Why Preferential Voting Works

I’ve lost count of the number of times I’ve written this out, but if I post it here, I’ll be able to refer back to it, and with the UK election coming up tomorrow, and electoral reform on the agenda there, it’s relevant again.

Many, if not most, voting systems in countries that are universally recognised as democracies are fundamentally undemocratic.  The problem is with the simple majority, or “first past the post,” voting system.  You might think that it’s perfectly fair that whoever gets the most number of votes is duly elected.  However, that all depends on what you consider to be a majority.

Consider an election result that looks like this:

Candidate A:    40%
Candidate B:    35%
Candidate C:    25%

Under a simple majority system, Candidate A would win because he polled more votes than any other candidate.  The problem is that an overwhelming majority of 60% want someone else.   That’s just in a three-candidate election. If there were five, someone could potentially win with as little as 21% of the vote.  How is that democratic?

Simple-majority voting doesn’t elect the most wanted candidate.  It elects the least unwanted.  There’s a big difference.

Preferential voting elects the most preferred candidate.  Under a preferential system, instead of ticking a box, the voter ranks each candidate in order of preference, giving a 1 to the most preferred, a 2 to his second preference, and so on.  If no candidate has an absolute majority of 50% +1, then the candidate with the least votes is eliminated, and his second preferences distributed.  This continues until one candidate has a 50%+1 majority.

In the example above, Candidate C’s second would be counted.  Let’s say that four fifths of the people who voted for Candidate C chose Candidate B as their second preference and one fifth chose Candidate A.  That would mean that after preferences were counted, the result would be

Candidate A:    45%
Candidate B:    55%

And candidate B is duly elected as the most preferred candidate.  Although he does not have a majority of first preferences, he is clearly the most preferred candidate.

There are some places that go part way towards this system by holding run-off elections when no-one gets a clear majority.  However, these run-off elections are usually between the top two candidates.  If there are lots of candidates on the ballot, it’s possible that even the top two have less than 50% of the vote between them.  It could be that someone further down on the primary vote is actually the most preferred candidate.  Then there is the obvious expense and inconvenience of a second election.  This is why some advocates in the US refer to preferential voting as instant run-off voting.

Preferential voting also removes the problem of third-party candidates splitting a demographic, and the need to vote tactically to avoid such a split allowing less preferred candidates to sneak in.  Let’s take the example of the 1992 US presidential election:
Bill Clinton won 43.01% of the vote, George Bush won 37.45% and Ross Perot won 18.91%, which is a massive result for an independent candidate. (Source)  Perot’s vote was enough to tip the election one way or the other.  While it’s impossible to say for sure, it’s fair to suggest that Perot took more votes from Bush than from Clinton.  If two thirds of Perot voters had chosen Bush as their second preference, that would have given the election to Bush. 

I choose that example because I know a few people have me pegged as a mad, screaming leftie.  And I won’t deny that Clinton was the most preferable of the three options as far as I’m concerned.  The fact remains though, that the overwhelming majority of US voters in 1992 voted for a conservative candidate, but because there were two of them, the election went to the not-quite-so-conservative candidate.

It was, of course, a similar situation in 2000.  Ralph Nader’s 2.73% may have been paltry in comparison to Perot’s result, but that election was so close that it would still have been enough to tip Bush or Gore over 50% and the farce in Florida would be neither here nor there.  While it’s arguable as to which way Perot’s preferences might have gone, I would be gobsmacked if any Nader voters would have preferred Bush. 

With preferential voting, those who want to vote for an alternative to the major parties can do so safe in the knowledge that their preferences are recorded and their vote will not translate into a simple majority for a candidate representing the opposite party.

The movement for electoral reform in Britain seems to be advocating a system of proportional representation.  While it may be an even more representative system for electing an assembly, I have doubts about its workability.  Proportional representation makes it far easier for minor parties to be elected in numbers proportional to their support, rather than being squeezed out by the major parties on preferences.  Proportional representation almost always delivers a hung parliament.  Advocates say that hung parliaments lead to consensus government.  I can see their point.  Critics say that it allows minor parties to hold government to ransom and wield a power far greater than their numbers.  I can see that point too.  Every system has its flaws – some more than others.

In Australia, we elect the House of Representatives by preferential voting and the Senate (the states’ house and house of review) by proportional representation.  I see this as the best of both worlds.  All parochialism aside, I believe the Australian federal electoral system is the fairest there is.