Does RCV Actually Make the Difference?

Rep. Jared Golden (D-ME)

This past week, Gov. Janet Mills (D-ME) made history, though you'd be forgiven for not knowing so in the midst of national conversations about imaginary hurricanes in Alabama & a (possibly also imaginary) cancelled Taliban meeting at Camp David days before 9/11.  Mills declined to sign a law (thus making it a law...politics in America is weird) that essentially made Ranked-Choice Voting legal in the Pine Tree State for next year's presidential election.  As a result, Maine will become the first state ever to employ ranked-choice voting in a presidential election.  This is a triumph, in my opinion, for democratic choice and a step in the right direction in helping to enfranchise more ideas in American politics, but I also was curious-how often does ranked-choice voting actually come up as an issue, and while we don't know if Maine's decision to do this will make a difference in 2020 (since we don't know how the other 49 states will vote), is there precedent for it potentially deciding a presidential election.

Before we get further, I want to point out exactly what Ranked-Choice Voting (RCV) is, and why it's an issue.  We frequently hear in presidential elections when it comes to third-party candidates "don't waste your vote"-it's a well-understood maxim, since in America there hasn't been a plausible third-party candidate who could since 1992, and a third-party candidate hasn't actually won a state's electoral votes since 1968.  Therefore, a vote for a third party candidate is considered a wasted vote since you aren't voting for a candidate who could plausibly win.  However, with ranked-choice voting, you don't run into this issue.  Let's say it's 2016, and you really don't like Hillary Clinton but you REALLY don't like Donald Trump.  Instead, you cast your vote for Jill Stein. With most states, you'd simply be sending a protest signal to both major party candidates, but ultimately not be voting for someone who has a chance at winning.  With Ranked-Choice Voting, you're actually ranking the candidates, so you'd vote for Stein as your first choice, and then Clinton as your second choice, so in case we're in a situation where your vote might make the difference, you have voted for one of the final two candidates.  Thus, you can vote for your preferred candidate (Stein) without having to worry about spoiling the race by throwing a vote to your least favored candidate (Trump), costing your second favorite candidate (Clinton) the race.  RCV allows you more leeway with your voting without punishing you for going outside the two-party system.  It's also better than Instant-Runoff Voting (where the top two challengers go to a runoff, similar to systems in Louisiana and Georgia), because people don't have to vote twice (frequently IRV runoffs will see a drop-off in voter participation).

I stated above "where your vote might make the difference" and with RCV that's when one candidate doesn't get to 50% of the vote.  Obviously if one candidate is 50%+ voters' choice for first place, he or she is the deserved winner regardless of who the second or third choices of the remaining candidates are, since they already got the majority.  It's in cases where no candidate can get the majority of the vote where RCV would need to come into effect.

Eliot Cutler (I-ME)

This has been a particular problem in Maine, where they actually started doing RCV in 2018 in congressional races.  Maine has a pretty robust third-party tradition (they are one of only two states to have an Independent Party senator currently), which has caused problem for moderate and left-leaning voters in recent years.  In 2010 and 2014, Gov. Paul LePage won both of the elections with less than 50% of the vote, but thanks to the Democratic candidate and a left-of-middle Independent candidate (Eliot Cutler) splitting the vote, a candidate whom it's probable the majority of blue-tinted Maine didn't support was able to hold the governor's office for eight years; it's probable that in both 2010 and 2014 that either the Democrat or Cutler would have won the governor's office had RCV been in effect.  Maine actually had a major race go to the RCV when the second congressional seat race ended up not having a majority for either incumbent-Rep. Bruce Poliquin (R) or State Rep. Jared Golden (D), as two minor candidates deprived them of a majority.  While Poliquin would have won under the old system, having won 2,000 more votes than Golden initially, Golden overwhelmingly was the second or third choice of the independent candidates who were dropped off for too-little support in the RCV, and thus he ended up winning the race.

But Maine is unusual-not all states have such a robust third party operation, and so I was curious to see when, especially in federal elections, RCV makes the difference.  It turns out most races do, in fact, have a candidate who hits the majority.  While there are some anomalies (Poliquin might not have made it to the House in the first place had this law been in effect in 2014 since he didn't get 50% in that race either), with Will Hurd in particular standing out for never having won 50% of the vote in his congressional career, no House election has ever featured enough contests that would have gone to RCV to make the difference for the majority in the past seven cycles.  There were likely instances where the results could have changed, but not enough that who the Speaker would have been would have changed.

The same cannot be said for the Senate or the White House, though.  Since 2006, 24 Senate races were decided by less than a majority of the voters, ranging from races you'd assume were in such a situation (Franken vs. Coleman in 2008) to races you might not remember (Jeff Flake was much closer to losing in 2012 than any of us give credit).  Senators like Michael Bennet & Lisa Murkowski have never, in fact, won a majority of their state's votes.  While cumulatively runoffs would have clearly made the difference (since senators run for six-year terms, eventually you alter enough races you get to the point where RCV always matters), if you look at just a specific cycle, both 2006 and 2016 would have made the difference had the Democrats won enough seats.  In 2006, where the majority was decided by one vote, both Democrats Claire McCaskill (MO) and Jim Webb (VA) won with less than 50% of the vote (though both probably could have taken enough second places from liberal-third party candidates to still win...it's impossible to know for sure), while in 2016 when the Democrats needed 2-3 seats, Republican Senators Lisa Murkowski, Roy Blunt, & Pat Toomey all failed to get 50% of the vote.  2016 becomes particularly interesting when you remember the Merrick Garland of it all.  While Donald Trump won in 2016, the lame duck session means that there are 17 days before the election, the new Congress is seated.  It's entirely possible that if Democrats had won, say, Missouri & Pennsylvania in runoffs that Chuck Schumer (with Vice President Biden casting the tie-breaking vote to give the Senate the judicial gavels) might have been able to push Merrick Garland through the Senate during that time frame before Mike Pence would have handed over the Senate to the Republicans.

Though, it's worth noting that if RCV was in effect, Mike Pence probably never would have been the Vice President.  That's because for the White House, we've had four major recent elections all of which would have gone to a runoff in a number of states.  While it was safe to assume that Bill Clinton's two elections could be altered by RCV (with Ross Perot in the mix as a pretty significant third-party candidate), 2000 and 2016 both would have gone into overtime enough so that the candidate who received 270 electoral votes would have had to receive it via RCV in select states.  In 2016, Trump received less than 50% in states totaling 104 electoral votes (ranging from states that you'd expect like Michigan & Wisconsin, but also including Utah, which disproportionately went for Evan McMullin, holding Trump to his lowest winning margin).  While Hillary Clinton also had states like Maine, Virginia, & Minnesota that would have also gone to second-and-third options for third party candidates, it's entirely possible that Clinton would have been able to pull off the White House thanks to second or third choice candidates in 2016 (putting to test the #NeverTrump theory).  And it certainly would have impacted 2000, when many people assume that the third party votes given to Ralph Nader (where it's largely assumed most of those voters would have gone for Al Gore otherwise) would have given Gore Florida, a state where George Bush failed to gain 50% of the vote.  The fact that four of the last seven elections would be decided by RCV second choice voters makes Maine's decision intriguing, and potentially a template the rest of the country should consider for the future.