Two dragons, one stone. How can we just leave it lying on the ground?
First, the electoral college. As a basic part of our federalist constitutional architecture, the electoral college was created to give some protection to smaller states, so they wouldn’t simply be drowned out by larger ones in the selection of the president.
Its structure is derived, of course, from Congress, which was shaped by the same dynamic. The number of electors — currently 538 — equals the number of senators and House members, plus three electors for the District of Columbia. Through this blended approach to representation, each category of state — large and small — had some comfort that they would have protection against the uncertainty of outcomes flowing from the democratic process. Those hedges allowed them to embrace an uncertain process — a.k.a., democracy — over guaranteed outcomes in their favor.
But things went awry in 1929. After we capped the House then, the national population of course kept growing, while the numbers of representatives did not, and so neither did the number of electors. Slowly, what had been a modest weighting in favor of less populous parts of the country has become an ever-more-robust one in their favor. This has increased the probability of outcomes where the electoral college vote and the popular vote will be at odds.
This problem is just going to get worse and worse, as long as the House fails to grow with the population.
Many recognize this problem. And many say, well, why don’t we just end the electoral college and go with the popular vote? I am not among those who think that would be wise.
Massachusetts, where I live, is a small state. It is likely to get smaller still, relative to the rest of the country. I think it’s good that small states get a smidgen of additional protection, but I think it should just be a smidgen, not the hefty dollop that has developed over time. A popular vote system wouldn’t strike the right balance, either.
We all think of ourselves as Americans, and that’s good, but the Founders’ federalist vision of our now-continent-spanning country was prescient. Think of the yawning red-blue divide we speak about so often: We continue to be a union of states and not a single, undifferentiated people.
Our states have different cultures. Our different jurisdictions produce quite different ecosystems for life. If we’re going to hold this thing together, we have to be honest about the continuing importance of states to our structure. This means the electoral college, weird as it is, remains necessary as a form of glue keeping our federal system intact.
What is true, though, is that the imbalance that has emerged by virtue of the capping of the House is a serious problem and requires correction. Fortunately, the correction is simple. A larger House of Representatives will reduce the advantage of small states to an appropriate level.
Once again, the logic of our dysfunction points to the same key solution: Let’s grow the House.
The problem of gerrymandering is harder than the issue of the electoral college. But we can slay this dragon with the same kind of stone — if we’re willing to pick up a bigger one. To reduce the corrosive effect that aggressive gerrymandering of district lines has on our system, we could grow the House still more ambitiously.
In earlier columns, I argued that a good target for the House would be 585 members, the number proposed in a bill put forward by of Rep. Earl Blumenauer (D-Ore.). This number would achieve what I called “deferred maintenance” — essentially, it would restore the members that states have had to give up so that representatives could be spread around to growing parts of the country under the fixed cap of 435.
Solving gerrymandering would require a larger number — 1,305, specifically. Before I dig into the details, let me note the good news: Our Capitol could accommodate even this number with some redesign of the interior space. As a matter of the architecture of our democracy, a solution to gerrymandering is available to us.
Why is 1,305 the magic number? You’ll notice that it is exactly three times 435. At 1,305, we could simply take all the districts we currently have and let them elect three representatives instead of one.
That alone won’t kill the gerrymandering dragon, however. The solution comes in deciding how we would elect those three members. Nothing improves if we run an election in a three-member district as if it’s three separate elections, and the majority in the district effectively gets three people to represent it. That wouldn’t change anything. It certainly wouldn’t disincentivize gerrymandering.
Instead, we need to elect those three members in a single election using ranked-choice ballots, which allow voters to record their first, second and third choices. With ranked-choice voting, nobody’s vote can be counted for more than one candidate, so you can’t have majorities piling multiple votes on three candidates from their party or ethnic group and blocking everyone else. (To avoid having the third seat go to real outlier constituencies, it’s necessary to set the win threshold at 25 percent, and to require every candidate to get over that level by factoring in second- and third-place rankings.)
Let’s imagine how this would tend to play out. In such a system, sizable minority constituencies within a district will generally be able to elect at least one representative — regardless of whether those minority constituencies are partisan minorities or racial, ethnic or cultural minorities. In other words, districts that we currently think of as “all red” or “all blue” would probably end up represented by two members of Congress from one party and a third from the other.
When districts have three seats, drawing lines to avoid this outcome is nearly impossible. Gerrymandering will cease to be worth the effort. At five seats, it would become mathematically impossible, but then we would need a new Capitol building or accept a virtual Congress.
More important, this result would helpfully change the dynamics of accountability.
For starters, those three members — two from one party and one from another — would need to figure out how to work together on behalf of their shared district. And when the two major parties caucus in Congress, every geographical region of the country would be represented. Now, when Republicans caucus, there is no one from Massachusetts or Hawaii there. And when the Democrats caucus, there is no one from Oklahoma or Arkansas. On this model, every state would have some representation from each major party. This would make for much healthier practices of debate and deliberation.
