When the price mechanism meets the ballot box

Politico reports that a plurality of Americans believe betting on election outcomes should be illegal — a finding that, on its surface, looks like ordinary public squeamishness about gambling. But I think something more interesting is at work here, and it is precisely the kind of institutional question that political economy is fitted to examine.

I have spent considerable effort arguing that free exchange, operating within a proper framework of law and honest dealing, tends to produce outcomes no single planner could improve upon. The price that emerges from willing buyers and willing sellers carries real information. A prediction market on, say, the harvest price of wheat, or the creditworthiness of a merchant house, does genuine work: it aggregates dispersed knowledge and disciplines the speculator who guesses wrong. I do not quarrel with that mechanism in its proper domain.

But markets require, for their honest function, that the underlying thing being priced is independent of the act of pricing it. When I buy a share in a pin manufactory, my purchase does not change how many pins the factory will produce. An election is different. A sufficiently large and visible market in election outcomes may influence those outcomes — by depressing the turnout of those who believe their side has already lost, or by giving a wealthy faction a second instrument, beyond mere political spending, with which to shape public perception. The price would no longer be a report on the world; it would be a tool for remaking it. That is a corruption of the mechanism itself, not merely of the election.

There is a second objection, which belongs less to political economy than to moral philosophy. In The Theory of Moral Sentiments I argued that sympathy — the imaginative capacity to enter into another person's situation — is the foundation of the moral order, including the moral order that underpins contract. When citizens vote, they are asked to exercise exactly that sympathetic imagination on behalf of the public good. Surrounding the act with a betting market invites them instead to treat the outcome as a financial instrument — to ask not what is right for the commonwealth but which way is the smart money running. The two questions are not the same, and habituating the public to the second at the expense of the first does lasting damage to the civic character that free institutions require.

I do not say prediction markets are without any merit. As a matter of inference, not recollection, I am told they have at times produced more accurate forecasts than polling. That is not nothing. But accuracy of forecast and legitimacy of outcome are different goods, and a device that sharpens the former while corroding the latter is a poor bargain. The Americans surveyed by Politico may not have articulated the argument in these terms, but I suspect their instinct points in the right direction.

The institutional question, then, is not simply whether to permit or prohibit, but what framework governs the exchange. At minimum, any such market should be prohibited from accepting wagers from candidates, their campaign committees, or any entity capable of materially affecting the outcome being priced — just as we would regard it as fraud for a merchant to buy futures contracts in a commodity he intends to corner. The deeper question is whether the democratic process, as a public good, deserves the same kind of protective framework that I argued in The Wealth of Nations applies to defense, justice, and basic education: not a market, but an institution maintained at public expense precisely because its value cannot be captured in any private price.