Another voting system to wrap your brains around: ‘Multi-Candidate First Past the Post’ (MC-FPTP)

(After writing this post, I discovered that this ‘new’ system I’d just dreamt up already exists, although it hasn’t been tried out in any major election. It’s called Approval Voting.)

Following on from my previous post on the Alternative Vote (AV), I’ve been playing around in my mind with different possible voting systems. As I’ve said before, the best system, in my opinion, is STV, favoured by the Liberal Democrats (who I won’t be voting for, by the way, because of their duplicitous abandonment of support for a referendum on Britain’s EU membership) and the Electoral Reform Society.

However, if the single-member constituency system is retained (STV relies on multi-member constituencies), this means, by definition, that only one candidate can win. So you need the best way to find out which is the candidate who is genuinely the most popular. AV doesn’t do this, despite the Scottish British-English prime minister’s hypocritical assertions to that effect, as I argued before: it merely determines the least unpopular candidate, or – technically – the candidate that a majority of voters whose selections remain in play within the AV system (i.e. not necessarily the majority of all those who voted) are prepared to give their grudging consent to, even if that candidate is only their third-, fourth- or even fifth-choice candidate.

The best way to determine who is genuinely the most popular candidate, even if not the first choice of a majority of voters, is as follows:

  • All voters can vote for as many candidates as they like: they simply mark an X beside the name for each candidate they wish to vote for
  • Each vote carries the same weight, i.e. there is no attempt to determine people’s ‘preferences’: if you vote for someone, this means you are making a positive choice for that candidate. They may not in fact be your personal ‘first choice’; but they are a candidate who you would be ‘happy’ to see elected and to consider as someone for whom you had made a definite, deliberate choice
  • The winner is the candidate obtaining the most votes, whether a majority or not.
Adopting a system like this means that there are no excuses. If you vote for someone, this is not a tactical vote, or a second or third best: you’ve voted for them, so you can’t complain if they win. It means you can vote for candidates you personally like, even if you don’t like their party very much, as well as voting on the basis of the political party you want to win the election overall. And it means, above all, that the winner is genuinely the most popular candidate: the one whom most people in the constituency are willing to support in a positive way.
Like AV, this would not be a proportional system when translated to a national level, simply because it continues to exclude the smaller parties from winning any seats, and because candidates can still win on a minority share of the vote. However, it is equally true, under my system, that many candidates would secure the support of the majority of those voting, even if not a majority of the total vote, based on the fact that more than half of voters would have indicated the winning candidate as one of their choices (or their only choice), whereas that candidate’s total share of the vote would be diluted by the fact that some voters had selected more than one candidate.
Similarly, my proposed system would change the results of elections in some unpredictable and quite dramatic ways. For instance, in the constituency I live in, the sitting Tory MP will probably be re-elected under FPTP with a majority of the vote (e.g. around 55%). However, the Liberal Democrat is likely to come a close second, and the other parties (in previous elections at least) have been nowhere in sight (I think Labour polled around 10% last time). If existing Labour voters decided to vote Labour and Liberal Democrat, under my system, this could be enough to swing the seat in favour of the Lib Dems. You could say this is just engineering the system to favour the Lib Dems at the expense of the Tories. However, if the value of each vote is identical (i.e. each vote is an equal, positive vote for the candidate(s) of the voter’s choice), there is actually no way you could say that if more people voted for the Lib Dems than for the Tories under this system, this was a ‘false result’, like those under AV. That would amount to insulting, indeed slandering, the electorate by saying that voters didn’t really prefer the Lib Dem candidate to the Tory, whereas the results prove the opposite: one vote = one positive choice.
Similarly, this system would empower voters to mark their approval for all of the candidates / parties they liked, meaning, among other things, that the share of the vote obtained by the smaller parties would rise: if you support UKIP but would be worried, under the existing system, that voting for them would let the Labour candidate in, under my system you could vote for both UKIP (to express your support for their positions on the EU and immigration, for instance) and vote Tory to keep the Labour candidate out. But then you couldn’t grumble about your Tory MP, because you’d made a positive choice for them. If you don’t want the Tory, don’t vote for them; but then don’t complain if the Tory wins on a minority of the vote, because the system has at least enabled the most popular candidate to win.
So I offer this new system (but horrid name) for the votes of my readers: better than standard FPTP or not; and better than AV or not?

