John O. McGinnis
Prediction markets are platforms where people can bet on future events, such as election results, the fate of legislation, or even economic outcomes conditional on the passage or defeat of that legislation. The idea behind prediction markets reflects the core insight of Friedrich Hayek: given the right incentives, the minds of many can aggregate the information in the world and make better assessments than any individual. Information is dispersed, and the market mechanism provides the best way of focusing its accumulation in the transparent form of a price. Prediction markets thus expand market order beyond commercial matters to the social sphere.
Prediction markets are instructive, time-saving, and even entertaining. With the advent of Predictit, a non-profit market run out of New Zealand, I now avoid the endless chatter of confident pundits claiming to know what will happen in the mid-terms as well the daily news article about the state of various political races. Instead, I regularly consult the prediction markets and watch a measured ebb and flow of politics as one party or candidate profits from events and another falters. The markets also provide me with a sense of what is driving politics. For instance, while I believe the Supreme Court’s decision in Dobbs was entirely correct, it is also clear from the markets that it had an adverse effect on the likely fortunes of the Republican Party in November. Immediately after the decision, the likelihood of Republicans controlling the House of Representatives as of 2023 went down sharply, although they are still substantially favored to win. It is useful to be graphically reminded that getting the law right and being politically popular have little to do with one another.
Prediction Markets and Planning
Prediction markets serve far more important purposes than my own pleasure. First, they promote liberty and the rule of law in a democracy where the government can change that law. One of the central purposes of the rule of law is to allow citizens to plan for the future. If government were not bound by rules, but instead made decisions at the whim of those in power, planning would be much more difficult. Not only would economic growth suffer but so would an essential aspect of liberty. Only the capacity for future planning can protect our agency in the world.
Our Constitution protects the rule of law and planning by requiring legislation to authorize any government action. But Cong-ress can change the rules (and now so can administrative agencies under bro-ad congressional delegations), and the results of e-lections are the most important indicators of how those rules will be revised. Thus, citizens need to have as good a sense of the future of politics as possible if they are to plan adequately. That understanding is good for the economy because people can make more rational choices and hedge against legislation that may adversely affect them. For instance, if a company has a more precise assessment of how likely a party with a regulatory agenda inimical to its flourishing is going to come to power, it can better figure out what steps to take in mitigation. Indeed, they could buy shares in the party’s potential victory to pay for some of the mitigation!
Prediction markets also promote liberty more generally as they allow us to be more rationally prepared for various eventualities. Liberty requires protecting our sense of agency through time. We have greater agency the more we know about the future. At the extreme, if the future were completely unpredictable, we would feel helpless. When we have a better sense of the obstacles and opportunities, we become more confident in our agency, like a skier who can see slalom gates on the course he must traverse.
By providing the most accurate forecast of who will be in charge of Congress, a prediction market gives a better sense of the future gates we are likely to have to navigate—that is, the policies we will face in the coming two years. Prediction markets have even emerged that gauge the likelihood of passage of legislation setting important policies such as tax rates. If these markets were open to large sums of money (they are not currently), individual citizens could hedge against the future, just as farmers now do against the weather, making their fortunes less dependent on vicissitudes of politics which they control no more than farmers do the skies.
Prediction markets can also improve the results of politics. Politicians continually claim that their programs will have all kinds of good effects. Their opponents claim the opposite. President Biden has argued that his recent plan to spend hundreds of billions of dollars in subsidies for climate-friendly products and to permit Medicare to negotiate on drug prices will reduce inflation. It would have been useful before the passage of this legislation to have had a market predicting inflation rates conditional on its passage and conditional on its defeat. That market would have allowed the most informed about the actual effects of the legislation to inform the public with their bets.
Yet another advantage of prediction markets is that, because they are the results of many minds disciplined by their interest in winning money, they have more credibility than partisans of either side. If they became an accepted part of the social landscape, they would cut through some of the polarization that has undermined good policy.
The CFTC’s Road-blocks to Prediction Markets
Unfortunately, while prediction markets could help make government pursue sounder policies and better reconcile its power to change the law with the values that undergird the rule of law and liberty, the Commodities Future Tra-ding Commission (CFTC) has so far frustrated their growth. It has misinterpreted a congressional statute that prohibits any transaction that “relates to, or references terrorism, assassination, war, gaming, or an activity that is unlawful under any State or Federal law.”
First, the CFTC has declared that any contract is a prohibited “gaming” contract if “the predominant use of the contract is speculative as opposed to a hedging or an economic use.” But, as discussed above, prediction markets about politics have public benefits that go beyond their economic use in promoting the ability of citizens to plan. Even under the CFTC’s own definition, however, since the outcomes of politics inevitably affect the economy, political predictions should be entirely outside the ambit of the prohibition. For the same reason, they meet any requirement that a market be in the “public interest.”
Yet the CFCT has prevented robust competition and innovation from developing in prediction markets. Instead, it has allowed only a few companies to operate intermittently und-er no-action letters. No-ac-tion letters state that the ag-ency will not prosecute a specific prediction market, but they offer no assurance that such markets are generally legal.
The no-action letters also specify strict conditions that limit the amount of money that any participant can bet. But that limitation makes the markets less robust and accurate. Apparently, the CFTC is concerned that prediction markets may undermine elections if someone puts a lot of money on a candidate, skewing the odds in his favor and influencing others. But if someone spends wildly in the hopes of bolstering a candidate then he is inviting someone else to put money on the other side at favorable odds. As Robin Hanson has observed, making wrong predictions makes a market participant a sheep to be devoured by wolves.
And, worst of all, the CFTC can simply revoke its non-action letter without giving any reasons, as it did in August with Predictit, telling it under the penalty of prosecution to wind up its affairs by this coming February. For prediction markets to develop fully, they cannot live under such a sword of Damocles. The CFTC’s regime prevents innovators from planning a future course for markets that help citizens plan their own actions. The CFTC actions ironically demonstrate why we need prediction markets!
Predictit has sued the CFTC demanding some reasoned explanation for the revocation of its no-action letter. Unfortunately, it is not at all clear that they have a good case under administrative law, which generally allows agencies to exercise their prosecutorial discretion without providing reasons. And Predictit, being a relatively small non-profit, likely does not have the resources to put up a full fight against the administrative Goliath that is the CFTC.
The greatest hope for the continued presence of prediction markets currently rests with Kalshi, a for-profit exchange that has made prediction markets in such things as who will win at the Grammys. It has a petition to create a political prediction market for congressional control this Nov-ember, which would allow participants to invest far more money in positions than previous markets.
But even if Kalshi gains a no-action letter, the CFTC will still be impeding the public interest in prediction markets. There is a vast number of political markets from which the public would benefit—markets in whether pending legislation in taxes, antitrust, and labor law will be enacted, as well as conditional markets in what various economic metrics will be depending on whether such legislation is passed. No one will invest heavily to build the infrastructure needed for such a wide range of markets without a firmer legal foundation than a no-action letter provides.
A more permissive attitude toward prediction markets would let those who cl-aim to know the most about the likelihood of future ev-ents put their money where their mouth is. In doing so, they would advance both liberty and democracy for the rest of us.