Prediction Markets: How they work and what they're good for

Betting markets have become a more popular source for real-time developments of the upcoming U.S. elections. But not everyone seems to have caught on.

The Economist quotes Nate Silver,

Venues such as Betfair are populated by foreigners who may be less well informed. Nate Silver, an elections forecaster, has derided taking such markets seriously as checking in with the “Scottish teens”."

An earlier post by self-declared election expert Andrew Gelman says that he does not bet on elections because it’s unfair to take money of the other participants, who are more ignorant than he is:

The related moral issue that, to the extent that I legitimately am an expert here, I’m taking advantage of ignorant people, which doesn’t seem so cool.

ABC News director of data analytics, G Elliot Morris, suffers a similar delusion to Gelman:

All three of Silver, Gelman, and Morris seem to believe they would easily make money in these markets. As Gelman says, his only fear would be “collecting” his winnings. It’s hard to say whether Silver’s, Gelman’s, and Morris’s dismissiveness towards betting markets reflects ignorance or disingenuousness. One of their assumptions seems to be that since they are smarter than the dumbest participants in the market, then they are smarter than the market.

That’s just not how markets work. In fact, it’s the opposite of how they work. To beat a market requires being smarter than the smartest in the market. And here’s why.

Market Dynamics: Incentives, Value, and Information

Market participants respond to incentives: every trade reflects both a buyer’s and a seller’s response to complementary incentives, the buyer’s incentive to buy and the seller’s incentive to sell at a given price. In responding to their incentives, traders seek to extract value from the market. In exchange for the value they extract, traders contribute either liquidity by injecting capital and driving volume or information by helping to bring the market price into closer alignment with the underlying event’s true probability of occurrence. The different motivations of market participants create a robust dynamic that enables markets to serve the public interest for hedging risk and aggregating information.

Whether a prediction market adequately serves the above hedging and information aggregation purposes is largely driven by who its participants are. Attempts to restrict or regulate market participation in favor of certain participants or specific participant motivations distorts market dynamics and undermines the market’s ability to serve its purpose.

Event contract participants fall into three general categories: hedgers, gamblers, and vultures.

Hedgers buy and sell contracts for the purpose of managing exposure to risk on future contingent events. Gamblers buy and sell contracts for recreation or speculation. Vultures buy and sell contracts seeking financial gain, by using superior information and pricing methods.

The involvement of participants from all three categories is necessary to the healthy function of any market. Wherever a market exists, there will be participants motivated by profit and those who only want to gamble. This is true of all currently regulated and operating stock, commodities, and futures markets in the United States and abroad.

From a regulatory perspective, markets are often permitted on grounds that they are a vehicle for hedging risk or aggregating information. Although gamblers and vultures may seem contrary to this stated market purpose, markets can only serve these purposes if participants from all three categories are able to trade.

Hedging risk and aggregating information are two primary purposes of event contracts with clear public benefits. Achieving either purpose requires a healthy demographic split among the three participant categories above. A party wishing to hedge risk must find a counterparty willing to take on the risk being hedged. In general, however, one cannot expect that the risks of participants will align so perfectly that every dollar seeking to be hedged on one side can be matched with a dollar seeking to hedge the other side. There is often an imbalance among those seeking to hedge. Gamblers help to resolve this imbalance, as their desire to speculate leads to the injection of the additional liquidity needed for the market to service those looking to hedge risk.

But Gamblers and Hedgers alone cannot serve the market’s information aggregation purpose. Gamblers and hedgers are neither informed nor price sensitive enough to generate a sharp market price signal. Vultures are sophisticated operators who often have access to better information and more reliable pricing methods than their non-profit seeking counterparts. Their deep understanding of the market and precise approach to trading allows them to trade profitably by identifying price discrepancies. For example, an event that has a 30% probability of occurring but for which a contract is selling at $0.20 offers the savvy trader an expected profit of $0.10 for every $0.20 invested, for an expected 50% return on investment. In exchange for the opportunity to profit, these informed participants contribute to a more stable, accurate market signal. More accurate pricing, in turn, has the positive internality of providing additional liquidity to the market and fairer pricing to those seeking to hedge risk and the positive externality of informing the general public (many of whom do not directly participate in the market, and therefore enjoy this benefit at no cost).

Understanding the interplay among the above three categories of market participants is essential when understanding how and why markets work (and also why the experts who do not test themselves against the markets would likely get embarrassed if they were to try).