Two days after the polls closed, the accounting began. OpenSecrets puts the 2026 midterms at $11.9 billion, the most expensive midterm in the country’s history, up from $8.9 billion in 2022. Outside spending is projected at $3.9 billion, an 85 percent increase in four years. Dark money, the contributions routed through entities that disclose no original donor, has passed $1.47 billion and will finish above the 2024 presidential cycle. One man gave $445 million.

That money bought an argument in perhaps eighteen House districts. The Cook Political Report rated 18 of 435 seats as toss-ups as the cycle closed, with 368 seats rated solid for one party or the other. The 2024 cycle was worse in one respect: Unite America’s analysis found that 87 percent of House races were effectively decided in the primary, by roughly 7 percent of eligible voters.

Those two paragraphs are usually filed under different headings, money and competition, and treated as separate diseases with separate cures. They are one disease. The country has built an election system in which the decisive electorate is small, self-selected, and cheap to reach, and in which the general election is mostly a formality staged for the cameras. Campaign finance reform, ranked-choice voting, and open primaries are three ways of changing who that decisive electorate is. That is why they are worth taking seriously together, and why each one alone disappoints.

Four symptoms, one cause

Start with what is actually wrong, stated in numbers rather than adjectives.

The primary is the election. In a district drawn to favor one party by fifteen points, the November result is not in doubt, so the only contest that matters is the spring one. Unite America counted 169 members of the 2024 class who ran unopposed in a primary for a seat their party could not lose, which means they were elected without winning a single contested vote.

The primary electorate is small and unlike the country. Closed primaries restrict the ballot to registered partisans. Twenty-two states held closed presidential primaries or caucuses in 2024, and 15 states close their congressional primaries; in those, 15.7 million registered independents could not vote in the election that often decided the outcome. Pew’s April 2026 survey found 30 percent of adults call themselves independents and another 14 percent something else, against 28 percent Republican and 27 percent Democrat. The gate is small, and the people outside it are the plurality.

The money is concentrated and partly anonymous. Candidate committees are capped at $7,000 per donor per cycle and still spent about $3.7 billion this year. The uncapped side is where the growth is: the Senate Leadership Fund spent $489 million and MAGA Inc. $353 million. A voter cannot audit a coalition assembled from shell nonprofits, and increasingly cannot even name it.

Nobody is enforcing the rules. The Federal Election Commission lost its quorum on May 1, 2025, and has not recovered it. It cannot open investigations, issue fines, write rules, or give advisory opinions. Its enforcement backlog doubled from 154 cases to 307, and it has not approved a dollar in civil penalties in that time. Craig Holman of Public Citizen described the result to CBS News: “Everyone knows you can do whatever you want this time around.”

Put those together and the pathology is not polarization in the abstract. It is audience selection. A member of Congress with a safe seat answers to the several thousand people who vote in a low-turnout primary, plus the donors whose money reaches them through committees they do not control, and to almost nobody else. Persuasion is not the job. Turnout among the already-convinced is the job. That is why the country’s politics sounds like a rally and governs like a whip count.

Campaign finance: counter the money, do not chase it

The constitutional constraint is real and it is not going away. Citizens United (2010) and the D.C. Circuit’s SpeechNow decision that followed treat independent spending as protected speech and permit unlimited contributions to committees that do not coordinate with candidates. Any reform platform that begins by promising to reverse those cases is promising a Supreme Court appointment, not a policy. The workable half is the half that does not restrict speech.

Disclosure first, because it is cheap and it works. Alaska’s 2020 ballot measure paired its new election system with “true source” rules that force the disclosure of the original funder behind a chain of political entities. Those rules have done visible work: reporters traced $1.5 million from Elon Musk into the state’s 2026 repeal campaign, and $10.4 million from out-of-state donors into the campaign against it, because the law required the chain to be traceable. Ballot Measure 2 would have repealed the disclosure provisions along with ranked-choice voting and open primaries. Alaska voters kept all three.

Then build up small donors rather than capping large ones. New York City’s matching program pays participating candidates public funds against small contributions from city residents, so a $10 gift from a resident can be worth as much as $90 to the campaign. The effect shows in the donor pool. In the 2013 citywide elections more than two-thirds of contributions were $175 or less, and more than 90 percent of the money raised came from individual contributors rather than committees, unions, or PACs. The Brennan Center’s study of the program found that small donors to 2009 City Council candidates came from a far broader array of city neighborhoods than small donors to 2010 State Assembly races did. That is a policy that changes whom a candidate must talk to, which is the only thing that reliably changes what a candidate says.

And fix the referee. A six-member commission with a four-vote threshold and a Senate that will not confirm is not a regulator. Restoration is unglamorous and it is a prerequisite: disclosure laws without an agency to receive and publish the filings are press releases.

Ranked-choice voting: a majority, not a miracle

Ranked-choice voting does one thing well and is oversold on the rest, and both halves matter for whether it survives a campaign.

It removes the spoiler. Under plurality rules, a voter who prefers a third candidate must decide whether to waste the vote or risk electing the candidate they like least. Under ranked choice, that calculation disappears; if the first choice is eliminated, the ballot moves to the second. It also produces a majority winner in a single election instead of a low-turnout runoff. That is the case, and it is a good one.

