Were the Midterms a Dress Rehearsal for Futarchy?

A laptop screen displays trades on the Kalshi website, April 16, 2026, in Portland, Ore. (AP Photo / Jenny Kane)

For decades, election betting was largely the domain of academic researchers running small-scale experiments. Today, it is a mainstream media product. By the time polls closed on November 3, traders had wagered more than $1.48 billion on the US midterms. 

Ahead of the results, platforms like Kalshi and Polymarket gave Democrats a 93 percent to 95 percent chance of winning the House and roughly a 75 percent shot at the Senate. They correctly predicted almost every individual winner, while traditional polling underestimated the scale of several Democratic victories.

Markets are decent at the yes/no question, but shaky on the how much, and they miss in ways that are hard to see in advance; some high-profile misses during the US primaries and in Brazil’s presidential election have shown they are fallible.

Prediction markets are a good description of the way the political atmosphere is behaving, but they should not be the decider. They were originally built to tell us what might happen, not what should happen. But as they have gained attention and influence, some economists have started asking whether markets deserve a bigger role, and have considered changes to the governing structure that would take better advantage of their growing power.

Robin Hanson, an economist at George Mason, first wrote “Shall We Vote on Values, But Bet on Beliefs?” in 2000. He starts from the idea that every institution is an information machine, and “our info institutions are those within which we induce, express, and evaluate the acquiring and sharing of info.” He claims speculative markets do this role well, while democratic institutions they rely on fail in many ways. 

Hanson’s proposed alternative is futarchy. Voters would vote based on their values and what they want, and betting markets would say how to get it. He describes it in his manifesto as: “elected representatives would formally define and manage an after-the-fact measurement of national welfare, while market speculators would say which policies they expect to raise national welfare.”

In practice, elected officials would define a measure of national welfare (he starts with GDP but suggests adding things like health, inequality, and the environment), and traders would bet on which policies raise it. A policy gets adopted only if the markets expect it to improve that measure. Mechanically, that means a pair of markets for each policy: one conditional on it passing and one on it failing. You compare the two prices, and the bets pay out later, when the welfare number comes in. 

Hanson claims futarchy is ideologically neutral. It could produce anything from extreme socialism to minimal government, depending on what voters want and what speculators think delivers it. Unsurprisingly, the paper walks through dozens of design issues and even offers gentler small-scale versions, like advisory markets for company decisions and monetary policy. Critics have also pointed out apparent logical gaps, so futarchy can’t be more than a new lens for looking at the prediction markets and evaluating their role in politics. The midterm markets are a tiny, accidental version of Hanson’s idea, sans the governing part. As of now, they only predict who wins, not which policy works. 

Even in that limited role, the rapid integration of betting markets into political journalism presents its own complications. Kalshi, which serves as CBS’s official prediction market sponsor and holds partnerships with CNN and CNBC, now sees its live odds broadcast directly alongside actual election returns. By placing speculative prices next to hard vote counts, networks risk turning market sentiment into a self-reinforcing part of the story, granting traders outsized authority over the public narrative.

The midterms showed that prediction markets can generate valuable signals, yet they also highlighted the steep barriers to anything resembling futarchy. Markets missed margins of victory, misread foreign elections, and, crucially, still relied on news organizations like the Associated Press to officially declare the winners so payouts could be settled.

In an environment where betting odds share screen time with real-time results, and where most Americans remain skeptical of wagering on elections, handing genuine policymaking power to financial speculators would place immense trust in an institution that is still young, relatively thinly traded, and heavily shaped by the media covering it. For now, prediction markets are best treated as one imperfect data point alongside traditional polling and on-the-ground reporting.

Hanson’s idea is valuable only because the power of markets is growing so much, and we must consider what that would entail and how much risk we are taking on.

The Zeitgeist aims to publish ideas worth discussing. The views presented are solely those of the writer and do not necessarily reflect the views of the editorial board.