TANDQ 14: Vote of No Confidence

In 2014-2015 I wrote an education column called "There Are No Dumb Questions" for the website "MuseHack". As that site has evolved, I have decided to republish those columns here (updating the index page as I go) every Wednesday. This fourteenth column originally appeared on Thursday, April 30, 2015.

How can I influence a democratic election?

Vote. That’s your best shot, particularly in light of voter apathy. And when you do vote, make it an informed decision, particularly if you’re getting others around you to vote too. I’ve noticed lots of voting talk lately, from the United Kingdom’s general election this May, to the Canadian federal election later this year, to the Hugo Awards ceremony in August. Oh - if you haven’t heard, that last is getting more press now due to the “Rabid Puppies” group (sometimes confused with the “Sad Puppies” group, but they’re in their third year). But take care, as not all democratic voting systems are the same! Let’s delve deeper.

Many countries that are (or were) British colonies will be familiar with the “first past the post” (FPTP) system. It’s more formally known as the “single member district plurality” system, and basically involves marking an “X” next to your candidate of choice - the candidates having been previously chosen by their political parties. When the votes are finally tabulated, whoever has the most “X”s wins - even if that individual did NOT collect an overall majority. To illustrate, let’s say we have three candidates: A, B and C. The election results are A: 25%, B: 35% and C: 40%, so C is put in charge of 100% of the district, despite the fact that 60% of people chose someone else. We can see from this that there are problems with FPTP - not the least of which is that in the long term it encourages a two party system (see the video link at the end). But it gets worse.

Mathematician Donald Saari has shown that the results of a plurality voting system can produce an outcome that is the exact reverse of actual voter preference. See his website for some introductory lectures, or search the internet for the famed “milk-beer-wine” example. As an abbreviated version: Given my election results above, consider a case where all those voting A and B would have ranked C dead last (meaning 60% of voters now really dislike the result) while all those who did vote for C would have picked A as their second best choice (meaning NO voters would rank A in last place - despite A being listed as “least popular”!). Obviously, ignoring those rankings is a big issue! Can including them fix the problem? Well, not entirely. But a ranking system is what they presently use for the final ballot of the Hugo Awards.

Rank and File

What ranking does fix is the problem of “strategic voting”. For instance, if a person really doesn’t want candidate C to win, then even though they like A best, they may choose to throw their support behind B, who has the better chance of beating C. This “spoiler effect” is what can lead to recent UK news articles like “Vote Conservative in seats Ukip can’t win”. I suspect it’s also behind the latest trend in overall election campaigns, which don’t seem to say “Vote for me, B!”, so much as “Don’t let C win!”. Fortunately, this issue evaporates with what we call “Instant Runoff Voting” (or in Australia, “The Alternative Vote”). It’s still a plurality/majority system, but you don’t mark a single “X”: you rank your candidates by preference. This means that, given the same election results above - we’d have no clear majority. So we drop out “A”, and the SECOND choice on their ballot is redistributed. If they all hated C, this results in B: 60% and C: 40%, meaning candidate B is declared the winner!

Of course, this preferential voting system doesn’t eliminate issues like gerrymandering, or address the problem that all voters could have been satisfied with “A”, and the system is still susceptible to mathematical paradoxes. (Not all relationships are transitive in nature: If A beats B and B beats C, it’s possible that C beats A.) But this system does mean that you don’t need to worry about how the other people in your district are going to vote! Which implies that candidates have to campaign based on their platforms, not against someone else’s. The final voting for the Hugo Award gets even more interesting here, in that “No Award” is a valid ranked choice. So to win, the first nominee who ends up with over 50% of votes must be subsequently tested to ensure that the “No Award” choice was not ranked higher than them on relevant ballots. To read more about the Hugo voting system (and to see how they pick their runners-up), go to their website here. Note also that their award nomination process is not the same procedure.

At this point, having torn down Plurality systems, what’s left? Why, the Proportional Representation systems, used by many European democracies. The idea here is that you get two votes - one for the governing party, and the other for a party candidate. Ergo, if the final results show a preference of 50% for Party 1, 45% for Party 2, and 5% for Party 3, the seats are filled to that proportion with the top candidates, as chosen by the second (simultaneous) vote. It’s more mathematically fair... but it’s not perfect either. Proportions cannot be exact, and there is a need for larger districts, meaning less local representation. Of note, in 2007, the Canadian province of Ontario held a referendum on whether to adopt the “Mixed Member Proportional” variation on this system - a system with double the seats compared to the regions - so feel free to watch Rick Mercer’s analysis for more. (The proposal was defeated by 63% of Ontarians, and “First Past the Post” remains in effect.)

Wait, doesn’t that referendum mean that Ontario held a vote on how voting will take place? How very meta. And yet, it’s only by proposing such things that any (necessary?) democratic reforms can occur. Which brings us back to the introduction: Vote, and try to do it in an informed way. It’s probably your best shot at influencing things... short of becoming a candidate yourself.

For further viewing:

1. Why Democracy is Always Unfair

2. Some Sad Puppy Data Analysis

3. The Problems with First Past the Post Voting Explained (Video)

Got an idea or a question for a future TANDQ column? Let me know in the comments, or through email!