Plurality voting, our usual system, means selecting government officials by who gets the most votes, where voters can choose only one candidate, regardless of how many candidates appear on the ballot. Here is a simple illustration showing why plurality voting is stupid.
Let’s say Jesus, Gandhi, and Hitler are on the ballot, and our polity has only 99 people. In our polity, 34 people are morally corrupt, so they vote for Hitler. While 65 people are morally virtuous, and so would much rather have either Jesus or Gandhi elected than Hitler, they are nearly evenly divided on which of those candidates to support. So Jesus gets 33 votes, Gandhi gets 32 votes, and Hitler wins with 34 votes.
Unless you are insane, you will immediately see that that’s a bad outcome. Yet that is the voting system in use almost everywhere throughout the United States, the one that yields potentially insane outcomes.
Now rerun the election with approval voting, meaning that voters get to vote for as many candidates as they want. Now Hitler loses badly with only 34 votes, while Jesus wins handily with upwards of 65 votes, as most people who vote for Gandhi also vote for Jesus. (Gandhi might win depending on how many crossover votes there are.) In this scenario, a seemingly slight difference in the voting rules means either that Hitler barely wins or that Jesus wins in a landslide.
The least worst
Ah, but isn’t a runoff between the top two vote earners just as good as approval voting? Often it is not, even aside from the extra cost and hassle of the runoff election. Consider a different hypothetical election.
Now nine candidates are on the ballot. One is Hitler, one is Stalin, one is Jesus, one is Gandhi, and the other five are some of the best people you’ve ever met or want to meet. In this polity of 99 people, only 24 people are morally corrupt, and they split their votes evenly between Hitler and Stalin, so each of those candidates gets 12 votes. The other 75 people are morally virtuous, so they would much rather have any of the other 7 candidates elected than Hitler or Stalin. But they spread their support nearly evenly among those 7 candidates, so each of those candidates gets only 10 or 11 votes. In this scenario, the 7 candidates preferred by 75 people all lose, while Hitler and Stalin move on to the runoff election. Again, that is an insane outcome.
That is how Pueblo runs its mayoral elections. CPR reports, “Nine candidates sought the position. None received more than 50 percent of the total votes, which means the top two are headed for a runoff on January 23, 2024.” In this case, the top candidates earned 22.6 and 21 percent of the vote, so they outperformed Hitler and Stalin in our hypothetical election. Maybe that means one of the real candidates actually had broad public support. Or maybe not. Maybe the 56.4% of voters who chose someone else would rather have had any of the other candidates win. We’ll never know, because voters were not allowed to well-express their preferences.
Okay, so what about ranked-choice voting, in which typically voters rank their first, second, and third choice? Surely that’s as good as approval voting, right? Not necessarily, although it’s definitely better than the usual system.
Boulder used ranked-choice voting in its mayoral election, and in this case it seems to have worked pretty well. Let’s look at the real election before considering potential problems with the system. Three candidates, Aaron Brockett, Bob Yates, and Nicole Speer, earned substantial votes during the first round, with Yates taking a commanding lead. But then, with Speer (and a fourth candidate) eliminated and those votes going to the second-choice candidate, Brockett pulled ahead, because most of the people who favored Speer favored Brockett over Yates. (See the Axios article.)
Approval voting probably would have yielded the same outcome in this case. If we can’t have approval voting for every election, I’d at least like ranked-choice voting. Notably, either system solves the problem of minor-party candidates “spoiling” the election for major-party candidates.
Don’t be like Alaska
But, as you might anticipate, there are some scenarios that ranked-choice voting does not handle well. Let us again say we have a polity of 99 people. Candidate A appeals to the bigotries and biases of 35 people, Candidate B appeals to the bigotries and biases of 33 different people, and Candidate C, a basically reasonable person, not only appeals to the other 31 people but is the second-choice candidate for the 68 people. In this scenario, someone who votes for Candidate A would be okay with Candidate C winning but would absolutely hate Candidate B winning.
What’s the outcome? Under plurality voting, Candidate A wins. Bad! Under ranked-choice voting, Candidate C is eliminated in the first round, so either Candidate A or B wins. Bad! But under approval voting, Candidate C wins. Whew!
Something like this problem with ranked-choice voting happened in Alaska last year. Nathan Atkinson and Scott Ganz explain: “Ranked-choice voting makes it more difficult to elect moderate candidates when the electorate is polarized. For example, in a three-person race, the moderate candidate may be preferred to each of the more extreme candidates by a majority of voters. However, voters with far-left and far-right views will rank the candidate in second place rather than in first place. Since ranked-choice voting counts only the number of first-choice votes (among the remaining candidates), the moderate candidate would be eliminated in the first round, leaving one of the extreme candidates to be declared the winner. This is exactly what happened in the special election in Alaska.”
True, no form of voting completely gets rid of strategic voting. But the problem is less severe with approval voting. With plurality and even ranked-choice voting, a voter sometimes has to worry about accidentally helping to elect the worst candidate (from the voter’s perspective). With approval voting, a voter has to worry about helping a second-choice candidate.
I’ve written about approval voting before, and I’m sure I’ll write about it again, as the political class seems surprisingly uninterested in heading off potentially catastrophic election results. They say democracy is the least-bad political system. We can lower the risks of downsides even more by adopting approval voting.
Ari Armstrong writes regularly for Complete Colorado and is the author of books about Ayn Rand, Harry Potter, and classical liberalism. He can be reached at ari at ariarmstrong dot com.
