Prediction markets are in the regulatory crosshairs. In the United States, within the span of a few months this year, federal prosecutors brought the first criminal insider trading case involving an event contract against an Army master sergeant with a top-secret clearance who allegedly traded on classified information about an upcoming military operation. Next the CFTC and DOJ pursued a Google engineer who made about $1.2 million trading contracts on Google’s unreleased “Year in Search” rankings. A House Oversight Committee investigation and at least eight bills in the 119th Congress quickly ensued.

Most of the legislative proposals envision importing some version of insider trading law from securities markets to prediction markets. In a new paper, we argue that adjustments in regulatory approach should be made to reflect the very profound differences between securities markets and prediction markets. The case for strict insider trading regulation in securities markets does not automatically translate to prediction markets. The distinctive regulatory challenge that prediction markets do present is that some, but not all, contracts on prediction markets create a serious moral hazard  of giving market participants an incentive to engage in corrupt, illegal, or dangerous actions in order to rig the outcome of the contract on the prediction market. These corruption-prone contracts, combined with other institutional features of prediction markets such as their small size and limited social utility, indicate that prediction markets require different regulatory priorities than securities markets.

The two markets differ profoundly in their economic and social importance. Securities markets allocate capital across the economy: U.S. public equities alone represent roughly $60 trillion in value, public markets supply about three-quarters of the financing for non-financial corporations, and a majority of American households hold equities. Prediction markets, by contrast, are zero-sum side bets with less than $500 million in outstanding contracts. They allocate no capital, finance no businesses, create no employment opportunities, and safeguard no precious savings or investment capital. Their one and only social product is informational: the probability estimate embedded in the contract price.

This distinction should make a difference in regulator priorities. In securities markets, the insider trading prohibition is best understood not as a mandate of informational “fairness”—a rationale the Supreme Court abandoned in Chiarella and Dirks—but as protection of property rights in valuable corporate information, a theory completed by the misappropriation doctrine in United States v. O’Hagan. Manipulation, meanwhile, as scholars have observed, is a somewhat remote, second-order concern on securities markets: Traders cannot manipulate prices because manipulative trading causes an artificial price spike that is inevitably too brief to allow profit-taking by a would-be manipulator.

While insider trading is the main priority on securities markets and manipulation is a minor concern, the regulation of prediction markets presents the opposite concerns.  Because the price of event contracts is the product, informed trading on such markets is not a system bug at all. It is the feature, the only feature, that makes the product interesting to market participants and socially valuable to the rest of us. Insider trading is not as serious a problem on prediction markets as it is on securities markets. Recent empirical work shows that roughly 3 percent of accounts—persistently skilled traders processing public information faster, arbitraging across contracts, and betting against the crowd’s biases—drive price accuracy, while amateur, recreational traders supply volume but almost no information. Insider trading in the legal sense does occur, but it is localized and sporadic: fewer than 1,000 flagged accounts out of 1.7 million, about two-tenths of one percent of volume.

In contrast, the available empirical evidence in finance indicates that insider trading is routine on securities markets. Such trading is estimated to occur in one in five mergers and acquisitions and one in 20 earnings announcements, at least four times the prosecution rate. Insider trading on securities markets is more likely when the information is more valuable, when more people possess it, and in more liquid stocks. Certain structural features of prediction markets, particularly traders’ lack of systematic access to informational advantages, thin order books, conspicuous positions, and severe non-market penalties for the very people who generally possess event information limit insider trading on prediction markets.

This is not to say that prediction markets are free of regulatory challenges. The payoff for corruption-prone event contracts turns on events a market participant can materially influence: a referee’s next call, the particular words a CEO utters on an earnings call, a congressman’s attendance at the State of the Union, whether a website suffers an outage. Such contracts are an invitation to corruption that after-the-fact trading surveillance cannot efficiently police. We note that this problem is not unfamiliar. Insurance law’s insurable-interest doctrine, dating to the Life Assurance Act 1774, answered the wager policy—a stranger’s bet on another’s life—not by prosecuting murder more vigorously but by refusing to allow such contracts at all. Thus, the appropriate deterrent is ex ante, not ex post. We propose the same regulatory approach for corruption-prone event contracts on prediction markets: the exclusion of such contracts at the listing stage in order to prevent the corrupt incentive from arising in the first place.

The hard cases confirm rather than undermine this framework. Trading on misappropriated corporate secrets is already punished under CFTC Rule 180.1, as the Google prosecution shows. Trading on classified information is a breach of public trust properly addressed by national-security, ethics, and fraud law.

We recommend a regime built from existing tools: Exclude corruption-prone contracts through rebuttable, categorical presumptions at listing; abandon  inequality of information as a listing consideration; leave participant-conduct rules primarily to exchanges, which internalize the relevant tradeoffs; police breaches of confidence at their source; and adopt a no-parity rule. The Commodity Exchange Act, organized around the listing decision, already contains the architecture this approach requires. What is needed is not new law, but attention to the right regulatory challenges.

Jonathan R. Macey is Sam Harris Professor of Corporate Law, Corporate Finance, and Securities Law at Yale Law School, and Luca Enriques is professor of business law at Bocconi University. This post is based on their recent paper, “Different Markets, Different Regulatory Priorities: Insider Trading, Manipulation, and Corruption-Prone Event Contracts in Prediction Markets,” available here.