Third in an occasional series on ranked-choice voting.
In discussing the arguments for and against ranked-choice voting, I’ve perhaps emphasized too much the urgency of assuring that the winner of an election have a majority of the votes, for whatever benefit such a result might add to the legitimacy of a winner’s mandate to govern and to whatever extent a majority winner might have to govern on behalf of a broader public.
“Majority rule” sounds like a fairly fundamental premise of democracy, but our American system has many exceptions to it built deeply in (take the bizarre Electoral College system, for example, or the various supermajorities required by the Constitution). There’s also this:
If we were considering whether to adopt RCV for statewide elections, the question of majority rule or plurality rule would be more relevant. For statewide elections (as for U.S. senator, governor, statewide constitutional offices) and for legislative elections, Minnesota uses the first-past-the-post, plurality-wins system that is common in most of the country. And, since Minnesota has a fairly strong third party, many of our recent governors and senators have been elected with a plurality, not a majority of the vote.
But since RCV in Minnesota is, for now, mostly about municipal elections in Minneapolis and St. Paul – and most especially about the hotly-contested Minneapolis mayoral race — it is important to note that the system in use for such elections before RCV also virtually guaranteed that the winner would receive a majority of the votes.
The old system, in case you need a reminder, was this: There would be an open, nonpartisan primary in September. Then the two biggest primary vote-getters would face each other one on one in November. In a two-person runoff, leaving aside matters like spoiled ballots or abstentions, the winner would essentially be guaranteed of having a majority of votes cast in the race.
Turnout a factor
The larger argument over whether RCV is an improvement has more elements than just the issue of majority rules. But I asked Jeanne Massey of FairVote Minnesota (the leading advocate of RCV) why, since Minneapolis already had a system that produced a majority winner, she believes RCV is an improvement.
Massey’s main response was about the difference in voter turnout between a primary and a general election. Historically, in the odd-numbered years when municipal elections are held, the November general election turnout is double or more of the turnout for primaries, Massey said.
Here are the turnout figures, expressed as a percentage of the registered voter population of Minneapolis, in the last four quadrennial elections when the mayor’s office was on the ballot.
1993: General: 46 percent; primary 23 percent
1997: General: 46 percent; primary 15 percent
2001: General: 40 percent; primary 27 percent
2005: General: 30 percent; primary 15 percent.
As you can see, it’s normal for about twice as many voters to turn out for the general election in November as for the primary. And, with the primary (which used to be held in September) moved up to the first week in August, the primary turnouts are expected to fall further.
It’s true that both the old nonpartisan primary system and RCV are both designed to produce a majority winner, Massey said. “The reason ranked-choice voting is such a vast improvement is that no one turns out for the primary,” she said.
People who argue against RCV tend to be political insiders who prefer the old system because they find it easier to manipulate. Allowing a relatively few voters to decide for the rest who will be the final two choices makes it much more of an insider game, Massey said.
Under ranked-choice voting, all the major decisions — who will be the final two candidates and which of them will take office? — are made on the day when turnout is highest, so “many, many more voters are participating in the decisions about who’s going to govern,” Massey said.
Over the course of many conversations about RCV in preparing for this series, I had a good one over lunch with Todd Rapp of Himle Rapp & Co., a PR/consulting firm that does a lot of political work. Rapp is the Democrat of the two name partners.
Rapp isn’t rabidly opposed but seems mildly against RCV — “not a big fan” is how he put it — at least on general election day for a reason that hadn’t occurred to me earlier.
Rapp believes that to really get the best kind of mandate to govern, the process has to end with a binary choice. You have to build in a period of at least two or three months at the end of the process for two candidates to identify and explain and clarify and justify their differences and describe their policies, their goals and their visions in a one-on-one environment where the voters understand that they have to choose one or the other.
The pre-existing system did that. The single round of RCV doesn’t, Rapp says. He said RCV might be used on primary day to winnow the field down to the final two. But he wants “two candidates to face off and duke it out” for those last months.
After I had obtained Massey’s argument about the small turnout for primaries, I ran it by Rapp to see if it changed his feeling. He was respectfully unmoved. Massey certainly has a point, he said. Many people vote on general election day and, under his preferred system, those voters would have only two choices.
That decision not to vote in a primary is, of course, up to those voters. Having the smaller group of more heavily involved voters choose the two finalists isn’t ideal, but he believes that’s what’s necessary to provide a clear choice.
Massey says there are some in RCV circles talking about a two-step process. A primary would be used to winnow the field down to four, then RCV would be used to elect the winner on the final round. But she didn’t sound too excited about it, and said it isn’t currently in use anywhere.
Effective Democracy is a year-long series of occasional reports supported by the Chicago-based Joyce Foundation, as part of a grant made to MinnPost and the Wisconsin Center for Investigative Journalism.