The Taxonomy of Loss: Citadel Securities, the Single-Name Event Contract, and the Jurisdictional Ghost in the CFTC's Machine

The Taxonomy of Loss: Citadel Securities, the Single-Name Event Contract, and the Jurisdictional Ghost in the CFTC's Machine

On a Tuesday in spring, in a document that ran to fewer than twenty pages, Citadel Securities asked the Securities and Exchange Commission to do something no one in Washington had been willing to do cleanly for four years: decide what an event contract actually is. Not what it does. Not whether it is useful. What it is — in the dry, binding grammar of the Securities Exchange Act of 1934.

The request was narrow on its face and tectonic underneath. Citadel did not ask the Commission to prohibit anything. It asked the Commission to assert that contracts tied to the fate of individual public companies fall within its jurisdiction — and that the Commodity Futures Trading Commission cannot, through a procedural shortcut called self-certification, simply wave them into existence alone.

The market barely moved. That is the tell. In a sideways tape, where perpetual funding sits flat and the majors grind through a slow, unremarkable bleed, a jurisdictional dispute between two federal agencies should register as noise. It does not register as noise. It registers as the sound of a foundation being poured under the next cycle — and, for those of us who watch liquidity rather than price, as the sound of a fault line opening in the floor of the American derivatives market.

Here is the paradox worth sitting with for a moment: we spent a decade teaching markets to price everything — elections, hurricanes, the survival of a chief executive, the probability that a merger closes — and then acted surprised when some of those prices began to behave exactly like securities. We call the product an event contract because that name is legally neutral. But a name is not a taxonomy. A name is a hope.

I have spent the last several years auditing infrastructure I did not design, hunting for the stress point that everyone else politely chose not to look at. This is one of those stress points. And like most stress points, it is invisible until it fails.

The Machine Nobody Was Watching

To understand why a market maker as large as Citadel Securities would spend political capital on a product as niche as event contracts, you have to first understand the mechanism that let those contracts exist at all. It is called self-certification, and it is one of the more elegant — and more dangerous — pieces of financial plumbing in American law.

Under Section 5c(c) of the Commodity Exchange Act, a designated contract market — a DCM, in the jargon — may list a new product by certifying to the CFTC that it complies with the Act and its regulations. There is no prior approval. There is a review window, and unless the Commission affirmatively finds a violation during that window, the contract goes live. The exchange is, in effect, its own gatekeeper. It submits the paperwork and opens the door.

Think about what that means structurally. In most regulated finance, the default is prohibition: you may not offer a product until a regulator says yes. In the self-certification regime, the default is permission: you may offer a product until a regulator says no. The burden of action sits with the government, and governments are slow.

This mechanism is not a loophole. It was designed. Congress built it so that commodity derivatives could innovate at the speed of agriculture and energy markets, where hedging needs move faster than rulemaking. And for decades it worked precisely as intended, because the products flowing through the pipe — corn futures, crude swaps, interest rate contracts — sat comfortably inside the CFTC's jurisdictional lane.

Then two things happened at once. The first was the rise of prediction markets — venues like Kalshi and, in an earlier, more chaotic form, Polymarket — which treated real-world events as the underlying asset. The second was the maturation of a legal architecture, completed by Dodd-Frank in 2010, that had already drawn a sharp line between swaps (CFTC territory) and security-based swaps (SEC territory).

Dodd-Frank created that binary because it believed the two categories were exhaustive. A derivative was either a swap or a security-based swap. There was no third door. But the drafters were thinking about rate swaps and credit default swaps, not about a binary contract on whether a specific company's chief executive would survive the quarter. They did not anticipate a product whose underlying was not a commodity, not an index, and not quite a security — but the event surrounding a security.

That is the classification gap. And gaps in the law are not neutral. They are invitations. For four years, prediction-market venues have been pouring product into that gap, self-certifying their way past the CFTC's review window, and growing volumes that now rival mid-sized commodity exchanges. The CFTC, for its part, has largely allowed this — partly because the CFTC has historically been the more permissive of the two market regulators, and partly because its own jurisdiction over event contracts is defended by a standard so vague it is nearly unfalsifiable.

I have read a great deal of statutory language in my career, and I will tell you this: the phrase "contrary to the public interest" — the standard Congress gave the CFTC in Section 5c(c)(5)(C) to prohibit event contracts involving unlawful activity, gambling, or the public interest — is not a rule. It is a mood. It is the kind of phrase that permits an agency to ban whatever it finds distasteful on a given afternoon, and to defend that ban on grounds no court can easily test. That ambiguity is exactly why the fight Citadel just triggered matters, and exactly why it will not be settled by the CFTC deciding to be stricter.