The record is now long enough to test. Maine adopted it statewide in 2016 and has used it since 2018. Alaska adopted a top-four primary and ranked-choice general in 2020 and used it in 2022, electing Mary Peltola to the House under the new rules. FairVote counts 49 jurisdictions using ranked-choice voting as of March 2026, reaching nearly 14 million voters, including two states, the District of Columbia, and Multnomah County’s first use this November. Washington, D.C. used it in its June primaries and saw its highest primary turnout in thirty years.

Alaska also shows the price of adoption. Repeal was on the ballot in 2024 and failed by 743 votes. It was on the ballot again in 2026, with the repeal campaign funded at $2.5 million against $10.4 million for its opponents, and with nearly every precinct reporting it trailed 51.9 percent to 48.1 percent. A system that survives two repeal attempts by a combined margin of about four points has not been accepted. It has been tolerated, and only while its defenders outspend its opponents more than four to one.

The honest objections deserve a hearing, because the sales pitch usually skips them. Ballots can be exhausted: if a voter ranks only one candidate and that candidate is eliminated, the ballot stops counting. MIT Election Lab’s study of Maine’s 2018 gubernatorial primary found 10.7 percent of ballots eventually exhausted through undervotes, with 0.7 percent carrying a double undervote that exhausts immediately. The errors were frequent and their effect on that outcome was small, but the pattern is not nothing. The evidence on turnout is mixed at best, and ranked choice does not guarantee that a final-round winner holds a majority of all ballots cast. It is a repair to the spoiler problem and to plurality winners, not a cure for civic disengagement.

Open primaries: the gate itself

If the primary is the real election, then the question of who may vote in it is the most consequential question in American politics, and it is the one almost nobody asks.

The numbers are stark. In 15 states, 15.7 million registered independents are barred from congressional primaries. Unite America’s polling of those excluded voters found 77 percent consider their exclusion unfair, 87 percent support opening primaries to independents, and 70 percent believe both parties are too extreme. Those are not swing voters waiting to be courted by a better message. They are voters who have been told, by rule, that the only election that matters is not for them.

The mechanism matters as much as the principle. A top-two primary, as California and Washington use, sends the two leading vote-getters to November regardless of party; it opens the ballot to everyone but can produce two candidates from the same party and squeeze minor parties out of the general entirely. A top-four primary, as Alaska uses, keeps four candidates alive into a ranked-choice general, which preserves space for independents and minor parties. Massachusetts voted on a top-two version this November and rejected it, 61.1 percent to 38.9 percent, after a campaign in which both parties opposed it and two former governors from opposite parties supported it. The parties will defend their gate, and they have the machinery to do it.

That result is not an argument against open primaries. It is a map of the difficulty. The reform has to be passed by the same voters who have been told for a century that primaries are a party function, against the two organizations best equipped to mobilize against it, and it has to survive a campaign in which the opposition can spend without limit.

What the three reforms do together

Taken separately, each is a partial measure with a plausible attack. Campaign finance reform cannot touch independent spending and is dead at the federal level while the FEC has no quorum. Ranked-choice voting exhausts some ballots, and the evidence on turnout is mixed. Open primaries can be repealed, as Alaska has now tried twice, or rejected outright, as Massachusetts did.

Taken together, they change the same variable from three directions: the size and composition of the electorate a politician must satisfy.

  • Open the primary and the incumbent’s electorate widens from the party base to everyone, including the 44 percent who belong to neither party.
  • Rank the general and a candidate must seek second-choice support, which means speaking to voters whose first choice is someone else, which is the definition of persuasion.
  • Match small donors and disclose the chain, and the money a candidate needs can come from many people instead of a few, which makes the widened electorate worth courting rather than merely enduring.

That is a coherent program, and it is also a long one. The federal path runs through a Supreme Court that has already ruled, an agency that cannot meet, and a Senate that will not confirm. The state path runs through ballot measures, which is how Maine, Alaska, and every city on FairVote’s list got there, and which is why the Alaska result matters more than any op-ed: a state that adopted all three at once, in 2020, has now defended the package twice against well-funded repeal. It is the only working proof that the combination is politically survivable as well as administratively workable.

The objection worth taking seriously

The strongest argument against all of this is not that it fails. It is that it works unevenly, and that the unevenness is not random.

Ranked-choice voting helped Peltola in Alaska and helped moderates in Alaska’s legislature, which is why the state Republican Party and the president campaigned against it. Open primaries widen the coalition a member must hold, which is a benefit if you want governing and a cost if you want a movement. Campaign finance matching funds mostly help candidates who can raise many small donations, which favors candidates with networks and disadvantages candidates with neither networks nor money. Every one of these reforms is a change in who holds leverage, and the people who currently hold it will describe that change as a rigging.

They will also be partly right about one thing: no balloting rule manufactures courage. A member who wants to run for president will still play to the primary electorate of a national party, and no state law reaches that. What these reforms can do is narrower and real. They can make the local electorate larger, make the money traceable, and make the second choice worth something. A republic that cannot do those three things is not being governed by its voters. It is being governed by the people who selected the choices.

Sources

PRH | huffmanwrites.org | © Philip Huffman