Skip to main content

Key Finding

Prediction markets should protect informed trading to preserve their informational value while restricting manipulation by excluding event contracts that give participants incentives to influence the outcomes being traded

Abstract

Securities markets and prediction markets are fundamentally different. Securities markets allocate capital, finance productive enterprise, and provide the infrastructure through which households save for retirement. Prediction markets do none of those things. Their principal social product is the information contained in their prices: a continuously updated estimate of the probability that a specified event will occur. While prediction markets could, in theory, be used to manage risk by allowing traders to hedge, thus far liquidity constraints have meant that traditional derivatives contracts (options and futures) rather than prediction market contracts are used.

Those differences have important regulatory implications, suggesting little or no regulation of informed trading as such on prediction markets and significant regulation of manipulation on these markets. The institutional features of securities markets indicate the opposite: strict regulatory attention to insider trading, which can undermine valuable property rights in information, but comparatively little concern for manipulation, which is rare and largely self-correcting. In securities markets, unauthorized insider trading may appropriate valuable corporate information, generate agency costs, widen spreads, and impair markets that allocate society’s capital. In prediction markets, trading based on superior information usually improves the only product the market creates. Empirical surveillance can identify anomalous trading and even informational advantage, but trading data alone cannot establish whether the advantage was obtained or used in breach of a duty. A general prohibition on trading while possessing nonpublic information would likely suppress, rather than protect, the market’s informational function.

Manipulation presents a different problem. Some event contracts motivate participants to influence the occurrence, description, or settlement of the event on which payment depends. This Article dubs these instruments corruption-prone event contracts: contracts whose payoff depends on events that one or more market participants can materially influence through their own conduct or through the conduct of persons subject to their influence.

The appropriate model is the insurable-interest doctrine. Life insurance is socially useful when the beneficiary would suffer a genuine loss from the insured’s death. A wager policy on the life of a stranger creates a different incentive, and the law responds ex ante by refusing to recognize the contract rather than relying exclusively on prosecution after the harm occurs. Prediction-market regulation should use the same institutional logic.

Existing misappropriation, agency, ethics, and classified-information rules should govern genuine breaches of confidence. Lawfully informed traders should remain free to trade. Exchanges and the Commodity Futures Trading Commission should, however, exclude corruption-prone event contracts at the listing stage, particularly contracts on granular outcomes that market participants can influence or control, and events whose listing creates severe incentives to engage in socially harmful conduct or to disclose confidential information.

Related Working Papers

Subscribe