One of the more recent trends of modern politics has been calling to have the First Past the Post (FPTP) electoral system abolished. If you happened to see my previous blog post, (thank you for reading), then you’ll know that I stress the importance of third party candidates to the health of a democracy. In a FPTP voting system though, third parties not only lose the vast majority of the elections they enter, but they assure that their voters will likely get a less desirable elected official in the long run due to the spoiler effect. In addition, FPTP voting schemes inevitably result in two-party systems which more easily leads to an “us vs. them” mentality—a hallmark of polarization. The ensuing parties also rarely capture the true interests of the majority of the constituency and are instead made up of a number of different voters who often vote for candidates they simply disagree with the least.
The United States uses an FPTP system to select its president[1] and is hardly alone in this. Many countries have retained this system despite the criticisms against it. It may seem that the downsides of FPTP are obvious, but changing the system is often very difficult. Countries’ methods of electing federal officials are almost always codified in a national Constitution which are, for good reason, difficult to amend. For example, Canada, which retains a FPTP voting system for its Members of Parliament, has five methods of Constitutional amendment, but the most realistic is the “750” method where both the upper and lower legislative chambers and the legislative assemblies for at least two thirds of the provinces (assuming they represent at least 50% of the total Canadian population) must approve of an amendment before it takes effect. Taiwan, which holds FPTP presidential elections, requires that at least three quarters of three quarters of all legislators approve an amendment to their Constitution. The United States has a similarly arduous process where two-thirds of both houses of legislature or two-thirds of all state legislatures have to approve of a proposed amendment before it is ratified.
But let’s say, hypothetically, that popular support makes changing electoral systems a feasible option. What alternatives are present[2]? One of the most popular is the ranked choice voting system. Such a system seems perfect at face value. In theory, it eliminates the dangers of the spoiler effect and maximizes the happiness each voter has with the results of the election. However, there are some circumstances where public preference may be distorted by this system. I’ll illustrate my thinking with an example. Suppose five candidates (A through E) are running an in an election for national executive of their country. For simplicity, let’s say that there are 25 voters in this small country with each voter ranking each of the candidates in order of their preference and the resulting preferences looking like this:
Here are the totals for each party:
And here are the election results:
In the first round of the example of the ranked choice voting system, no candidate won a majority of the vote for their party. According to the rules of ranked choice, the candidate with the least amount of votes, which is D in this case, is cut from the race and all voters whose first choice was candidate D get their second choices added to the voting pools of the remaining competitors. In this case, all voters who chose D as their first choice chose E as their second. At the end of the second round, E still does not have enough votes to obtain a majority, so, in the third round, the candidate with the next lowest votes (B) is cut from the race. Three of the voters whose first choice was B had E as their second choice and the remaining voter had C as their second choice. The results are added in the third round accordingly with E being declared the victor.
This seems like ranked choice voting working at its best, but if you take a moment to look back at what each voter voted for, you’ll see that, while a little over half of the voters chose E for either their first or second choice, a little less than half chose E as their last choice. Conversely, while C received only five first-choice votes, they received fourteen second-choice votes and six third-choice votes with no voters choosing them as a fourth or fifth choice. Putting it in a different way, let’s say that every first choice vote that a candidate received awarded them five points and every second choice four points and every third choice three and every fourth choice two and every fifth choice one. In this case, B would actually be the candidate in last place with only 65 points. D would be second to last with 66 points. A would be third to last with 72 points with E barely beating them with 73 points. C would be the clear victor with a total of 99 points.
We have arrived at what I see is the essential failing of the ranked choice voting system. It Is often lauded as being the system that most closely aligns with voter preferences and creates a multi-party system, but this example shows that it’s possible that a slim majority’s first choice candidate can beat an overwhelming majority’s second choice candidate. It is true that a majority has still chosen candidate E in this example, but, and this is up to personal opinion, is the ranked choice system really reflective of voter preferences?
We often discuss democratic erosion as something that’s happening now. It’s easy to fall into the trap of thinking that the framers of the world’s strongest democracies created balanced systems, but is this truly the case? Perhaps the presence of a flawed electoral system makes some degree of democratic erosion inevitable. I mentioned above that two-party systems and ensuring polarization are the inevitable end state for FPTP systems. Is it possible that detrimental consequences, unforeseeable at present, are inevitable for alternative voter systems? I won’t deny that institutional corruption, polarization, disinformation, populism and so many other forces play their part in democratic erosion. However, I also think that looking critically at a country’s basic institutional framework could reveal some weaknesses that need to be addressed before they create unmanageable problems.
[1] The United States uses a majoritarian system to select its president, but I still see it as a de facto FPTP system because there were only three instances where the House of Representatives served as a tie breaker and they all occurred in the 1800s and dealt with popular third party candidates. Now, the general electorate seems aware of the spoiler effect these candidates can have and have adjusted their voting behavior accordingly. The U.S. is, for all practical purposes, still in a FPTP system.
[2] I’m suggesting alternatives here for elections of national executive leaders. For reasons I won’t get into here, I support a proportional system of representation when it can be applied (as it can for legislatures). However, with races that can only produce a single winner, such a model, unfortunately, cannot be.
Amber Germain
Hi Matthew, your post follows sound logic. The main argument is that First Past the Post systems of voting are inherently flawed, but ranked choice voting may not end up being better. According to your sources, 45 other countries use this system for picking their executive seat including Canada. But of these 45, Canada seems to be one of the few on the list that is not currently experiencing some form of democratic erosion or a status not free (according to Freedom House). I agree that Public preferences may actually be distorted based on ranked choice voting and this is illustrated well by your easy-to-follow example of a country of 25 people choosing National Executive candidates A-E. In your example candidate E won the majority, but as a first choice E had 7 votes while C had five. As a second choice, though, E had 6 votes while C had 14. Based on the rules of ranked choice, because C did not have a majority in the first round, the votes for Candidates B and D went to Candidate E, even though most people ranked C their second choice. The argument is that this method actually does not reflect the will of the people. Math is not my strongest subject and I would be interested to know if the fairness evens out as the population size gets higher. For example, if there is a population of 25 with 5 candidates running maybe ranked choice is less fair than say a population of 1000 with 5 candidates running… I don’t know. Your argument may be strengthened by either citations on the mathematical theory you are using to prove the rank choice less effective than some scholars perceive, or to use a political science article that defends concepts like that of Levitsky and Ziblatt. They write, “one of the great ironies of how democracies die is that the very defense of democracy is often used as a pretext for its subversion. The concept that this supposedly fairer way to vote could as a means of actually undermining voting fairness is good food for thought. Your article is concise and engaging.
Christian Carlow
Hi, Matthew. As someone who is greatly interested in electoral reform, I really enjoyed reading your article. However, I disagree with your main argument that ranked choice voting is not necessarily better and more democratic than first past the post. In a previous post, you emphasize the importance of third parties to a representative democracy. If third party participation and viability is that important of a measure of democracy, ranked choice voting is no doubt better for democracy than the first past the post system. In a first past the post system, candidate E still would have won. But candidate C, who you argue should receive the most seats as they have the highest “score,” would have been completely excluded. And maybe E in your example did not have the highest “score,” but the fact is that they received the most first place votes of anyone else. Winning matters: they should be rewarded accordingly. So, while I agree that it’s important to look at the harm institutional frameworks can do to a democracy—as you point out with first past the post voting—I disagree that ranked choice voting is not clearly more democratic than first past the post.