That primary reform initiative

Several states are looking to push parties further out of elections; here's why I think that's the wrong call

Recent debate at the University of Denver over Prop. 131, featuring Kent Thiry, Nick Troiano, Jason Lupo, and Candice Stutzriem

Several states, including my state of Colorado, are considering some sort of primary election reform on this year’s ballot. This is an important topic, as reforms like this tend to pass pretty comfortably, and they’re becoming more popular across the country. I’d like to focus on the reform in Colorado — an all-party top-four first round election followed by a ranked-choice runoff in November — but recognizing that the lessons apply elsewhere.

To get something out of the way first: I’m not a big fan of this kind of reform. I make a point of not telling readers here how to vote, and my guess is that this initiative is going to pass in Colorado anyway. But I’ll outline my thoughts on this.

Right now, Colorado has a primary system generally described as “semi-closed.” Democrats can only vote in the Democratic primary, and Republicans can only vote in the Republican primary, but unaffiliated voters are given the choice of whether to participate in the Republican or Democratic primary (but not both).

Under this proposed reform (Proposition 131), the state would switch to a “top-four” system, in which the “primary” election would become a contest in which all candidates for the same office, regardless of party, would compete directly against each other, and the top four vote winners would advance to a runoff election in November. So the runoff election could consist of three Democrats and a Republican, just four Republicans, two Democrats and two Greens, etc.

The November runoff election would be conducted under ranked-choice voting (RCV), where voters would rank the four candidates in order of preference, the least popular first-round candidate would be eliminated and their second-choice votes redistributed, etc., until there’s just one candidate remaining for the contest. This set of rules is similar to what Alaska uses now, and what several other states, including Idaho and Arizona, are currently considering.

One of the main arguments for adopting such a reform is that it would, theoretically, disempower parties and allow for the election of more moderate officeholders. We actually have some evidence on this sort of thing, and the track record is mixed. California adopted a top-two runoff system — not exactly the same thing but similar in spirit — more than a decade ago, and it seems to have reduced polarization in the statehouse slightly, which is no small feat in what was the most polarized legislature in the country. The main effect is when there are same-party runoff elections, and more moderate candidates tend to win those. But some other states, including Washington, have similar top-two systems without any detected effect of reduced polarization.

The ranked-choice voting aspect of Colorado’s runoff proposal is another interesting tweak. Again, the argument for this is that it can help reduce polarization, and yes, it can do that in some cases, but it can also increase to polarization in others. It can also sometimes produce unusual outcomes when one party is better at coordinating than another, such as Democrat Mary Peltola winning in comfortably-red Alaska’s 2022 House election. As Lee Drutman wrote here, RCV is especially helpful for picking multiple officeholders from a crowded field, such as in at-large city council elections. Most of the time in a single-winner election, it doesn’t produce anything different from a simple plurality rule, and once in a while it produces something weird. RCV also asks a lot from voters — ranking candidate choices is harder than picking a D or an R, and some people respond to the demand for more work by simply declining to vote.

Also, for those seeking to disempower parties, sometimes reforms like this produce surprises. One of the unanticipated effects of California’s top-two reform was that the parties started making pre-election endorsements of candidates, and those had a strong effect in picking winners. As I write in my book The Inevitable Party, the shelf life on party reforms tends to be pretty short, since parties often adapt. That’s part of the reason that even if this initiative passes, I doubt it will have a significant long term impact on who gets elected in Colorado and who doesn’t.

One of the worst-kept secrets in Colorado politics is that kidney dialysis company DaVita’s CEO Kent Thiry, who has been heavily bankrolling this and other party reform initiatives for years, has been doing all this as a way of changing the state’s election system to make it easier for a wealthy moderate like him to someday become governor. Thiry now says he’s not interested in doing this, and honestly it strikes me as a pretty inefficient way for a wealthy person to become governor anyway. But it’s not clear that this is a reason to oppose the initiative — any reform benefits someone, and there are all sorts of unanticipated consequences to such reforms that make it far from a slam dunk.

But honestly, my main objections to this initiative are normative. I believe that if anyone should be deciding who should and shouldn’t be running for office it should be political parties. Decades of reforms weakening parties have generally made the pool of candidates worse. A stronger Republican Party around, say, 2015 — one capable of screening out an ideologically incoherent billionaire with weak commitments to democratic principles — would have been good for both the party and the country. Reforms like Proposition 131 just weaken parties further and make it a bit easier for factional candidates and wealthy celebrities to win.