Comment Text:
My name is James Cust, and I am a Senior Economist in the Office of the Chief Economist for the Africa Region at the World Bank. My research focuses on how to achieve sustainable development and the economics of natural resource wealth. I have traded on various prediction markets for years, when I lived in the UK, and now more recently on Kalshi. I am writing to support Kalshi’s bid for a legal, well-regulated prediction market on election outcomes.
The beauty of prediction markets is their ability to provide a valuable market-driven data point. Talk is cheap–people, including pundits and experts, can assert that “x event will happen” or “z candidate will win” with impunity. In many cases, without money on the line, predictions become merely a form of preference expression. In other cases, other incentives (e.g. a career or social desirability incentive to say one outcome is likely to happen) dominates the incentive to actually be correct. For example, someone may say “I think X party will win” as a way to gin up controversy and draw views to their show. As a result, the world of news media, consultants’ reports and “expert” models often can be wildly inaccurate. Nowhere is this more true than in politics. Anchors hype up polls that drive engagement (such as an outlier poll showing an underdog up in the race), and more partisan-aligned commentators have strong incentives to emphasize all races as close, but winnable. As a result, the election discourse is replete with misinformation and its lesser cousin, inaccurate or lacking-context information.
Enter political prediction markets. For those who are not using the contract to hedge their election risk, the only incentive is to be right. In other words, political prediction markets meet what economists call incentive compatibility, which means that the optimal behavior of participants is to act according to their true beliefs. If the market is trading at 25c for an outcome you believe has only a 20% chance of occurring, you will bid the price down. And the incentive to figure out the slight edge–between 20 and 25–drives one to consume large amounts of news, analyze reports, crunch numbers and update constantly. Thus unlike pundits or experts who steadfastly stand by their previous predictions even as new information has rendered it outdated, prediction market participants have every incentive to update their beliefs as new information emerges, in order to be right. As a result, the market-based estimate of the state of the race is likely to be more accurate than what exists in the news media.
Importantly, markets seek to see through all the misinformation or context-poor information out there. They focus on the signal, not the noise, because they have a strong incentive to do so. As a regular citizen, it is a lot of effort to listen to a pundit, then search up their track record to see if their history of predictive accuracy, then weigh their credibility against pundits saying the opposite. But market participants are more than willing to do so, since they have skin in the game. This is the means by which the market creates more accurate information than the traditional news media.
The CFTC appears worried that this constitutes “gaming”. This criticism is false. While it is true that many people enjoy predicting, there are many more who use the market in part for hedging, or price-basing purposes. The value of the “predictors” then is that they provide liquidity and competitive pricing to those who use the market to hedge and manage their risks. In addition, even when people are not even trading on the market, the data-point can be used for pricing other services or financial assets. The number of people who can cite or quote the probability for their business purposes (especially if it gets picked up by a newspaper or television channel) can be vast. It would be inaccurate then to only look at the share of participants in the market using it for hedging as the only people using the product for non-speculative purposes as it would exclude the large number of people who benefit from accurate reporting. As an economist who specializes in natural resource economics and development, I know how politics is inextricably linked with the price of natural resources. Government policies related to taxation, leasing rights, fuel subsidies, and environmental regulation all have real and sometimes immediate effects on the supply and demand for energy commodities, and thus have a clear and obvious impact on its price. As a consequence, the price point provided by the market can help participants in other markets better price commodities and financial assets.
Prediction markets incentivize informed participation as opposed to those who treat participation as a game of chance. And as such, they produce useful, actionable, and over time, reliable information that can approximate a likely outcome better than existing tools.
These markets therefore have the potential to bring large, positive social externalities, and should be permitted by the US government. It would be hugely disappointing if the CFTC were to reject this valuable tool for researchers, businesses, and the broader public.