The Name at the Center of the Storm

Strip away the rhetoric and the Citadel letter is a single legal argument wearing a suit: a contract whose payout is tied to a single issuer's event is not a commodity event contract at all. It is, or it may be, a security-based swap — and security-based swaps are the SEC's exclusive jurisdiction.

The operative language is Section 3(a)(68) of the Exchange Act, which defines a security-based swap to include an agreement that provides for payment based on the occurrence of an event related to a single security issuer and that directly affects the issuer's financial statements, financial condition, or financial obligations. Read that definition slowly. It is not about indexes. It is not about broad baskets. It is about the single name.

This is the pivot that everyone is missing. The public debate has framed Citadel's intervention as a question of investor protection or systemic risk or the morality of betting on corporate death. All of those frames are political. The legal frame is narrower and much more powerful: the moment an event contract's reference is one company's merger, one company's earnings surprise, one company's regulatory outcome, one company's survival, the contract stops being a neutral wager on the world and starts looking like a derivative on that company. And derivatives on a single issuer sit in the SEC's house.

The SEC already owns an older, related instrument: the single-stock future, dormant in the United States since 2001 for reasons that were themselves jurisdictional and structural. And it owns the security-based swap, the workhorse of post-crisis credit and equity derivatives. What it has never cleanly owned is the binary event contract on a corporate outcome. That is the empty chair at the table. Citadel just pulled it out and pointed at it.

But here is where the analysis becomes uncomfortable for anyone who wants a clean answer. The Howey standard — the test from SEC v. W.J. Howey Co., 328 U.S. 293 (1946) — asks whether there is an investment of money in a common enterprise with an expectation of profit derived from the efforts of others. A binary event contract on a corporate merger arguably fails parts of that test. You are not investing in a common enterprise. You are not relying on the efforts of a promoter. You are taking a position against a counterparty on an outcome. Howey, for all its reach, does not obviously capture a wager.

This is the technical heart of the dispute, and it is why the answer will not come from regulators alone. The event contract lives in the space between two definitions that were never designed to touch: the wagering contract and the security-based swap. Citadel's argument is that proximity to a single issuer pulls the product into securities law. The prediction markets' implicit argument is that the absence of a common enterprise keeps it out. Both are correct about their own premises. Neither premise resolves the case.

And this is precisely the environment in which I have learned to stop asking what the law says and start asking who bears the cost of being wrong — a habit I picked up the hard way.

What the Alameda Ledger Taught Me About Classification

In 2022, I spent the worst month of my professional life reconstructing the hidden leverage inside Alameda Research's balance sheet. I used cross-collateralization ratios pulled from on-chain data — a mathematical archaeology that required me to assume the worst about every footnote and then look for the places where the footnotes were themselves the lie. I found a discrepancy of roughly $1.2 billion in unallocated stablecoin reserves. The number mattered less than the method. What I learned was that systemic failure almost never announces itself as a violation. It announces itself as a classification error — an asset labeled as liquid that was not, a liability labeled as contingent that was not.

The FTX collapse was, at bottom, a taxonomy failure that everyone agreed to ignore because the taxonomy was profitable. The label said customer funds. The substance said something else. And the gap between the label and the substance is where the losses lived.

The event contract fight is the same shape. The label says event contract. The substance, for single-name products, may say security-based swap. And if that reclassification ever happens — if a court or the SEC declares that a company-linked event contract is a securities instrument — then every contract in the existing inventory becomes a classification error with a legal price attached to it. The positions are already open. The collateral is already posted. The clearing has already happened. Reclassification does not unwind the past. It only makes the past illegal.

That asymmetry is what Citadel is really buying. Not a ban. A definition. Because whoever writes the definition writes the rules for who is allowed to make markets in the aftermath.

Let me be clear about motive, because the coverage has been credulous. Citadel Securities is the largest market maker in American equities. It is also a designated market maker on major venues. If single-name event contracts are securities, Citadel cannot simply ignore them — it would either have to register for the appropriate securities activities and comply with the full apparatus of securities law, or watch a competitor build the market without it. The defensive reading is the correct one: a reclassification that pulled these contracts into SEC territory would lock out the venues that self-certified their way in, and hand the franchise to whoever is already equipped to operate under securities rules. That is not a conspiracy. It is market structure logic, and it is rational. But we should name it accurately. This is a land grab dressed as a jurisdictional clarification — and that does not make it wrong. It makes it strategic.

