Sunday, May 9, 2021

Does Ranked-Choice Voting Really Work?

All else equal, it would be good to reduce divisive partisanship in politics and to get elected officials who are more generally representative of their constituents' views. At least it seems like that would be good. Upon typing it I immediately have my doubts that I can both defend those goals as meaningful and desirable. And even if I can, it might be ridiculous on its face with the "all else equal" qualifier being impossible.

After all, all else equal, I would love to be worth $100 billion.

Regardless, let's briefly explore one possible method of improving elections, ranked-choice voting (RCV); aka, instant-runoff voting. From Ballotpedia ranked-choice voting is
an electoral system in which voters rank candidates by preference on their ballots. If a candidate wins a majority of first-preference votes, he or she is declared the winner. If no candidate wins a majority of first-preference votes, the candidate with the fewest first-preference votes is eliminated. First-preference votes cast for the failed candidate are eliminated, lifting the second-preference choices indicated on those ballots. A new tally is conducted to determine whether any candidate has won a majority of the adjusted votes. The process is repeated until a candidate wins an outright majority.
Let me state up front that I very much like this as something that might help break the two-party duopoly that corruptly prevails in America today. This is a goal that isn't necessarily in line with the reasoning that RCV could reduce divisiveness, etc., but I think it is consistent with those goals. 

Yet a simple but extreme thought experiment gives me some doubts about RCV as a cure for the supposed ails. 

Consider this election ballot: 
  • Jesus
  • Almost The Devil
  • A Goofball
  • The Devil
Now suppose that we hold the election and the first-place results are:
  • Jesus - 38%
  • Almost The Devil - 30%
  • A Goofball - 20%
  • The Devil - 12%

Under RCV we don't yet have a winner because no candidate has a majority. So, we eliminate the lowest first-place vote getter and give his second-place votes to the remaining candidates. Presumably in this hypothetical everyone who voted for The Devil, who has now been eliminated, put Almost The Devil as second place. Therefore, the new results are:

  • Almost The Devil - 42%
  • Jesus - 38%
  • A Goofball - 20%
We still don't have a majority vote getter; so we now eliminate A Goofball whose voters equally split their votes for Jesus and Almost the Devil as second place. Therefore, the new results are:
  • Almost The Devil - 52%
  • Jesus - 48%
My hypothetical has resulted in Almost The Devil defeating Jesus. That seems bad on its face. Additionally it probably does not satisfy the Condorcet criterion (a known short coming of RCV among other systems). This is just one way RCV might not live up to our dreams. The Volokh Conspiracy at Reason explores another, similar dampening of expectations for what RCV can achieve in heavily contested elections. 

My hypothetical concern here might be rejected for at least one of two reasons (or both): 
  1. In repeated experience voters should get better and better at using the new system. This is similar to rejecting the argument that a simple one-time experiment in a classroom "proving" prisoner dilemma problems mean people will fail at coordination ... therefore, government force is required to make them make the right choices. 
  2. Don't let the perfect be the enemy of the good. If this is a 90% solution to known problems with elections and democracy, it might be worth this hypothetical, unlikely risk.
Perhaps mine is an uncharitable view on the electorate in general. The thing is when I apply the labels (Jesus, The Devil, etc.) so as to be self evident about the hypothetical candidates you get an unfair look into the future that is not visible to the voters ex ante. Reasonable minds will be rationally ignorant about specific candidates, and no amount of homework done ahead of time will tell us the future with certainty. 

Perhaps it is an argument against democracy--one could charitably say it is simply an argument in favor of less democracy. I think it is certainly an argument for less government power. Electing Almost The Devil as Advisement-Only Czar to the Local Private Firm Dog Catchers minimizes his tyrannical reach. 

I still think RCV is a big step in the direction of improvement. But this thought experiment has given me some moderation in my expectations of what it can achieve. 


P.S. See also this podcast from Building Tomorrow for an overview on RCV and its potential implications. 



No comments:

Post a Comment