After the Binary Ban: The FCA, Prediction Markets, and the Regulation of Truth

In April 2019, the UK Financial Conduct Authority did something that felt almost painfully responsible. It banned the sale of binary options to retail consumers. The move was backed by the regulator's own arithmetic: by its estimate, roughly three out of every four retail clients who touched these products lost money. Binary options had become a digital slot machine dressed in financial terminology, and the FCA, to its credit, pulled the plug. I remember the period well. I was not yet building my education platform, but I was already deep in the philosophical trenches of decentralised finance. The ban felt clean, decisive, and easy to defend. What I did not fully appreciate at the time was that the FCA had not simply banned a product. It had drawn a line across an entire category of human behaviour: the desire to put money behind an opinion about the future. That behaviour does not retire because a regulator tells it to. It migrates. It renames itself. And sometimes, it grows into something that regulators can no longer ignore. Now, five years later, reports indicate that the FCA is reconsidering that position. The regulator is said to be weighing whether to lift or revise the ban as it applies to prediction markets. Noise fades. Value remains. The question is whether London can build a framework that tells the difference between a gambling product and a truth-finding instrument. The first thing worth understanding is what the 2019 ban actually targeted. Binary options are contracts that pay a fixed amount if a particular condition is true, and nothing if it is false. You bet yes or no on an outcome, the contract expires, and the world renders its verdict. In principle, that is almost identical to buying a share in a prediction market. Buy a share that pays if a candidate wins an election, hold it to expiry, and you have effectively traded a binary option. The payoff profile is the same. The structure is the same. What differs is the cultural context, the continuous price discovery, and most importantly, the intent of the people building the market. The FCA's original intervention was a product intervention under the Financial Services and Markets Act. It followed years of complaints about aggressive marketing, unlicensed brokers, and platforms that profited from churn rather than accuracy. The regulator was right to act on those abuses. But the ban also had an unintended side effect: it made the UK a hostile jurisdiction for one of the most interesting experiments in collective intelligence since the invention of the opinion poll. What changed between 2019 and today is not the technology. Prediction markets have existed in various forms for centuries. What changed is the scale, the reliability, and the mainstream visibility of the infrastructure. In the last few years, prediction market platforms have processed tens of billions of dollars in volume. Election nights became live demonstrations of probability aggregation. Suddenly, these markets were not just a crypto curiosity. They were a source of real-time information that newsrooms, hedge funds, and political campaigns were watching. That shift forces a regulatory reckoning. When a market becomes systemically informative, it stops being purely a retail entertainment product. It becomes infrastructure. And infrastructure invites oversight, not simply prohibition. The core question for the FCA is deceptively simple: what distinguishes a harmful binary option from a legitimate prediction market? If the answer is nothing, then the ban should stand. If the answer is something, that something must be definable in rules, enforceable in practice, and implementable in code. Let us start with the uncomfortable technical truth. At the level of smart contract logic, there is almost no difference between the two instruments. A prediction market share is a token that resolves to a payout based on an outcome. A binary option is a contract that does precisely the same. The distinction lives in the surrounding mechanism and the market's purpose. A well-designed prediction market is not trying to maximise turnover on short-dated noise. It is trying to produce a continuous, honest estimate of the probability that an event will occur. The payout is a secondary effect. The primary effect is calibration. That is why prediction markets have historically been studied as tools for forecasting, not as instruments of speculation. This is also where the philosophical difficulty begins. If the FCA classifies prediction market tokens as financial instruments, then platforms will need to comply with a dense layer of conduct rules designed for securities. They will need orderly book building, disclosure obligations, possibly a prospectus for certain offerings, and strict marketing restrictions. That regime can accommodate a prediction market, but only if the market operator has the legal and engineering resources to carry the compliance burden. If, on the other hand, the FCA treats prediction markets as a form of regulated betting, then the jurisdiction shifts toward the Gambling Commission, and the markets inherit a different set of constraints: licensing, responsible gambling controls, and a ceiling on how large the positions can become. That approach may protect consumers, but it risks strangling the informational value of the market. If a trader cannot put enough capital behind a view to make the price meaningful, the market becomes a poll, not a pricing mechanism. There is a third path, and it is the one I find most credible. The FCA could create a sandbox or a pilot regime for prediction markets, allowing a small number of licensed platforms to operate under close supervision. Within that sandbox, the regulator could test whether the structural safeguards of contemporary prediction markets are sufficient to protect retail users. Do longer expiry horizons reduce impulsive trading? Do collateral requirements discourage reckless leverage? Does the transparency of an on-chain order book create better outcomes than the opacity of a traditional broker's internal matching engine? These are empirical questions. They deserve empirical answers. Based on my own experience auditing various DeFi protocols, and my years of studying how prediction markets behave under stress, I would argue that the mechanism is far more resilient than its reputation suggests. The fragility of a prediction market almost never lies in the market itself. It lies in the oracle. The oracle is the mechanism that determines whether an event actually occurred. In a decentralised market, that function is often performed by a network of reporters who stake tokens on the accuracy of their answer. If the oracle is corrupted or politically captured, the entire market becomes a tool for manufacturing false consensus. This is where the crypto industry needs to be honest with itself. The same properties that make prediction markets valuable, their openness and their permissionlessness, also make them vulnerable to manipulation by well-funded actors who want to move the price of opinion. A trader