The Death of Deference and the New Arithmetic of Arguments

The reason this fight is happening now, rather than five years ago, is a single Supreme Court decision. In 2024, in Loper Bright Enterprises v. Raimondo, the Court overruled the Chevron doctrine — the forty-year-old principle that courts should defer to an agency's reasonable interpretation of an ambiguous statute it administers.

The practical effect is seismic and underappreciated. Under Chevron, when a statute was ambiguous — and "contrary to the public interest" is nothing if not ambiguous — the agency's reading usually won. Courts deferred. That gave the CFTC enormous latitude to expand its jurisdiction over event contracts through interpretation rather than legislation. It allowed the agency to treat its vague public-interest standard as a governing rule, because no court wanted to second-guess the regulator's reasonable reading.

After Loper Bright, that deference is gone. Courts will interpret the statute themselves. And when a court reads the text of Section 5c(c)(5)(C) without a thumb on the scale, it will find a standard that is broad, discretionary, and — critically — structurally weak as a basis for expansive jurisdiction. Meanwhile, the SEC's textual basis in Section 3(a)(68) is narrow, specific, and doctrinally durable: payment tied to a single issuer's financial condition. Text, not deference, now decides the winner. And text favors the party holding the more specific definition.

The consequence is that the CFTC's historical method of expanding control — bend the public-interest standard, add a special rule, self-certify, repeat — has lost the legal scaffolding that held it up. For a decade, the agency's institutional habit substituted for its statutory precision. Loper Bright removed the substitute. What remains is the statute. And the statute does not, on a plain reading, give the CFTC exclusive control over a contract that is effectively a bet on one company's balance sheet.

There is a precedent that matters here, and I want to handle it carefully, because it is often misread. In Kalshi v. CFTC, a federal district court held that the CFTC exceeded its authority when it tried to block Kalshi's congressional-control contracts, reasoning that the agency's action fell outside its statutory power. Read the holding precisely. The court did not say the SEC has no jurisdiction. It did not say event contracts are commodities. It said the CFTC cannot arbitrarily forbid them. That is a shield, not a grant of territory. Kalshi secured the freedom to list; it did not secure the freedom from securities law.

Citadel's move is to exploit exactly that gap. It is using the very precedent that constrained the CFTC to argue that the SEC — not the CFTC — is the correct authority for single-name contracts, which would simultaneously validate the Kalshi holding and neutralize the CFTC's expansionist reading. It is a clever piece of legal jujitsu, and it tells you how sophisticated the parties have become about the new post-deference landscape.

The Economics of a Fault Line

Now let me do what I actually do, which is to ignore the lawyers for a moment and follow the money. Because the classification question is downstream of an economic fact: the products at issue are the ones that generate the most volume, the most spread, and the most information content.

Macro event contracts — elections, rate decisions, inflation prints — are interesting but thin. They attract episodic volume around scheduled catalysts and then go quiet. Sports contracts are deep and liquid but capped, socially, by gambling law. Climate contracts are nascent. The product that changes the economics of the entire category is the single-name contract: the corporate event. It is continuous. It is information-rich. It draws hedgers, speculators, and anyone who wants exposure to a binary corporate outcome without touching the equity or the options chain.

That is why the jurisdictional fight is really a fight over the most valuable slice. A regulator who claims the single-name product claims the category's growth engine. A market maker who wins the definition wins the franchise. And a venue that loses the classification loses its best product line overnight — not through enforcement, but through illegality.

The clearing dimension is the part almost everyone is underestimating. If a company-linked event contract is reclassified as a security-based swap, then the clearing house that stands behind it may find its own registration status and compliance posture challenged. Clearing houses are the load-bearing walls of the derivatives market. When the load-bearing wall's legal condition becomes uncertain, the cost of capital across everything it touches rises — quietly, at first, then all at once. I have watched a clearing-adjacent failure before, and I can tell you the sequence: first the legal question, then the collateral haircut, then the run.

And there is an infrastructure angle that ties directly into the work I have been doing on settlement. In 2025, I built a liquidity model around BlackRock's BUIDL fund integrated with Ethereum Layer 2s, and I quantified something that surprised even me: tokenized real-world assets compressed traditional settlement times by roughly 94 percent while preserving regulatory compliance. That result taught me that the bottleneck in modern markets is almost never the technology. It is the permission. The settlement tech works. What does not work is the entitlement — the legal right to hold, transfer, and net an instrument across jurisdictional lines. Event contracts are running headfirst into exactly that wall. The rails are ready. The permission is contested. And a market built on rails without permission is a market that can be shut off with a single letter.

