'Maximal' ban on insider trading would hurt prediction markets, says researcher
Research from Stevens Institute of Technology argues prediction markets should use calibrated insider-trading enforcement rather than a total ban. A formal model shows accuracy is “hump-shaped” with enforcement: weak policing lets insiders crowd out other traders, while overly strict rules remove useful private information. The optimal level is an interior balance that preserves participation and informational value. Enforcement should vary by information source: independently researched edges should face little or no punishment, misappropriated or leaked information should face stronger action, and trades by people who can influence the outcome should face the toughest enforcement because of manipulation risk. This view aligns with recent market changes, including Kalshi requiring employment disclosures in sensitive markets and assigning higher risk scores to markets prone to insider trading.
