Our Elections Are Threatening Our Democracy

Finding an Alternative to the First-Past-the-Post Electoral System

The dominant electoral structure in the United States is the Simple Plurality system, or commonly called First-Past-the-Post (FPTP). Its advantages are in its simplicity: whoever gets the most votes wins the election. (A plurality just means the biggest slice of the pie, so to speak. A majority (50%+1) is a type of plurality.) FPTP is easy to understand and selecting winners is easy as well.

But essentially, that’s where the benefits of the Simple Plurality system end. There are a lot of problems with the system that harm the fairness and integrity of elections all around the world.

  • The winner doesn’t have to get over 50% to win. Hell, they could win with 10% of the vote, as long as each of their opponents got 9.9% or less. In 2015 for instance, the Conservative Party of the UK won a majority of seats in Parliament with only 36.7% of the votes [Source]. In November 2015, the Canadian Liberal Party won a majority of seats in Parliament with just nearly 40% of the vote [Source].
  • But what is more problematic is that there is no opportunity for the voting blocs to compromise or appeal to moderation, because centrist candidates were bumped off long ago in the primary elections and only extreme candidates remained (which is why no candidate gets over 50% in big elections until there’s only 2 left in the race; they never have wide appeal).
  • Many people feel that their votes are insignificant or wasted in FPTP elections, because many votes in these elections do not change the election results (ex. if Candidate A already won a majority of the votes, any additional votes would be “wasted” basically they won’t make Candidate A win any more than they already have).
  • The system is highly vulnerable to tactical voting: meaning people will change their vote depending on the (perceived or actual) votes of others. Not only does this make the election insincere, but it leads to the strong showing of candidates that no one is really happy with, and a weak showing of candidates people might
    actually really like.
  • perot 1992 prez button
    1992 Ross Perot Presidential Button [Image Source]
    Tactical voting also allows for what is called the Spoiler Effect. It’s the idea that if a voting bloc is split between two similar candidates, they will spoil each other’s chances of victory, because a dissimilar candidate (supported by a unified voting bloc) will win. Even if more people voted against that candidate than for them. It’s what happened in 1992 with Ross Perot; he and the Reform Party stole votes away from George H. W. Bush and the Republicans, letting Bill Clinton and the Democrats take the victory.
  • Finally, FPTP inevitably leads to a two-party system. In political science, this is called Duverger’s Law. Because of the tactical voting, people will forego voting for their preferred candidates for one who has a stronger chance of winning and keeping a hated candidate out of office. This leads to a us-versus-them mentality of voting, which leads to a “us” Party and a “them” Party. All other parties are pushed out for being considered “too weak” to win, or they fuse with another party to achieve a better chance of spreading their agenda within a stronger party structure.

two party system xkcd
Me and my friends while I was researching for this post [Image Source]
There are many other voting systems other FPTP which may be more fair or advantageous, depending on what you are trying to accomplish with your election. Are we looking for a system that makes the top vote-getter the winner no matter what? Do we what a system that elects a compromise or consensus candidate over a candidate that is loved by some and hated by others? Do we want to ensure that tactical voting doesn’t lead to a Spoiler Effect?

For the purposes of this post, I want to just focus on a new election structure for general elections, for example, district elections for Congress or mayoral elections. I’ll save tackling the Presidential Election (and the Electoral College), voting structure within Congress, and Primary elections for another post.

Voting and Election theorists have developed a number of new voting systems to accomplish different goals throughout history. So if I am looking for a new system to replace FPTP, first I have to identify what kind of elections and elections results I desire in order to narrow down my selection.

approval ballot.png
Example of Approval Voting Ballot [Image Source]
Firstly, I blame a lot of the problems with FPTP on the fact that I can only vote once. It’s why people feel compelled to vote tactically, it leads to wasted votes, and it creates the opportunity for the Spoiler Effect to influence an election result. So my first criterion for an alternative system to FPTP is that you must be able to vote for more than one candidate. If I can vote for multiple people (shown on the right), it would eliminate the penalty for voting for a minor Party (because I could just vote for a minor candidate I love and a major candidate I wouldn’t mind winning). Additionally, it could lead to more moderate candidates having better chances of victory, because again, people could vote for an extreme candidate they might favor and a moderate candidate they wouldn’t mind winning. Voting for multiple candidates at once weakens the power of the two-party system, because it gives minor parties more chances for success, which is important to me, personally.

Example of Range Voting Ballot [Image Source]
Secondly, multiple voting could be more expressive of my preferences if I could rank my choices. I think this attribute is necessary in electoral structures that allow for multiple votes, because voters don’t view all candidates as equal nor do they care equally who wins. So it follows that there must be a mechanism that allows voters to show their level of agreement with candidates. So my second criterion: there is some kind of ranking system to better express preferences. There are lots of different ways to rank choices: A-F scale, 0-10 scale (similar to example shown above), Agree-Disagree scale, Excellent-Poor scale, etc. Or maybe just number the candidates in the order I would most like to see them win; thus the scale is dependent on how many candidates there are. I particularly like the ballot pictured above because it puts all candidates on the same scale. But I would change 0 from “worst” to “no preference”, and make 1 the “worst” ranking. That way we do not have to force voters to rank every single candidate; candidates not ranked can just be considered to be a 0: “no preference”.

schluze computation
The complicated computation of the Schulze Method, created by Marcus Schulze (1997) [Image Source]
Thirdly, I actually want to go back to one of the advantages of FPTP: its simplicity. I think that one of biggest reasons the United States doesn’t switch to another system is that a lot of the available alternatives are too complicated for the average citizen’s taste. They like the clarity and seeming transparency of FPTP; no one can mess with the system without being easily caught. Additionally, it is easily to count votes in FPTP, and thus, follow and predict elections as they are being counted. I don’t think that is one characteristic of American election season that should have to be sacrificed. Finally, if I am trying to sell to the American people a new electoral structure, I want it to be simple. So that’s my third criterion.