can lose money on a wrong prediction. But a society can lose something more important when a prediction market is used to manufacture a false sense of certainty about an event that has not yet occurred, or to legitimise a fabricated outcome. Code executes. Ethics sustain. A regulated regime, done intelligently, could actually improve this situation. If the FCA requires prediction markets to use transparent oracle mechanisms, to publish their resolution criteria in advance, and to maintain an audit trail of how outcomes are determined, that would raise the cost of manipulation without destroying the market's utility. The danger is not regulation. The danger is regulation that is so blunt it recreates the 2019 problem in a new form. There is also a quieter risk that the market itself must confront. If the FCA lifts the ban and prediction markets become regulated financial products, the economics of the space will change. Compliance costs are not trivial. Licensing fees, legal counsel, KYC obligations, transaction monitoring, and ongoing regulatory reporting will demand a serious operational budget. Small, idealistic teams may find themselves priced out of the market. Meanwhile, well-capitalised institutions with experience navigating financial regulation will be able to enter and dominate. That outcome would be ironic. The prediction market movement began as a decentralised, anti-establishment experiment. Its early advocates believed that crowds, not committees, were the best judges of future events. If the space is regulated into the arms of large financial institutions, it may become more credible but less radical. It may become a clean, sterile information market where ordinary citizens spectate while institutions trade. Silence speaks louder than pumps. And when there is no meaningful public participation, the market loses the very diversity that gives it predictive power. The contrarian view, which I hold with increasing conviction, is that legalisation may be more dangerous to the ethos of prediction markets than prohibition ever was. Prohibition kept the space pure. It attracted only people who understood the technology deeply and believed in its mission. Regulation, by contrast, will attract capital. Capital brings sophistication, but it also brings a different set of incentives. A hedge fund that uses a prediction market to hedge an election outcome does not care about the philosophy of collective intelligence. It cares about basis risk and execution quality. That is perfectly rational. But it means the market's centre of gravity will shift from truth-seeking to risk transfer. I do not say this as an argument against regulatory clarity. The industry has spent years complaining about uncertainty, and it deserves a mature, principled framework. But the UK has an opportunity to do something more interesting than simply import the standard financial services playbook. It can design a regulatory regime that recognises the dual nature of prediction markets: they are simultaneously financial instruments and information services. A regime that treats them only as securities will miss half their value. A regime that treats them only as gambling will miss the other half. The FCA has shown, in recent years, a willingness to engage constructively with innovation. It has established a sandbox, supported the growth of fintech, and published thoughtful work on cryptoassets. The fact that it is even considering a change to the 2019 position suggests that someone inside the institution has been reading the academic literature on information aggregation. I hope that whoever is leading this review also takes the time to study the failure modes of the technology. Prediction markets are not magic. They fail when resolution is ambiguous, when oracles can be captured, and when liquidity is so thin that a single large trader can distort the price. These are solvable problems, but they require careful engineering and honest governance. What would I advise if a platform asked me how to prepare for a regulatory opening in the UK? The answer begins with humility. Build the compliance infrastructure before it is required. Maintain meticulous records of how every market is constructed and resolved. Publish a clear framework for oracle selection and a clear path for dispute resolution. Do not treat KYC as an afterthought. Treat it as a design principle. A prediction market that cannot verify its participants cannot claim to be a responsible part of society. It will also be the first target of a regulator looking for an easy win. And for the regulator, my advice would be this: do not attempt to define the technology. Define the outcomes you care about. If you care about consumer protection, focus on leverage, marketing, and the opacity of resolution. If you care about market integrity, focus on oracle manipulation and insider trading in political event contracts. If you care about the public good, focus on ensuring that the informational output of these markets remains accessible to everyone, not just paying customers. The next twelve to eighteen months will be telling. If the FCA publishes a consultation paper, that will be the first real signal that this is more than regulatory noise. If a licensed prediction market emerges in the UK, it could become a template for other jurisdictions that have watched the American legal battles around Kalshi and the CFTC with a mixture of fascination and fatigue. Britain has a chance to lead by crafting a regime that is principled without being paternalistic, rigorous without being rigid. The deeper question, the one that keeps me awake at night, is not about legal classifications or oracle designs. It is about who gets to participate in the construction of civic reality. Prediction markets are, at their core, a technology for disagreeing honestly. They let people express a view with their capital, and they convert aggregate disagreement into a measurable probability. That is a remarkably healthy social function in an era of performative consensus. If the FCA bans it again, the trading will simply move elsewhere, to jurisdictions less concerned about consumer welfare. If the FCA regulates it wisely, it will have taken a stand for the proposition that truth, even when it is inconvenient, should be allowed to find a price. London built its financial reputation on the trading of risk. The next frontier is the trading of certainty. Whether that frontier becomes a genuine public market or a private club for institutions will depend on choices being made in quiet rooms right now. Regulation is not the enemy of decentralisation. Bad regulation is. And the best regulation is the kind that recognises the difference between a rigged bet and a sincere question. The markets are ready. The technology is ready. The only remaining question is whether the architecture of trust we are building is designed for the benefit of the few or for the clarity of the many. Code executes. Ethics sustain. And in the end, every framework the FCA writes will be a reflection of what it believes people are capable of. I hope it believes we are capable of handling the truth.