The National Popular Vote interstate compact (NPV) fails to account for differences among the states. These could lead to Equal Protection violations—where voters are treated differently in the same election—and exacerbate uncertainty and distrust. Even one of NPV’s creators shares these concerns.
A new and related problem comes from ranked choice voting (RCV). This election process allows voters to rank multiple candidates. If no candidate has a majority on the first count, then votes for the least popular first choice are canceled and those voters second choice is counted as their first (if voters fail to rank additional candidates, their ballots are thrown out). This process is repeated until one candidate gets a majority. Maine and Alaska are both set to use RCV to choose their states’ presidential electors in 2024.
The trouble is, adding RCV totals to any other vote totals is not mathematically meaningful. It’s like combining percentages, which likewise produces a result that just doesn’t mean anything. As my colleague, Sean Parnell, has explained, hundreds of thousands of votes (or more) could be erased by states using the RCV process—changing the result of an election under NPV.
In a presidential election, RCV could literally erase hundreds of thousands or even millions of votes for major party candidates if there is a strong third-party or independent candidate (which would be more common under the NPVIC). Any state with RCV where the Democrat or Republican finished in third place would re-allocate votes from the third-place finisher to the top two candidates. In 1992 at least 390,000 votes have been erased in Maine and Utah if RCV had been in place at the time.
Combining percentages is a common math mistake, and now a prominent election group that backs both NPV and RCV is making the same mistake and using it to conceal the conflict between the two proposals. FairVote is an election reform organization on the political left. It has long supported NPV, even sponsoring swanky trips for state legislators to lobby them on the issue.
Because FairVote supports both plans, and has only recently convinced Maine to enact RCV, it is desperate not to admit the conflict. Here is the group’s attempt to explain that everything is okay:
… Maine will use RCV to generate a winner that earns its two at-large electoral votes. When there is no first round winner, Maine will run its final statewide tally down to two presidential candidates and put those final round vote totals on the certificate of ascertainment. That means the vote totals for the top two candidates will be added to the overall tally for the national popular vote. This common-sense approach allows voters to enjoy the benefits of the ranked choice ballot while incorporating the results with the national popular vote tally in a way that reflects the intent of Maine’s RCV law.
Again, this isn’t a political problem—it’s a math problem. Maine voters may or may not want to use RCV, but the process is only legitimate for choosing their own state’s presidential electors. As Sean details (using 1992 as an example), RCV could arbitrarily shift large blocks of votes in the national total based on political conditions in an individual state.
It is no end of ironic that supposed advocates for democracy at FairVote are pushing a perversion of NPV—mixing in state RCV totals—that would eliminate the “one person, one vote” nature of NPV’s majoritarian plan.