4 Responses

  1. Umm….. not better.

    Think about it. Suppose a Labour candidate wins under FPTP with 51% ( an absolute majority) over Conservative with 39% and a BNP candidate with 1o%. Lib Dems skip the race.

    But then we put your system in. There’s a lot of talk of the BNP candidate running well and nearly half of the Labour voters cast a vote for the Conservative as well to help defeat the BNP. This puts the Conservative up to 64%. A quarter of the BNP voters do too, eager to oust Labour – so up to 66%. Meanwhile, Tory voters are more focused on ousting Labour. Only a quarter of them vote for the Labour candidate as well, and few BNP voters do either. So the final vote is 66% for the Conservative, 62% for Labour and let’s say 15% for the BNP.

    Not good.

    Or say you have a constituency where it’s roughly split between the Lib Dems, Labour and Conservative. How do you vote in your system? If you can vote for only one candidate, you aren’t going to vote against that candidate. If you vote for two, you effectively cancel out each other’s vote. The winner would be the candidate who most effectively persuaded supporters to not vote for anyone else.

    AV avoids all of that.

    • Thanks for the reply, Peter. I take your point, even though I don’t think your first example is very credible. In such a scenario, Labour would tell its voters that the best way to keep the BNP out would be to vote Labour only. A more likely scenario is the one in my constituency, where the Lib Dems poll around 35% and the Tories 55%; but if there were a swing to the Lib Dems and the Labour voters could be persuaded to vote Lib Dem as well, the Lib Dems could conceivably overtake the Conservatives, even though the Tories might win with more than 50% of the vote under FPTP. It makes people more likely to vote tactically because they can also express their support for their favourite candidate, who doesn’t have a chance of winning. However, tactical voting can cut both ways: you could get UKIP, English Democrat and BNP supporters also voting Tory to keep either Labour or the Lib Dems out. In a sense, then, Approval Voting (which I’ve discovered ‘my’ system is called) tends, I would think, to favour the two strongest candidates: one from the left of centre and one from the right.

      In your second example – a three-way split – voting for two candidates (e.g. Labour and Lib Dem) doesn’t dilute or cancel out each other’s vote: it adds to the total and increases their share of the vote – it’s just that if you vote for more than one candidate, each party’s share of the vote increases by a proportionally smaller amount than if you vote for one. But then the share of the vote obtained by candidates for whom people vote exclusively also increases by a proportionally smaller extent the more other voters select multiple candidates. So it is still an effective way to try to ensure that the candidate you don’t want to win is defeated.

      The question is, is such tactical voting fair or legitimate? Well, it happens under the present system, the only difference being that the tactical voter can’t then register support for their favourite candidate as well. It is legitimate in the sense that it’s entirely up to the individual voter how they vote, whether ‘positively’ or ‘negatively’ – and it might be just as, if not more, important for particular voters to try to ensure some candidates are beaten (e.g. BNP ones) than only to vote for their ‘favourite’ candidate or party.

  2. I’ll focus on the second example. I only vote for two candidates if my primary goal is to defeat the third candidate. If I really want one candidate to win, i will only for that candidate — even if I do think the second candidate is okay. And party ioytalists certainly would say “just vote for our candidate.” The end result is the voter foolish enough to vote for more than one candidate would help defeat their first choice.

    I also don’t like the idea of a candidate losing who would win 51% with FPTP.

    • I agree the voter needs to be aware whether voting for more than one candidate will risk defeating their preferred candidate. But if there’s any doubt, voters will just vote for their favourite, I would have thought. I agree that being defeated under Approval Voting when you might have won under FPTP is a bit dodgy. However, it’s also highly unlikely: if there is enough backing behind the winning candidate to ensure they get an outright majority under FPTP, it’s very improbable that that would get overturned by people voting tactically under Approval Voting, especially as some voters who prefer other candidates would also vote for the winning candidate if their candidate didn’t have a chance of winning.

      If this unlikely eventuality did come to pass, then it would reflect the fact that the winning candidate was also rather unpopular with a large part of the electorate: Approval Voting allows negative preferences to be expressed, i.e. voting to defeat a particular candidate. Similarly, you have to bear in mind that a slim majority under FPTP doesn’t necessarily mean that a majority of voters really prefer that candidate, because the votes received by that candidate will include tactical votes and votes from people whose preferred candidate doesn’t stand a chance, so they’re voting for their second or third choice.

Leave a Reply

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

You are commenting using your 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 )

Google+ photo

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

Connecting to %s

%d bloggers like this: