Technically Exists

The National Popular Vote Interstate Compact (NPVIC) aims to subvert the Electoral College and elect the president of the U.S. by popular vote. It would accomplish this by having the states that signed on give their electoral votes to whichever candidate won the popular vote. However, the way it defines the popular vote assumes that all states will use single-choice plurality voting in their presidential elections. This is a problem, because Maine has switched its presidential elections to use single-winner ranked choice voting (RCV-1), also known as instant runoff voting.

The NPVIC website doesn’t address this possibility, but promoters have suggested that this problem would be resolved by the state explaining how its votes should be tallied. This is an unsatisfactory answer for a few reasons. First, there are multiple options for how the votes could be added. One option is to ignore the use of RCV-1 and simply count voters’ first preferences. Another option is to have Maine run RCV-1 until all but two candidates are eliminated, then count voters as having cast a plurality vote for whichever of those two candidates they preferred.¹ A third option is to hold a nationwide RCV-1 tally where voters from other states are considered to have cast ballots that ranked only one candidate. Each of these options has advantages that will appeal to some people and disadvantages that will drive others away.

A second reason why this isn’t a good solution is that it may not work if other states adopt other methods. This probably wouldn’t be an issue anytime soon; the most likely alternative to be adopted at the state level is approval voting, and there are options for adapting RCV-1 to work with equal rankings. However, in a scenario where other methods gain traction in the future, this could become very difficult. Trying to add RCV-1 ballots, 3-2-1 ballots, and ranked pairs ballots together while respecting how those voting methods operate would be a nightmare.

The final reason I don’t consider this solution to be enough is that it relies on the states agreeing on a method for adding the ballots together. If different states have different priorities, they may not be able to agree on a method. For instance, Maine may want to stay as true to its chosen voting method as possible, while other states may want a simple method that gives all voters the same options. Even if the states are cooperative with each other and try to find a compromise, the process could still take a lot of time. If it isn’t finished by the time the next presidential election is held, the election results could be severely delayed.

I don’t know what the future holds for the NPVIC. Maybe it will be unable to get enough states to pass it and simply die off. Maybe Maine will repeal RCV-1 and join the other states in sticking with single-choice plurality. Or maybe this problem will need to be dealt with at some point. If that becomes the case, things could become rather unpleasant.