Headlines: Party primaries are a slow motion train wreck
Candidates winning with low percentages in low-turnout primaries undermine the "will of the majority"
We still don’t know whether David McCormick or Mehmet Oz will be the Republican nominee for the U.S. Senate in Pennsylvania. The primary battle remains too close to call, with thousands of mail ballots left to be tallied and a mandated statewide recount still to come.
But we do know this: Whether Oz or McCormick advances to the general election, either will have “won” with less than a third of the vote.
That means more than two-thirds of Pennsylvania Republicans will have preferred someone else. A bitterly contested primary won’t have pulled together a majority around a nominee. And if enmity continues over mail-in ballots, and the nominee’s victory becomes disputed and controversial, intra-party rancor could continue into the fall and affect the nominee’s chances for victory.
This “primary problem” has become one of the primary problems in our politics, visible this month in Democratic and Republican races in Ohio, Oregon, North Carolina and Indiana, among others.
Fortunately, there’s an easy way to avoid it: Ranked choice voting. It works so well that it has been adopted for statewide elections in Maine and Alaska, in big cities like New York and smaller localities across Utah, by Republicans in Virginia to choose statewide candidates like governor, and by several Democratic state parties for 2020 presidential primaries.
Ranked choice voting gives voters the power to rank the field in order. Then, if no candidate is the first choice of 50 percent, an “instant runoff” ensues. The last-place candidates are eliminated and their supporters’ votes automatically rerouted to their second or third-place choices indicated on their ballot. The process continues until someone secures a majority.
This one solution cures multiple issues. It eliminates the need for future runoff elections, which cost more money and tend to have lower voter turnout. It helps build party unity when a majority of voters feel they had a say in selecting the nominee. It also ensures that the candidate with the deepest and widest support – and therefore the best chance of prevailing in November – wins the nomination.
In Ohio, for example, J.D. Vance will be the Republican nominee for U.S. Senate after winning just 32 percent of the vote in a multi-candidate race. In other words, he’s the party’s pick even though some 68 percent of Republicans preferred a different candidate. Instead of celebrating a win, now Vance is spending valuable time trying to reunite Republicans back together rather than focusing on Democratic opponent Tim Ryan.
On an Ohio talk radio station, he admitted that “there are a lot of Republican voters with hurt feelings about the primary results.” Asked how he would reach those conservatives, Vance said he’d begin by saying “I might not have been your favorite candidate…” If Ohio had used ranked choice voting, Vance would have been talking to Ohio Republicans that way all along. He would have been looking for their second-place votes, knowing that they would be necessary to clear 50 percent in a crowded field.
Ranked choice voting also would have also prevented another spectacle in Pennsylvania this month, where multiple GOP gubernatorial candidates either dropped out of the race at the last minute (though their names remained on the ballot), or came under intense pressure from the party and major donors to end their campaigns after early voting had already begun. The state Republican establishment – after several candidates jumped into the race – finally feared the large field would lead to a plurality win for the most extreme candidate in the race, state senator Doug Mastriano. That’s exactly what happened. An open race in a swing state could now favor the Democrats, as Republicans fear moderate voters will likely lean away from Mastriano.
This failed attempt to consolidate around a single candidate may not have been any more successful with RCV. Maybe Mastriano would have won anyway. Either way, the outcome would have benefited everyone. Voters who wanted someone other than Mastriano wouldn’t have had to figure out who represented the most viable alternative. All the candidates who wanted to run could have stayed in the race until the end. And if Mastriano crossed 50 percent, he’d have a powerful argument to shut down naysayers after his win, that, indeed, the Republican electorate did believe he was electable.
This isn’t just guesswork. Virginia Republicans used ranked choice voting last spring to select their gubernatorial nominee. There were many similarities to the Ohio and Pennsylvania races: A deep and crowded field, enmity and mistrust between many camps, controversial candidates who might have won a plurality without being the strongest candidate statewide. Virginia politicos not only credit RCV for determining winner Glenn Youngkin had the deepest and broadest support, but believe that it helped the party unite faster to win an otherwise blue-trending state last November.
Ranked choice voting would have led to a fairer and more meaningful result in congressional primaries nationwide as well.
This week in North Carolina, five candidates for the U.S. House moved onto the fall elections despite winning just somewhere in the low 30s. Two additional candidates won with support in the 40s. In Pennsylvania’s 12th, Summer Lee won the Democratic primary by just over 800 votes and a non-majority 41.8 percent of the vote.
Oftentimes, these primaries are low-turnout affairs, which means that when a candidate wins less than 50 percent of the vote, in a district drawn to be a safe seat for one party in the fall, only a handful of voters are making a decision for everyone else.
Each U.S. House seat represents about 747,000 people. But a bright-red congressional district in Indiana with nine GOP candidates was won with just 21,000 votes. In Ohio, where seven Republicans vied for the 13th congressional seat, the nomination went to a candidate with 28.6 percent and barely 16,000 votes. That’s just over 2 percent of the district selecting a member of Congress for the other three-quarters of a million people.
Some electoral reforms are controversial, but ranked choice voting is good for everyone – for voters and for parties, for better campaigns and better government, for more representative results in primaries and general elections. So yes, we have a primary problem – but we can fix it by giving everyone a second choice.
David Daley, @davedaley3
It's not clear to me what the actual recommendation is here. RCV in party primaries will solve the minority winner problem in states that don't have run offs, and maybe modestly improve turnout in states that do. But with RCV there is no need for a party primary at all. Having one single election for each race would dramatically improve turnout in the determinant election.
Turnout in primaries has not improved in Maine (RCV), California (Top 2), or Washington (Top 2), to my knowledge.
But if we are to have fair choice in elections we need the same rules for ballot access and filing deadlines regardless of the party name--something that is not found in any US state at the moment (but is the norm in virtually all other democracies).
"In Ohio, for example, J.D. Vance will be the Republican nominee for U.S. Senate after winning just 32 percent of the vote in a multi-candidate race. In other words, he’s the party’s pick even though some 68 percent of Republicans preferred a different candidate."
The goal of a partisan primary should be to select the candidate with the broadest support. Such candidate will have the best chance of defeating other party's candidates. RCV accomplishes the goal.