I will add a related observation from the cost side, because it shapes who survives. The proving and settlement overhead for sophisticated on-chain derivatives is currently absurd — Layer 2 proving costs remain high enough that, outside bull-market gas conditions, operators bleed. Any venue that tries to build single-name event contract clearing natively on-chain, with securities-grade monitoring, will discover that the compliance cost dwarfs the computational cost. The real expense is not ZK proofs. It is the surveillance system required to detect insider trading on corporate events — and that is a human-intensive, not a hardware-intensive, problem.

Which brings me to the deepest structural issue of all.

The Oracle Is the Market

Every event contract requires a resolution source. For a macro contract, that is a government statistics release. For a single-name corporate event, that is something far more dangerous: a licensed market data feed, a corporate actions database, a settlement determination about what actually happened — did the merger close, did the executive resign, did the regulatory action constitute a material event.

The moment you reference a single issuer's price or condition, you have imported not just securities law but securities infrastructure: reference data licenses, corporate action feeds, price validation, manipulation surveillance with an equity-market analog. And this — this is where the RWA narrative and the event-contract narrative collide, because they are the same story told from opposite ends.

Here is my long-held position, and I will state it as a fact of institutional behavior rather than an opinion: the tokenized-RWA story has been a three-year exercise in storytelling, and the part no one wants to concede is that traditional institutions never needed a public chain to do it. BlackRock did not need Ethereum to move money. It needed a settlement layer it could audit, freeze, and recover — and a public chain, for all its elegance, is a liability before it is an asset when you are a fiduciary. The institutions will take the function of tokenization and leave the ideology behind. They always do.

The event contract market is about to learn the same lesson. If single-name contracts are securities, the resolution sources will not be permissionless oracles. They will be licensed feeds, controlled by the same data monopolies that dominate equity reference data. The "decentralized prediction market" will either retreat entirely into macro events — sovereign, not corporate — or it will quietly become a licensed securities venue wearing a crypto costume. The costume will fall off. It always falls off.

That is the decoupling, and it is where I part company with the consensus narrative.

The Decoupling Thesis: Not Convergence, but Partition

Everyone I talk to in this business believes the endgame is convergence. Prediction markets and traditional finance merge. The CFTC and SEC harmonize. The crypto-native venues and the Wall Street incumbents meet in a regulated middle and the category grows up. It is a tidy story. It is wrong.

What the Citadel letter sets in motion is not convergence. It is partition. And partition is the more likely outcome precisely because the legal frameworks cannot be merged without an act of Congress that nobody currently has the votes or the appetite to pass.

Follow the logic. The macro event contract — elections, rate decisions, inflation, geopolitics — sits comfortably in CFTC territory. It has no single issuer. It has no corporate insider with material non-public information. It is a genuine, legally clean, socially tolerable wager on the world. The CFTC keeps it. Volume grows. The venue model holds.

The single-name event contract — corporate mergers, executive departures, earnings outcomes, regulatory rulings on a specific company — will, I believe, migrate to SEC-registered venues or simply to products that are economically identical but legally dressed as securities: structured notes, binary options, exchange-traded conditional instruments. Same payoff, different wrapper. The wrapper is the whole point. The wrapper is what determines who loses when something goes wrong.

What happens when you partition a market this way is not harmony. It is a widening basis. You will have two products referencing the same underlying corporate event, trading in two separate legal venues, priced by two separate pools of capital, with two separate margin regimes — and a spread between them that reflects nothing but the jurisdictional wall. Arbitrageurs will live in that spread. Sophisticated players will exploit it. Retail will trade the wrong side of it without knowing it exists. This is not a prediction born of cynicism. It is what always happens when you slice a market along a legal seam rather than an economic one. The seam becomes a frontier. The frontier becomes a fee.

And here is the blind spot that no one in the debate is naming: the partition thesis assumes there is a coherent thing on each side to partition. But there isn't. The corporate information that drives a single-name event contract is continuous and cross-market. The equity price contains it. The options surface contains it. The credit spread contains it. The event contract would contain it too — which is precisely why it looks like a security and precisely why the SEC's claim is structurally coherent rather than opportunistic.

We are auditing the ghost in the machine's soul. The machine is the market. The ghost is the informational content that refuses to stay in one regulatory box.

And underneath all of it is the most uncomfortable structural fact of the decade, which is that the market is increasingly not a market of humans making discrete decisions. In 2026, I analyzed a dataset of ten million transactions executed between autonomous AI agents on blockchain networks. Sixty percent of them occurred with no human intervention at any step. Sixty percent. The counterparties to an event contract will not always be discretionary traders with an opinion about a merger. They will increasingly be agents executing a probability model at a speed no human can monitor and a granularity no human can review.

