On Monday, the U.S. Commodity Futures Trading Commission issued an order prohibiting the marketing of a new set of derivatives that would have enabled traders to bet on the winners of national elections. While these “political events contracts”—the latest development in the world of prediction markets—have their partisans, the CFTC’s order was based on fundamentally sound logic that should be applied more broadly. Financial instruments that serve primarily as a means of speculation rather than hedging should be banned, just as gambling is illegal in most contexts.
A simple, common-sense principle guides this distinction. Imagine two different scenarios. In the first, an oil producer faces the risk that oil prices will fall while a consumer faces the risk that they will rise. The producer and the consumer might use a futures contract to hedge this risk by contracting to transact in the future at a price fixed today, thereby effectively purchasing insurance against the risks they both face. The consumer no longer worries that a price increase will interfere with his daily commute, while the producer no longer fears that a price decline will force him to lay off his workforce. This insurance reduces the unpredictability of the individuals’ incomes, smoothing them out across different contingencies. Such transactions embody the valuable spreading of risk for which financial markets are justly praised.
Now consider an alternative scenario. Suppose that two individuals, neither of whom uses or produces oil, harbor different opinions about the future price of oil and decide to wager on it. Both parties willingly participate, because they think they’re each getting the best of their confused counterparty. Clearly, both of them cannot gain from this transaction, and the wager itself creates rather than reduces risk. While each party thinks it is getting the better of the other, both agree that on average both of them will be worse off because on average they will win and lose on the same number of bets, and both of their incomes will be less smooth and predictable on account of their wagering. As a consequence, this sort of speculation is socially harmful. That’s why gambling and wagers are heavily regulated or banned outright in nearly every country.
Some deride this common-sense economic logic, formalized recently by several economists, as “populist” or ignorant of the benefits that derivative securities supposedly bring. There is no doubt that many derivatives bring insurance and supply information to the market. But these benefits must be weighed against the costs of speculation. It is absurd to praise derivatives for reducing risk and allowing insurance without simultaneously condemning them when they achieve the opposite goal.
There’s a long tradition in the common law of drawing precisely this distinction between speculation and insurance. Back in the 18th century, the new-fangled “derivative” of the time was the life insurance contract. When life insurance policies were first marketed in Britain, people used them to gamble on the deaths of the great figures of the day—prime ministers, popes, explorers, and generals. It soon became clear that an instrument that was designed to spread risk—so that a dependent could survive if the breadwinner died unexpectedly—could also be used to speculate. To prevent such abuse, the British Parliament created the insurable interest rule, which provided that life insurance was only payable to an individual who faced a risk to her income from the death of the insured individual. The insurable interest doctrine cleanly cut through the confusion about the difference between speculation and insurance, banning the former while permitting the latter.
Our skepticism about speculation shouldn’t imply that we oppose all forms of gambling. In controlled and appropriate contexts, it can be a source of entertainment for people who are aware of and willing to accept the potential losses. But participants in financial markets are usually seeking financial security rather than entertainment, and they typically have little sense of the risks they are taking on. Also consider that speculators often effectively gamble with other people’s money, allowing the government and individual investors or creditors to pick up the losses while they collect the gains. And because of the systemic links in the financial system, the more risk that accumulates, the greater the chance of a systemic collapse like the one we experienced in 2007-08.
For all these reasons, we advocate the creation of a Financial Products Agency that would be charged with determining whether a derivative is likely to act primarily as a vehicle for speculation. This agency would have to weigh the harms created by speculation against whatever benefits would result from this new derivative.
In principle, for example, bets on the presidential election could be used for insurance. If you think you will be impoverished if Obama is re-elected, you could, in theory, hedge against his re-election by betting that he will win. But in reality, the effect of the presidential election on your income is probably close to nil. Thus, the CFTC was rightly skeptical of these new political events contracts—the mere possibility of insurance is not enough to refute the likelihood of large quantities of speculation.
A second potential benefit of allowing trading in derivatives is the information that they provide to market participants. The knowledge of the likely outcome of the presidential election provided by the wisdom of the crowds is useful for planning by businesses, individuals, and governments. But that information is only valuable to the extent that it enables real economic decisions to be made more effectively. Creation of derivatives may do as much to spur wasteful expenditures as to supply valuable information to markets in a timely fashion. Innovators pressing for the approval of a new product would have to show that the additional gains from the latter outweighed not only the harms from the former, but also any net costs from speculation compared to insurance.
Other arguments made in favor of allowing derivatives to trade unchecked are mostly variations of these simple tests. For example, it is often argued that derivatives help create liquidity in markets, making it easier for individuals to take on or unwind desired positions. But this just pushes the analysis back one step: Liquidity is neither inherently good nor bad. If derivatives make markets more liquid for speculators, they are harmful; if they make them more liquid for individuals seeking insurance, they are beneficial.
The simple logic of projecting the demand for a product arising from insurance compared to speculation provides a sound basis for determining whether derivatives should be traded. We know this is possible, not fanciful. The same logic was the basis for the CFTC’s order prohibiting political events contracts, and similar demand projections are already used by the Department of Justice and Federal Trade Commission to determine whether mergers should proceed because of the synergies they allow or should be blocked because they will reduce competition.
In a forthcoming article, we provide several examples of such analyses, arguing that credit default swaps, collateralized debt obligations, and many derivatives based on statistical properties of equity returns (such as correlation and volatility) should have been banned. On the other hand, we believe that index funds, derivatives based on real estate indices, and derivatives based on national income measures likely should have been allowed.
The larger question raised by the CFTC’s decision to ban political betting is why it does not use the same reasoning to regulate financial derivatives, which are conceptually no different from political events derivatives. The major difference between them is that financial derivatives have done a lot more damage.