Ranked-Choice Voting Is Central Planning for Elections

In a constitutional republic, elections are not the purpose of government. Liberty under law is. We hold elections to select representatives whose power is then bound by a written Constitution. So when ranked-choice voting arrives with promises to end polarization, eliminate spoilers, and rescue "wasted" votes, the question worth asking is not whether the counting sounds clever. It is whether the system makes self-government more accountable, or whether it applies central planning to the consent of the governed. My answer: it is closer to the second.

Ranked-choice voting, sometimes called instant-runoff voting, asks voters to rank candidates in order of preference. If no one wins a majority of first choices, the last-place candidate is eliminated and his voters' second choices are redistributed. The process repeats until someone crosses 50 percent. On paper, that seems reasonable. But the problem is not the arithmetic. The problem is the assumption that a single algorithm can take millions of incomplete, contradictory, and often casual human judgments and turn them into a clear expression of the public will.

The knowledge problem comes for the ballot box

Friedrich Hayek explained why that assumption fails. In his 1945 essay "The Use of Knowledge in Society," he observed that the information a society needs never exists in one place. It is dispersed across millions of minds as "incomplete and frequently contradictory knowledge." Markets work because prices condense all that scattered information into signals that no central planner could ever gather by hand.

Ranked-choice voting attempts the political equivalent of the planner's error. It presumes that voters actually possess a meaningful ranking of five or six candidates. Most voters do not. The typical person knows something about one or two names on the ballot, and often not much at all. Asking him to rank the rest is not revealing a real preference. It is requiring him to invent one. The counting algorithm then treats those manufactured rankings as if they were genuine expressions of judgment. That is not better self-government. That is worse information dressed up as sophistication.

Madison already knew the answer

The Founders did not expect political conflict to disappear. In Federalist No. 10, James Madison wrote that "the latent causes of faction are thus sown in the nature of man." They cannot be removed without destroying liberty. The remedy is not to eliminate faction but to control its effects: a large republic, representation, separation of powers, and the competition of interests.

Ranked-choice voting is an attempt to engineer consensus by algorithm. It sends a clear instruction to candidates: do not stand for too much, because you need to be everybody's second choice. The result is not a more civil politics. It is a mushier politics, in which the safest route to office is to blur differences and avoid hard positions.

Consider a simple example. One candidate is the first choice of 40 percent of voters and the last choice of the remaining 60 percent. Another candidate is the second choice of nearly everyone but the first choice of almost no one. In a ranked-choice system, the second candidate can win. He may not have persuaded many people he was right, but he offended almost nobody. That is not consensus. That is the triumph of the unobjectionable.

Madison wanted faction to check faction, not to be dissolved into a synthetic majority. A republic needs candidates who say what they believe so voters can hold them to it. When the system rewards the least objectionable candidate rather than the most principled one, accountability dies by a thousand second-choice transfers.

The seen and the unseen

Henry Hazlitt's Economics in One Lesson teaches us to look beyond the visible. The visible benefit of ranked-choice voting is easy to sell: no wasted votes, more names on the ballot, supposedly friendlier campaigns. The unseen costs are where the damage accumulates.

Consider Maine's 2nd Congressional District in 2018. Republican Bruce Poliquin won the most first-choice votes, just under 46 percent, but lost after lower-ranked votes from eliminated independent candidates were redistributed to Democrat Jared Golden. I am not arguing the result was illegitimate under the rules. I am pointing out what the rules revealed: the winner's "majority" can be constructed from voters who did not choose him first and who, in many cases, knew very little about him beyond a party label.

That is a second-choice mandate, and it is thinner than the clear first-choice mandate of a traditional election or a two-round runoff. It also changes how the winner governs. If your authority rests on being the backup option for half the electorate, the incentive is to avoid any stand that might cost you those backup rankings. That is not a formula for leadership. It is a formula for permanent triangulation.

Then there is ballot exhaustion. Every time a candidate is eliminated, some voters have no remaining ranked choice. Their ballots simply stop counting in the decisive final round. The system tells them their vote mattered, but when the actual choice was made, they were no longer in the room. That is not equal weighting of every citizen's judgment. It is a system where the depth of your ranked guesses determines how much your vote counts.

The constitutional layer

Originalism matters here. The Constitution leaves the time, place, and manner of elections largely to the states, but it does not authorize a state to dilute the value of a citizen's vote. The Supreme Court has long held that votes must be weighted equally. Ranked-choice voting sits awkwardly against that principle.

There is also a subtle but serious problem known in voting theory as non-monotonicity. In some ranked-choice eliminations, a voter can actually cause a candidate to lose by ranking him higher. That is not a quirk of one implementation. It is inherent to certain elimination algorithms. A rule-of-law republic cannot rest on a voting method whose results can move in the opposite direction of the expressed preference.

Justice Antonin Scalia would have asked the simpler originalist question: what does the text authorize and what does it presuppose? The Constitution does not require a pure majority runoff, but it does presuppose that elections produce a clear and intelligible expression of the people's will. When voters cannot explain how their ballot was counted, and when the winner's mandate is a tangled web of second and third choices, self-government becomes less transparent, not more.

What we should do instead

I am not romanticizing the old system. Third-party spoilers are real. Low-turnout runoffs can be messy. But the answer is not to centralize preference aggregation into an algorithm that pretends to solve faction. The answer is to strengthen the institutions that make political choices clear: parties that stand for something, primaries that produce coherent nominees, and runoffs where voters make a final choice with full information after the field has narrowed.

A traditional two-round runoff does something ranked-choice voting cannot. It gives voters a second look after the first round, when they know who the final contestants are and what those contestants actually stand for. It does not ask them to rank strangers before the campaign has even been joined.

Better yet, accept that in a free society, a plurality winner is not a scandal. If three candidates split the vote and the one with the most votes wins, that is a legitimate result. It may not feel mathematically pure, but it is honest. Honesty is worth more than a manufactured 51 percent.

The stakes

The people pushing ranked-choice voting are not evil. Many sincerely believe they are making elections more rational. But as Hayek warned in The Road to Serfdom, the road to lost liberty is often paved with well-intentioned central plans.

Elections are not a machine for producing consensus. They are a means by which a free people choose among real alternatives and then hold the winners accountable. Ranked-choice voting substitutes an algorithm for that process, and in doing so, it weakens the very thing a republic needs most: clarity.

I want candidates who say what they mean and mean what they say. I want voters who choose among them with open eyes. I want a system where the winner's mandate is plain enough to be judged. Ranked-choice voting gives us instead a muddled plurality dressed up as a majority, produced by a counting method most voters cannot explain.

That is not self-government. That is central planning for consent. The cure for our political divisions is not a better algorithm. It is a citizenry that remembers what it believes, and a political class that still fears the people enough to tell the truth.