An Even Better Idea for Reforming Presidential Elections

Research report

ballot

Imagine if there was no possibility of being a “spoiler” in the general election for President. Bernie Sanders could run as an independent, as he has for most of his career. Donald Trump also may have run as an independent, allowing the Republicans to choose a candidate more acceptable to the majority of the party.

Then in November you could choose from Sanders, Clinton, Trump, a Republican Party candidate and third party candidates, free of worry about “spoilers,” “wasting your vote,” voting for the “lesser of two evils,” or any of the other problems that saddle us every four years with only two realistic choices.

Two choices. That’s just one more than you would have had in the Soviet Union or some dictatorship somewhere. Shouldn’t we do better than that?

Last month I offered a proposal to use ranked choice voting in the general election for the President, greatly reducing the possibility that any candidate would become a spoiler. Since then I’ve had an even better idea. Using the Bucklin system, an alternative form of ranked choice voting, would address the spoiler problem, is easy to understand, and is feasible with current voting technology.

Replacing the plurality rule

As most readers probably know, the President is not directly elected by the voters but is chosen by the electoral college. The voters vote in November and, in all but two of the states, whoever gets the most votes in each state gets all of the electors from that state. The electors from all fifty states then vote to determine the ultimate winner.

The electoral college is mandated by the U.S. Constitution and so can’t easily be abolished. But each state can change by the way the popular vote decides which candidate gets the electors from that state.

Like most elections in the U.S., the current system of deciding the winner of the popular vote for President in each state relies on the plurality rule. This is the “whoever gets the most votes wins” feature. The advantage of an election using the plurality rule is that the process is easy to understand and to administer.

The main disadvantage is that it creates the conditions for the spoiler problem. When two or more candidates appeal to the same set of voters, they risk splitting their votes between them thus allowing another candidate to win with a minority of votes. The spoiler problem discourages some candidates from running and voters from supporting a candidate they otherwise would prefer.

The Bucklin voting system

An alternative to the plurality rule is the majority rule, under which a candidate needs more than half the votes to win. The problem with the majority rule is that if there are more than two candidates there might be no majority winner.

One way to solve this is with ranked choice voting (RCV). Instead of choosing a single candidate to vote for, each voter is allowed to rank the candidates in order of preference: my first choice is this candidate, my second choice is that candidate, and so on. If any candidate gets a majority of the first choice votes, that candidate wins. If not, then the other rankings are factored in to determine the winner.

Different versions of RCV do this factoring differently. The most widely used version of RCV is not currently feasible for the Presidential election due to limitations of voting technology. However, the Bucklin system of RCV is feasible.

Under the Bucklin system, if no candidate gets a majority of the first choice votes, the second choice votes are added to the first choice votes. If a candidate then has a majority, that candidate wins. If not, the third choice votes are added in. The process is repeated until either there is a majority winner or there are no more rankings to be added, in which case whichever candidate has the most votes wins.

Because no candidate can win with less than a majority of first choice votes, vote splitting is not an issue. The method thus would diminish the spoiler dynamic, give voters more leeway in expressing their preferences, and give independent and third party candidates a chance to receive serious consideration from voters.

Criticism of the Bucklin system

The most common criticism of the Bucklin system is that it results in “bullet” voting, where voters only use the first choice option and do not vote for second or third choices. The idea is that voters want to avoid helping someone other than their most preferred candidate, while hoping the other voters help their candidate in the further rankings.

But there is evidence to the contrary. The Bucklin system was named after James W. Bucklin, who promoted the method in the Progressive era of the early 20th century.  A 1915 analysis of cities where the system was adopted found that people did use the second and third choice options, and in rising numbers with each succeeding election, under certain conditions.

The conditions were that there was an educational campaign to inform voters of how the system works, there were more than just first and second choice options offered, and there were more than two candidates who attracted voter interest.

As the Presidential race is a highly visible election, a change to Bucklin voting is bound to get a lot of publicity and draw out more than the usual two viable candidates. My considered opinion is that it would function like this:

  • voters with a strong preference for a major party candidate (the Democrat or the Republican) will rank their candidate first and many will refrain from using the other rankings;
  • voters with a strong preference for a competitive non-major party candidate will split between those who rank this candidate first without using the other options and those who use their second choice for a major party candidate in case their candidate does not win the first choice;
  • voters with a strong preference for a candidate who does not seem very competitive (such as a minor party candidate) will vote for this candidate as their first choice and use their other choices to vote for other candidates; and
  • voters who like more than one candidate will use their choices to rank order their candidate preferences.

Conclusion

The end result would be a greater number of viable candidates to choose from, the injection of a wider range of ideas in the election campaign, voters being able to vote for their true first preference without worrying about spoilers, and a winner with broad support. And because voters could turn to other candidates if the major party candidates do not seem responsive to their concerns, the influence of money in the outcome of the election would also be diminished.

This article is a first pass at presenting this idea, and I invite your feedback. I will be publishing another article in the near future examining aspects of the proposal in more detail.

–Alan F. Zundel

Support the SEJ Review

Facebooktwittergoogle_plusredditpinterestlinkedinmail
Understanding This Political Moment
Robots Want Your Job

Leave a Reply

Your email address will not be published / Required fields are marked *