Those who pay close attention to the Democratic presidential horse race are salivating over recent polls showing that Sen. Elizabeth Warren (Mass.) is “surging” and has now hit 20 percent support, with Biden holding a slim lead at 21 percent. But the real news is actually the opposite: the current leader in the polls is not the favored candidate of 79 percent of the Democratic electorate: that is, 4 in 5 voters prefer someone other than the front-runner.
That should be a big problem for those who care about majority rule, as most of us do. Indeed, there is something deeply important in our Constitutional structure about the 50 percent threshold. It starts at the top: to be elected president, a candidate must receive not just more electoral votes than anyone else, but must receive over 50 percent of all available electoral votes. A majority, that is.
Indeed, law professor Ned Foley has shown that those Framers who tweaked the Constitution in 1803 by writing the 12th Amendment wanted “presidents to be elected according to the principle of majority rule.” This was a no-brainer for them: majority rule is a great way to ensure that leaders have sufficient support from the public to be effective. Following this principle, many of the first statewide elections for presidential electors required electors to receive not the most votes but actually reach 50 percent—and if they didn’t reach that threshold, there would be a runoff between the top two finishers to ensure that the winner crossed that magical number.
In the two centuries since, though, we’ve lost sight of the magic of 50 percent, which signals that at least half the voters are at minimum comfortable with a result. Instead, the winner of most elections is now decided by picking the top vote-getter off the list and saying that person’s the winner, even if the winner might actually be opposed by 60, 70, even, as the Democratic race stands, 79 percent of voters.
There is a wonderful and simple solution, though, called Ranked-Choice Voting, which lets voters rank candidates in order of preference. This allows voters to share the information required to ensure that the end-winner will have the support of at least 50 percent of voters.
To see how useful implementing ranked-choice voting would be, we wanted to understand how bad the no-majority problem really is and whether it affects one party more than another. We started by looking at primary and general elections for president and U.S. Senate since 1992. The results stunned us with just how common this problem is—throughout America, and for both political parties.
Start with presidential primary elections, which were very troubling for those who care about majority rule. More than half the winners of contested primaries since 1992 won with less than 50 percent of the vote. In some states, it almost always happens. For instance, the winner of the New Hampshire primary for at least one party has received less than 50 percent of the vote in every election since 1992, with only one exception. Seven other states have had this occur at least five times in that period. And some winners weren’t even close to 50 percent—in many elections, especially during the 2016, 2012, and 2008 Republican cycles and the 1992 Democratic cycle, winners barely cracked 30 percent support in many early states. Expect the same results this time on the Democratic side given how many candidates are running.
Things don’t get much better for presidential general elections, and, like in primaries, both parties have suffered. In 2016, neither Trump nor Clinton received a majority of votes nationwide, and the pattern held in most key swing states, like Florida, Pennsylvania, and Wisconsin, where neither candidate got to 50 percent. The same thing happened in 2000 and 1996, and the deviation from majority rule was especially bad in 1992, when Ross Perot ran a strong third-party candidacy. That year, Bill Clinton won with only 43 percent of the vote nationwide, and he crossed the 50 percent mark in only one single state: Arkansas, where he was previously the governor. Did a majority of the country actually prefer a Clinton presidency, or would a better gauge of the majority’s preferences have resulted in the election of George H.W. Bush or even Ross Perot? We’ll never know.
Things get somewhat better in the Senate, at least for the general election, but not by much. Since 1992, 49 senators from 27 states have been elected with less than 50 percent support. And Senate primaries are a majority-rule nightmare. Every cycle since 1994 has seen multiple nominees of both parties win with support way below 50 percent. Indeed, occasionally the primary winner receives as little as 28 or 29 percent of the vote. It’s no surprise that these deeply split primaries often result in weak candidates who lose in general elections. Candidates who don’t get any votes from 70 percent of their own party face an uphill battle in the general election.
Fortunately, there is an easy and relatively inexpensive solution to this problem called ranked-choice voting. Ranked-choice voting lets election officials run “instant run-off elections” so that if a voter’s first choice candidate doesn’t get a lot of support, the voter’s support goes to the second choice rather than being tossed out. In this way, all ranked-choice elections result in winners with at least 50 percent support.
This method is really catching on. It’ll be used in six Democratic presidential primaries or caucuses this year; it’s used for most elections in Maine; and it’s in use in a growing number of cities, including potentially New York if a ballot measure passes this fall.
Ranked-choice voting, then, is that rare thing in American politics: a modern fix that will not only benefit both parties, but will also enable us to fulfill the vision of majoritarian elections that our Founders had for us. And it’s relatively cheap and easy to do. Now there’s a solution that should get way over 50 percent support.
Jason Harrow is executive director and chief counsel of EqualCitizen.US.Victor Shi is a fellow at EqualCitizens.US.