Comment Text:
Dear Commissioners of the CFTC,
My name is Jason Crawford, and I am the founder of The Roots of Progress, a nonprofit organization dedicated to researching and educating the public on the history and philosophy of progress—scientific, technological, and economic. I am also a visiting scholar at the University of Texas at Austin. I am deeply involved in creating the Progress Studies community, which is a group of people dedicated to promoting scientific research, technological innovation, and economic development. I am writing this letter to support Kalshi’s submission to the CFTC regarding election prediction markets.
One of the core values of prediction markets is their ability to aggregate the wisdom of the crowd. We know from research from Professor Philip Tetlock that the predictive track record of pundits and topic-area experts can be wanting. Instead, a mechanism of group forecasts, preferably from individuals operating with an unalloyed profit incentive to be right above all else, will consistently outperform their better-credentialed peers. It is true that there is a panoply of data that seeks to forecast elections, but none are better at collating that information into a single point estimate than a prediction market. For instance, polling data is notoriously unreliable more than a few months before the election, and persistent issues with non-response bias haved rendered them imperfect in several recent cycles (though they still remain valuable in the aggregate). A community of profit-motivated forecasters thus has the best hope of agglomerating all the relevant data—fundamentals, polls, local-specific factors—into the best guess for the probability of one party winning the election or another.
Why is it so important? Why are so many people—including so many of those who have submitted comments to the CFTC prior to myself—so enthusiastic about the value of forecasting? Because accurate forecasts are essential to better decision-making. When a business or a non-profit plans a new program, they do so by making implicit assumptions about what government policy will look like in the future. If a nonprofit relies on a government grant for their wellbeing, it would be useful to have a measure that could help them know whether cuts (or expansion) of the grant are likely in the near future. Estimating that probability of cuts is often directly downstream of knowing which party will soon be in control. Those advising the government also could benefit from knowing whether a party is likely to lose, which may warrant encouragement to pivot to the middle or myriad other countermeasures. In short, having a more accurate picture of what the future may hold is essential to making the most informed decision one can.
One theme of some prior comments is a fear that these contracts may compromise the integrity of the election. However, the Commission should not let unwarranted fears override a more sober-minded analysis of the public interest and the law. These contracts have existed for decades in many other countries, as well as in the US historically, and there is no documented evidence that they have resulted in an increase in election manipulation, tampering or misinformation.[1] In addition, the claim that giving people a financial stake in an election will prompt them to attempt to game the election fundamentally misunderstands the status quo: people already have a massive financial stake in elections. Whether a party wins or loses could mean thousands to millions of dollars for families and businesses, through changes in taxes, subsidies or other policies. These markets do not introduce any additional financial stake that may incentivize manipulation (if anything, by giving people a vehicle to hedge their exposure, it may even decrease that incentive). Many of the other concerns, such as the risk that some people may flip their votes, hardly seem relevant to the outcome of a national race, let alone to a contract about the combined outcome of several hundred races.
In conclusion, the CFTC can best uphold the public interest by approving this contract.
Thank you.
[1] Paul Rhode and Coleman Strumpf. 2003. “Historical Prediction Markets: Wagering on Presidential Elections”.