So the real question the Citadel letter forces — the question buried under the jurisdictional headlines — is this: when the market no longer has a human on the other side, who is the investor that event-contract law is supposed to protect? The machine economy is arriving in exactly the sector where the investor-protection framework is least able to describe the participants. The ledger bleeds red when trust decays into code — and here, the code is the counterparty.

That realization forced me to confront something I had believed for years: that blockchain was, at its core, a technology for human financial liberation. I no longer think that is the primary use case. The primary use case is machine-to-machine settlement, and the human is increasingly a beneficiary and an observer, not the protagonist. It is not a comfortable conclusion for someone who came to this field looking for meaning. But it is the honest one.

What the Smart Money Is Actually Pricing

Let me return to the tape, because this is a sideways market and sideways markets are won by positioning, not by prediction. In consolidation, the winners are the people who understand structure before the structure is validated by price. On-chain, the majors are quiet. Funding is flat. The rotational capital that drove the last cycle has parked in yield and is waiting. Volatility is compressed. That is precisely the kind of tape where jurisdictional and structural shifts are priced slowly and then repriced violently.

What is the market actually pricing right now? It is pricing the probability that nothing happens. The event-contract category trades as if the CFTC's permissive posture is permanent. It is not permanent. The moment the SEC formalizes a position — and I expect it will do so within eighteen months, most likely through a concept release or a rule proposal rather than an enforcement action — the entire legal scaffolding under single-name contracts gets re-rated in a single week. And here is the asymmetry the market is mispricing: a rule proposal is cheaper for the SEC than enforcement, and more devastating for existing contracts, because a proposal can freeze new listings while the inventory problem festers. Enforcement punishes one venue. A rule punishes the category.

I have said before that liquidity is tightening and I have said it in the wrong cycle. I will not repeat that error. Liquidity is not tightening. Liquidity is reclassifying. It is moving from categories with legal clarity to categories with legal shelter, and it is doing so quietly, before the headlines. The single-name event contract has neither clarity nor shelter. That is why it is at risk. That is why Citadel moved first.

The operators who will bleed are the ones who built their product lines on the assumption that self-certification was a permanent right rather than a procedural default. Because defaults can be changed by rulemaking, and rights must be defended in court. Most venues have been defending nothing. They have been filing paperwork and opening doors, and the doors have stayed open long enough that they mistook the open door for a wall.

When the door closes — and it will close, not slam — the venues that survive will be the ones that already built the securities-grade monitoring, the reference-data licensing, the corporate-actions adjudication, and the insider-trading surveillance before they were required to. That is the window. It is open for six to eighteen months. After that, the compliance moat is finished, and the market is concentrated among the players big enough to have paid for it in advance.

Which is, of course, exactly what Citadel Securities wants. Not because the firm is villainous. Because the firm is early, and early is the entire game.

Takeaway

Three years of CBDC research taught me to read monetary infrastructure not as a technical system but as a political one — and to see that sovereignty, in the end, is the authority to name the thing. The ECB's digital euro prototype contained an offline transaction cap of three hundred euros, and I spent a month reading fifty thousand lines of code to understand that the cap was not a technical limitation. It was a governance decision expressed as an engineering constant. Sovereignty had been written into the arithmetic.

The Citadel letter is the same kind of artifact. It is a governance decision wearing a legal brief. It says: the power to name whether a corporate wager is a security is the power to decide who is allowed to run the largest market of the next cycle. Whoever holds that pen holds the franchise.

My work on algorithmic monetary policy — the report I published as The Sovereign Algorithm — projected that by 2030, roughly forty percent of global GDP would be governed by algorithmic monetary policies embedded inside central bank infrastructure. I still believe that projection. But I now understand its corollary with far greater clarity: the same algorithmic governance logic that will run monetary policy will also decide, by rule and by default, which financial products are permitted to exist. The taxonomy is the policy. The classification is the constitution.

So the forward-looking question is not whether the SEC or the CFTC wins. It is whether the market has begun to price the possibility that the answer to what an event contract is will be decided by someone other than the two agencies currently arguing about it — by Congress, by a court, or by a technological shift that makes the whole question obsolete before it is answered.

That is the position I am taking into the next cycle. Not a side. A posture: assume the definition will change, position in the assets that survive the change, and treat the ambiguity itself as the trade. Because in a sideways market, the only thing more dangerous than being wrong is being right too early about a taxonomy no one has finished writing.