How manipulatable is too manipulatable?

Only two commodities are banned by US law from futures markets: onions and motion picture box office receipts.

In both cases, lawmakers concluded that the underlying markets were too easy to manipulate.

They didn’t have to imagine how the onion market might be. In 1955, Vincent Kosuga — a New York onion farmer who also owned a seat on the Chicago Mercantile Exchange — bought nearly every onion available for delivery in Chicago. At the peak, he held 30 million pounds of onions in storage.

With the market cornered, Kosuga gathered 13 of the country’s largest farmers and distributors of onions and threatened to crash the price of their livelihood if they didn’t buy futures from him at the artificially inflated prices he had engineered.

They agreed to buy futures representing a small fraction of Kosuga’s onions — $168,000 worth — in exchange for a promise that he wouldn’t sell any more. He made the promise, sold them the futures, and then broke the promise.

Kosuga and co-conspirator Sam Siegel shorted as many onion futures as they could, as fast as they could, and then started selling their warehoused onions, as well. This completely broke the market. From an inflated peak of four dollars, the price of a 50-pound bag of onions fell to just ten cents.

The burlap bags the onions came in were worth double that. So the onions themselves were effectively valued at negative ten cents — slightly less than the cost of disposing of them, presumably.

Kosuga and Siegel were left with millions of pounds of rotting onions, but they had shorted enough futures that their corner reportedly turned a profit of $8.5 million ($106 million in today’s money).

Onion farmers lost much more than that — many lost their entire farm.

Kosuga was unrepentant. “If it’s against the law to make money,” he reportedly said in the congressional investigation that followed, “then I’m guilty.”

He wasn’t. The only consequence Kosuga faced was losing his seat on the CME. He retired a wealthy man.

But Congress agreed with his victims that speculators should not be allowed to ruin farmers. In 1958, they passed the Onion Futures Act, making it illegal to buy or sell onions for delivery at a future date.

It was too late to help devastated onion farmers, but an important precedent had been set: commodities that are easy to manipulate should not trade as futures.

In 2010, the law was amended to include box office receipts — the idea being that speculators might short a movie’s futures and then undermine its opening weekend.

Those remain the only two banned by law. Everything else is left to the discretion of the CFTC.

Enter prediction markets

Commodity regulators have long claimed jurisdiction over event contracts because they closely mirror traditional futures: both are contracts that derive their value from the outcome of a future event.

The CFTC therefore expects prediction markets like Kalshi and Polymarket to comply with Core Principle 3 of the Commodities Exchange Act: “The board of trade shall list on the contract market only contracts not readily susceptible to manipulation.”

I’m not sure how seriously the industry has taken that guidance, because many event contracts seem easy enough to manipulate.

Consider, for example, that Brian Armstrong only had to read out the list of what words he might say in a conference call to manipulate the outcome of the event contract “What words will Brian Armstrong say in his conference call?”

That may not be a transgression of the Core Principle 3  — it’s hard to say because the CFTC has not yet defined what “readily susceptible” means in the context of event contracts.

They will do soon, though. The agency is currently working on guidance specific to prediction markets that will establish what kind of events — if any — are too susceptible to manipulation for futures trading to be allowed. 

Prediction market-researcher Maksym Nechepurenko believes they should start by determining how much it might cost to manipulate each category of market.

He estimates, for example, it might cost about $100,000 to manipulate the outcome of a moderate-profile sporting event.

Another category of cheaply manipulated markets, he says, is “sub-national political events” — $50,000 might buy enough votes to swing a local election, for example.

Cheapest of all are “information release outcome” events, like company earnings reports or government statistics. These can be manipulated for free — you only need to be the person in charge of releasing the information.

All of these seem far easier than buying 30 million pounds of onions. But Nechepurenko doesn’t think they need to be banned. Instead, he recommends that prediction markets limit the scope for manipulation by implementing limits to leverage, position-size, and concentration based on how easy an event might be to manipulate.

As it happens, these are exactly the measures the CFTC says it’s considering as part of its new guidance.

Staying off the short list

In 1955, onions were not the most important commodity in the country (switching from onion soup to tomato soup wasn’t much of a hardship). 

But they were the most important futures contract — somehow, onion futures accounted for 20% of the CME’s volume at the time.

The CME therefore sued to overturn the Onion Futures Act, arguing that it unfairly restricted trade.

They lost. The courts determined that Congress has the ultimate say in which futures can and cannot trade. 

Thus far, they’ve only exercised that right in onions and motion picture box office receipts.

But that could change. Several states are suing prediction markets to stop them from offering sports-related markets. The military worries that prediction market betting could warn adversaries of surprise attacks. The DoJ is prosecuting insider trading incentivized by an event contract on Google searches.

Whatever the CFTC decides, prediction markets will have to self-regulate — and do some lobbying — to ensure they’re not added to the short list of federally banned futures.

— Byron Gilliam

_{Brought to you by:}

Robots and machines already outnumber humans. They will soon outnumber us onchain, too.

peaqOS turns machines into autonomous participants in the economy — a new, liquid asset class, with yield tied to real-world outcomes. peaq gives machines everything they need to do business on any chain and lets humans earn from automation.

Billions of machines. One layer.