Rep. Jamie B. Raskin (D-Md.) captures the stakes of our current pattern of gerrymandering exceptionally well in a vignette in his recent memoir, “Unthinkable: Trauma, Truth, and the Trials of American Democracy.” Watching Congress in operation, Raskin recalls a conversation he once had on the floor with a Republican colleague, Rep. Clay Higgins of Louisiana:
“‘What do you see when you look out over on our side and then over on your side?’ I asked Clay, who is a good-natured man with an awesome Cajun accent. He looked at me and said, ‘You tell me first.’ ‘When I look at our side,’ I said, ‘I see America today in all its glory: black, white, Hispanic, Asian-American, LGBTQ, straight, women, men, immigrant, native-born. And when I look at your side, I see . . . America in the 1950s.’ ‘When I look at your side,’ he said, ‘I see the coasts, New York and California. When I look at my side, I see the heartland.’”
Gerrymandering has accelerated our splintering. Congress operates the way it does because of it. But state legislatures are not likely to walk back from the mutually assured destruction of the gerrymandering arms race any time soon. But we can achieve bilateral disarmament by growing the House of Representatives.
What about party primaries?
Okay, one more renovation, since I think I may know what many of you are thinking: Will a bigger Congress really be any better than what we have now?
Let’s not mince words: Many of us already think Congress has too many nutjobs. It’s terrifying to imagine even more of them in a bigger Congress. And how could we even get this change through when any increase will require an act of Congress? It’s like a bad joke.
How many members of Congress does it take to change a lightbulb?
It’s going to be dark for a long time.
Any case for growing the House must include a path toward electing members willing to make this change. It must also include a pathway to sustaining or even improving candidate quality over time.
There is a solution, but here we have to shift focus to the states. Our national Constitution can lead us to forget that the greater part of the architecture of our democracy resides at the state level. We can make changes there that will cause the federal structure to shift. Major democracy renovations will require this.
The change we need at the state level is simple. We need to get rid of party primaries.
Instead of running a party primary and a general election, we need elections that start with an “all-comers preliminary” in place of the party primary, and then have a final round in November for the top finalists, regardless of party, from the preliminary.
In this case, we don’t even have to ask whether this change can be accomplished, because it already has, in Louisiana, California, Washington, Alaska and Nevada, though in slightly different forms. Louisiana, California and Washington move two on from their preliminary to the final election. Alaska moves on four. Nevada has recently voted in a mechanism to move five finalists forward. Both Alaska and Nevada chose an instant runoff for the final round, a version of ranked-choice voting.
Alaska is the most recent state to implement this method. In 2022, it resulted in the election of a moderate Democrat to the House, Rep. Mary Peltola, over the polarizing Republican Sarah Palin, and enabled the reelection of moderate Republican Sen. Lisa Murkowski. The state now has bipartisan representation for its federal delegation. A recent report documents that the system also resulted in winners in state legislative races who are more likely to work in coalitional, cross-partisan ways.
This shift of electoral method would fundamentally transform incentives in the way we need. Now, during primary season, candidates campaign to whatever sliver of an activist base happens to dominate their district. But with an all-comers preliminary plus final round, candidates have to appeal to the general electorate the whole way through the process. They will be rewarded for seeking to solve problems recognized by a majority of Americans — congressional dysfunction, for instance.
The shift would also change the tone and tenor of the media environment. Political consultants who work in California, for instance, run media differently there — seeking to connect to the whole electorate from the get-go. This reduces negativity and puts the pressure of greater public exposure on falsehoods and misinformation.
Now, listen really closely here. To argue for all-comers preliminaries instead of party primaries is not to say we need to get rid of parties. That isn’t necessary at all. We don’t even need to get rid of partisanship.
In these all-comer preliminaries, candidates can run with party labels, as they do in Louisiana, California and Alaska and will continue to in Nevada. Parties can still hold private caucuses and conventions and endorse candidates, and those candidates can show up on ballots touting their party’s endorsement. But there might be other candidates on the ballot, too — having earned their place via signature gathering — from the same party or from the other party or from third parties such as the Libertarians or Greens. And there might be independents who show up without any party label at all.
That’s what’s meant by “all-comers.” And a note on the nomenclature, which I think matters more than we realize: Sometimes what I’m describing is called an “open primary,” but that label is confusing, since some forms of traditional party primaries are described that way. Sometimes it’s given the ugly label “jungle primary.” Sometimes this all-comers method is called a “blanket primary.” That’s better but nonexplanatory. “All-comers preliminary” is best — it tells you what it is in plain language.
Something to remember about party primaries is that they are, essentially, state welfare for political parties in most states, in that the state government — state taxpayers, in other words — pays for a party’s private election. Preliminaries, in contrast, are for all voters. They are genuinely public and therefore represent a reasonable public expense. Public funding for candidates also makes even more sense in this context. And parties will have to up their games to make sure their endorsed candidates are always competing for all the voters at every moment in the process.
An electoral college that too frequently tips the election to the popular-vote loser and gerrymandering that lets the parties dig into invulnerable fortresses erode the legitimacy of our institutions and endanger our political system. We urgently need to address these problems, and a larger House of Representatives would do so. But our elected officials are trapped by their party primaries.
So to get that larger House, we’re going to have to free the politicians.
