Economist Writing Every Day

I have a good friend who is a professor in philosophy at another university. He was telling me about the struggle among his colleagues to determine the recipient of their annual department award. Every year the department chooses from among the graduating philosophy major students one to recognize for excellence. This year, they faced the challenge of incommensurables.

One student had a high GPA in the major, but had a severe case of senioritis and had phoned-in her senior courses. A second had a slightly worse GPA, but had face-planted the senior thesis. Still a 3rd student had merely a good GPA, but wrote an excellent publishable thesis.

The philosophy faculty could not agree. They each shared stories and arguments about the relative weights of the performance indicators and the relative value of the performances. I don’t know if you know any academics, but suffice it to say that they both A) tend not to be good administrators and B) tend not to be invited to productive meetings. I’m glad that I wasn’t in the room.

In fact, the faculty met twice! They were at an impasse. The department award winner is usually no contest. The person who excels in one area tends to also excel in the others. This year, the decision was so unclear and the faculty were so divided that they even seriously considered withholding the award entirely. None of the candidates was excellent on all counts.

Finally, trying to come to a decision – if not an agreement – they decided to adopt something that they’d heard good things about: Ranked Choice Voting. I was thrilled to hear this. What an opportunity to exhibit the nuance and beauty of this collective choice method! They agreed to adopt whatever the outcome would be. As my friend told me this, I was giddy with anticipation. What an exciting story! More good experiences with ranked choice voting may improve its popularity and make widespread its adoption.

If you don’t know, Ranked Choice Voting involves everyone ranking the candidates in order of preference. In this case 1 is most preferred and 3 is least preferred. Then, the candidate with the fewest first-ranked votes is eliminated from the running. The voters whose first preference was nixed now have their votes reallocated to their 2nd preferred candidates. Since only two candidates remain, one of them has won the majority and the election ends with an outcome that is usually considered better than the simple ‘just choose your favorite’ version that most of us use at our local polls.

How did the philosophers fare?

I was stoked that these professional arguers would adopt an algorithm to come to a conclusion. But then… my friend proceeded to tell me about how they assigned points: 3 points for the most preferred, 2 points for the 2nd preferred, etc. Oh No! I could see where this was going. They would sum the points, be upset, and then decry ranked choice voting as a bad system. And that is exactly what happened. They summed the points, didn’t like the outcome, disregarded it, and then continued to argue.

The primary reason that their method was bad is because it leans too much on the relative value of the points. The points imply that the first choice (3 points) is 1.5x more preferred than the 2nd choice (2 points), and 3x more preferred than the last choice (1 point). Furthermore, it implies that we care about the rank order of the voters who get their first choice anyway. Their lower point votes are still counted as if we care about forming an entire aggregated ranking. But the purpose of the election is to choose a winner, not to compare the relative values among losers. So, this points method fails to choose a winner in an acceptable manner and also performs the extra service of ranking losers. If the total points for the winner is ‘wrong’, then you can bet that voters won’t be happy and peaceful once they see the wrong points for their losers.

I’m not sure if my enthusiastic disgust rose to the level of horror, but you get the idea. Ranked choice voting isn’t complicated and it does tend to result in something closer to what people would expect as a consensus from a less structured method. But, by gross misapplication, these eminent scholars who form the intellect of young minds would thereafter inevitably give their students the impression that ranked choice voting is no good.

I told my friend that he’s obligated, as a matter of public service, to inform his fellow philosophers about the error of their ways. There’s always next year.