTC's own enforcement history to the collapse of leveraged venues, is that liquidity is not generated by filing. It is generated by trust, by verification, and by a reference price that independent participants can audit. Trust the code, verify the human, ignore the hype.
The most likely timeline is not fast approval or dramatic rejection. It is a slow, opaque process of comment letters, informal inquiries, and quiet pressure. During that process, offshore venues will keep growing, US users will keep waiting, and the gap between demand and supply will widen. Coinbase is not asking permission to enter a market. It is asking permission to define one. The answer will not come from a press release. It will come from the first 3 a.m. liquidation, the first false price, and the first time a market maker decides that the spread is no longer worth the risk.
That is the moment to watch. Everything before it is just noise.