Now I have three criteria for an alternative primary:

  1. Voters must be able to select more than one candidate.
  2. Voters must be allowed to rank their selections based on preference of those candidates.
  3. The system must be simple.

I want to a take a minute to say that of all the voting systems I have looked at which satisfy my first two criteria, I think a Range Voting system (which I’ll explain below) would work the best as an alternative to FPTP in the United States (honorable mentions include Borda Count (originally developed in 1770), Schulze/ Pathwise Method (1997), and Ranked Pairs (1987)). In theory, I think I prefer Schulze Method the most, but its process of selecting the winner is very complicated (shown above) and thus fails the third criterion.

So that is what I am advocating for: in general elections around the country, we should switch from First-Past-the-Post to Range Voting.

Now, we need to talk about the Plurality Principle. This is the idea that the person who gets the most votes must be the winner. It’s the foundation of FPTP. On the surface that seems like a good idea: who would want a unpopular candidate to win? But I think we must rethink the validity of this principle. (I define validity as the ability to do what is intended to be done.) As I noted earlier, the person who gets the most votes in FPTP may not necessarily be the popular candidate: they might have just run against several similar candidates who split the votes of the majority of the people (the Spoiler Effect). The goal of the Plurality Principle is that the most popular or supported candidates should win, but dictating that the top vote-getter win does not always satisfy that goal. Range Voting does not satisfy the Plurality Principle on a technicality: it provides the opportunity for the top vote-getter to win, but that does not necessarily always happen. But that’s because Range Voting takes into account the second and third preferences of voters, meaning that a consensus candidate could win instead. This means that Range Voting eliminates the Spoiler Effect, and also that Range Voting might be closer to complying with the spirit of the Plurality Principle than FPTP.

Let’s look closer into how Range Voting might look in a hypothetical election.

Example of (Simple) Election Results Using Range Voting:

% of Vote Candidate A Candidate B Candidate C
45% 10 6 1
25% 3 7 2
30% 1 5 8
Explanation: 45% of the voting electorate ranked Candidate A with a 10, meaning they would love if they won; Candidate B with a 6, meaning they wouldn't mind if they won; and Candidate C with a 1, meaning they really do not want them to win. 25% of the electorate gave A a 3, B a 7, and C a 2. And 30% of the electorate gave A a 1, B a 5, and C an 8.

In FPTP, Candidate A would have won, even though the majority of the electorate really didn’t like them (55% of the electorate gave Candidate A a 3 or less). I make this assertion assuming that each voting bloc would use their single vote in FPTP to vote for the candidate they ranked the highest (A gets 45%, B gets 25%, and C gets 30%).

But in Range Voting, Candidate B, who has high-to-moderate support throughout the whole electorate would have won.

Points for Candidate A Points for Candidate B Points for Candidate C
450 (45 x 10) 270 (45 x 6) 45 (45 x 1)
75 (25 x 3) 175 (25 x 7) 50 (25 x 2)
30 (30 x 1) 150 (30 x 5) 240 (30 x 8)
555 Total 595 Total 335 Total

Because of Range Voting, everyone is pleased with the election results, instead of a minority being overjoyed at the expense of everyone else’s disappointment. I will assert that this outcome is better than the outcome from the FPTP election.

If this system looks vaguely familiar, it’s because it’s already used in other situations. The Olympics use range voting for figure skating, for example. Each judge rates each performance with a score, instead of watching all the performances and selecting just one to win the whole competition. Internet sites like IMDb use range voting to rank movies, allowing visitors to share their preference for a particular movie on a scale from 1 to 10.

Of course, Range Voting is not some miracle solution. It does have disadvantages. But I want to note that it was fewer disadvantages than FPTP.

  • Range Voting is vulnerable to a different form of tactical voting called Bullet Voting. Bullet Voting is when voters only rank one candidate, because they don’t want to give points to other candidates. This effectively turns Range Voting into just Approval Voting.
  • Range Voting also fails the the Later-No-Harm Principle, meaning that giving a positive ranking to a less preferred candidate might lead to a more preferred candidate losing. This is directly related to Bullet Voting.
  • Range Voting does not work well when it’s supposed to provide more than one winner in an election (ex. a City Council or School Board election if all candidates run in the same election).
  • The validity of Range Voting, like many other voting systems including FPTP, is threatened by uninformed voters and low voter turnout.
  • Multiple voters, who share identical opinions of the candidates, might rank those candidates differently on the ballot, since each voter interprets the scale differently.

I believe Range Voting is more capable than FPTP at accomplishing the goals of democracy: ensuring the will of the people is supreme. Range Voting gives consensus and moderate candidates better chances of winning, it breaks the power of the two-party system, it protects against the Spoiler Effect, and it is simple and understandable to the general public.

There unfortunately have been few times Range Voting has been studied in real governmental elections. I fear that it’s because Range Voting doesn’t satisfy the Plurality Principle, which many people are uncomfortable with giving up. But the United States has never been a nation afraid of innovation or experimentation.

I say it’s time to try it out.

-Brian (Twitter: @iambriam)


2 thoughts on “Our Elections Are Threatening Our Democracy

    1. I don’t think I have anything against IRV. I just like that in Range Voting, you can express your preference for each candidate on a scale rather than in relation to other candidates. There is a lot of debate in election reform circles about IRV vs. Range/Approval Voting, which I am still going through myself.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s