Prediction markets have exploded in popularity, transforming from academic curiosities into multibillion-dollar, consumer-facing platforms that facilitate trading on politics, sports, geopolitics, and even clinical-trial outcomes and FDA approvals. To date, attention has largely been focused on whether prediction markets should be legal and if so who should regulate them. Yet one question has received little scholarly attention: What happens if a prediction market fails? We address this question in our new article.
Prediction markets, and the event contracts sold on them, include features of gambling, derivatives exchanges, and platform marketplaces. They also have features distinct from those areas, leading the federal government to treat them as derivatives under the auspices of the CFTC and states to view them as sports-betting enterprises subject to state licensing. A recent U.S. circuit court split on the legal status of prediction markets has created further legal uncertainty that has spilled over into bankruptcy law.
Bankruptcy doctrine generally depends on firms possessing assets and liabilities with readily-ascertainable valuations, predictable legal priority, and clear rules that apply during insolvency. Yet event-contract assets are difficult to value. If treated as derivatives, bankruptcy courts might be compelled to interpret ambiguous event contract provisions in a dysfunctional market while applying familiar derivatives netting and close-out rules where the quoted price may no longer be indicative of value. Since event contracts often depend on contingencies that have not yet occurred, a judge might need both to interpret the contract and to determine whether the event in question had occurred—a far cry from merely enforcing a preexisting agreement, and a far more difficult valuation task.
Basic components of bankruptcy doctrine are also difficult to map onto prediction markets. Blockchain-based governance arrangements used by some prediction markets may make it difficult to identify the debtor, determine who may represent the estate, classify the claims of token holders, and administer assets. Claims on unpaid event contracts would have uncertain priority because different priority is assigned to claims treated as customer claims on a platform, unsecured creditors with contractual claims, contingent financial claims with setoff rights, claimants with a custodial arrangement, equity holders in a speculative enterprise, or even tort claimants against an illegal business. The applicability of section 365 of the Bankruptcy Code, a powerful tool that allows debtors to reject executory contracts, is also uncertain because it is not clear that event contracts are executory or how damages for rejection should be calculated. Finally, even though settlement typically plays a central role in bankruptcy resolution, the tests that courts must apply to ratify a settlement likely do not fit uncertain event contracts.
Prediction markets also pose an antecedent question: Do the central tenets of insolvency law apply at all? If prediction markets are governed as derivatives marketplaces, then the safe harbors that exclude derivative contracts from bankruptcy’s automatic stay, avoidance, and ipso facto rules should similarly exclude event contracts. Yet the contested rationales for these safe harbors apply with even less force to prediction markets than they do to other derivatives markets. Alternatively, if prediction markets are treated as illegal gambling enterprises, they may be excluded from bankruptcy protection just as cannabis companies are. A third alternative may be for states to regulate prediction markets almost exclusively, as they do the insurance industry, though this would likely increase compliance costs, decrease market efficiency, and encourage regulatory arbitrage.
The practical and doctrinal problems that a prediction market insolvency would present contribute to two enduring debates in bankruptcy scholarship that extend far beyond prediction markets. First, valuation disputes are central to bankruptcy, and bankruptcy practice has evolved as courts have increasingly turned to markets for evidence of valuation. Scholarship has charted this trend, incorporating it into debates about the efficiency of bankruptcy’s longstanding absolute priority rule as compared with the perhaps fairer, but more-difficult-to-value process argued for by proponents of relative priority. These difficulties would be exacerbated in a prediction-market bankruptcy, since event contracts are inherently probabilistic assets whose value in insolvency could diverge materially from their quoted prices. In short, the value of a book of event contracts (and by extension of a firm holding them or even offering them) poses a valuation problem akin to that identified by critics of relative priority, but does so even under absolute priority. Thus, the lesson that relative priority is inapposite for prediction markets applies more broadly to other firms with high-variance assets, where valuation will necessitate more assumptions about future states of the world, such as investment firms, startups, or providers of catastrophic insurance.
A second debate in the bankruptcy world relates to the appropriateness of specialized insolvency regimes for financial firms. Securities brokers, commodities merchants, and banks are resolved outside of bankruptcy under procedures administered or overseen by SIPC, the CFTC, or the FDIC, respectively. These schemes are typically justified by a need to protect investors, prevent macroeconomic risks, and reduce the costs in time and money incurred while resolving insolvency as compared with those in a bankruptcy proceeding. While some may believe that these models could apply to prediction markets, the smaller size, market concentration, and concerns about moral hazard for speculative retail betting weigh against designing a specialized regime for prediction markets. Indeed, some may even wonder whether a wind-down of a prediction market is appropriate at all, since a legal ban on many or all of its contracts would perhaps result in little residual value to save. But if a wind-down is appropriate, there is a more fundamental reason to be skeptical of a specialized insolvency regime for prediction markets. This is because the uncertain nature of claims in prediction markets may suggest that thoughtful litigation over entitlements may actually be better for claimants than a quick resolution of an uncertain claim. This highlights an aspect of specialized resolution that is often overlooked in debates about whether to apply such procedures more broadly: Speed is only beneficial when parties feel confident in the value of the asset they are receiving.
The various challenges discussed above argue for thoughtful regulatory design with insolvency in mind. As regulators consider whether to treat event contracts as derivatives, gambling, or something else, we advocate for considering bankruptcy treatment as part of the analysis. Any prohibition on event contracts should be combined with considered rules on how to unwind such contracts in bankruptcy, as such a prohibition would not be self-executing.
We suggest that event contracts should be regulated by economic function rather than form. Taking a page from the recent joint SEC-CFTC rulemaking on cryptocurrencies that classifies them into separate categories based on their characteristics, uses, and functions, we encourage a functional taxonomy for event contracts based on whether the event contract transfers preexisting risk or creates new risk. This would allow event contracts in the former category to be treated as derivatives, those in the latter to be regulated as gambling, and others to be governed by unique rules appropriate for their subject matter.
A prediction market insolvency would require a bankruptcy court not merely to allocate value, but to decide, contract by contested contract, whether the wager was lawful and who is entitled to be paid. Uncertainty over legal classification could also expand, curtail, or even eliminate bankruptcy’s ordinary tools for resolving financial distress. We argue that to avoid the chaos, bankruptcy should not serve as a substitute for regulatory choices that ought to be made in advance.
William Organek is an assistant professor of law, and Nizan Geslevich Packin is the Newberger Berman/Zicklin Family Chair and professor of law, at the Zicklin School of Business, Baruch College, City University of New York. This post is based on their recent article, “Don’t Bet on Bankruptcy: Prediction Markets and the Limits of Insolvency Law,” available here. The authors welcome any comments on the article.