Therese Coffey, the Conservative MP for Suffolk Coastal, has kindly been responding to me in a discussion on the Alternative Vote (AV). It started with this tweet from her:
Bbc continues to pump rubbish that to win with AV requires >50% of vote when its own R5L test showed you didn’t. Bias? #no2AV
To which I responded:
Of course a candidate needs more than 50% of the vote. That’s fundamental to how the system works.
After all, it’s right there in the description of how the system works right? If no candidate has reached 50%, you proceed to another round until they do. If you didn’t have to reach more than 50% how would you know who has won?
So let’s say it is the second round and Conservatives have 48%, Labour 30%, Lib Dems 12% and Greens 10%. Nobody has more than 50% of the vote so we go to the third round. Greens are knocked out and now Conservatives have 52%, Labour 34%, LibDems 14%. The conservative candidate has more than 50% so has won.
But what would happen if you continued the system to its logical conclusion? LibDems get knocked out and all their next preference votes go to Labour. Now Labour have 48% of the vote and the Conservatives have 52% still. Obviously the Conservatives have still won.
You could even go another round. Labour get knocked out and now Conservatives obviously have 100% of the vote.
Wow! 100%! Everyone loves the Conservatives. Absolutely one hundred percent of everyone!
Of course not. That’s ridiculous.
But for the same reason neither does having greater than 50% mean that the majority of the electorate really support them.
So why is 50% important then? Well notice that the winner didn’t change in those last two rounds. Once the Tories had reached 52% of the vote, Labour was hardly going to overtake them and get 53% of the vote because otherwise the total number of votes would have to be 105%. Even if those LibDem voters have no further preference and so their votes are dropped from the total number of votes, the Conservatives can’t mathematically drop below 50% again. Their percentage can only stay the same or get higher. Once you hit the majority you are guaranteed to continue having the MOST votes in future rounds.
So getting 50% of the vote simply means, stop counting because there really is no point in continuing.
This is a statement made on the Yes to AV campaign website:
Your next MP would have to aim to get more than 50% of the vote to be sure of winning. At present they can be handed power with just one vote in three.
Therese Coffey takes issue with this. And you know what, so do I. While the statement is true, aiming for 50% is the only way you can guarantee winning with first past the post too! Sure, they can get in power with far less than that, but if they aimed for less they couldn’t be absolutely sure of it.
The second statement is more complicated, because it’s not clear what a vote actually means when there are potentially several rounds of votes. And it ignores eligible voters who don’t vote for anybody.
That 50% doesn’t mean anything different to saying they must get a 100%. The only reason we stopped counting at 50% was because we have now reached a stage where the winner won’t change.
What does 50% mean then?
Let’s say we’ve taken it to a round with just the Conservative and Labour candidates left. The results are Conservatives 70% and Labour 30%. All we can say is that 70% wanted the Conservative candidate to win over the Labour candidate. And that 30% wanted the Labour Candidate to win more than the Conservative.
We can’t say 70% of voters wanted Conservatives to win. Half of the Lib Dem supporters may have not put down either Labour or the Conservatives as a preference at all in which case their vote won’t be counted. And anyway, there is a difference between a vote that says I really want Lib Dems to win, but if not as long as it isn’t Labour I’m happy; than saying I support Conservatives.
What Therese Coffey wants to do though, is compare that final count of votes to the total number of people who voted in the 1st round (i.e. ‘total cast ballots’) including those dropped LibDem votes.
Me:
… stating a candidates final round count as a % of 1st round votes cast is meaningless. Each round is separate
I completely disagree it is the share of the people who actually cast a vote - not meaningless at all but smokescreen from yes2AV
A smokescreen perhaps, but if it isn’t meaningless what DOES it actually mean?
Much of this debate is based around this mock election from BBC Radio 5. The results are presented in a table on the Guido Fawkes website.
Okay, so in this example the winner apparently got 49% of total votes cast. The final result was Labour 49%, Greens 41.7% and no preference 9.3%. If this seems odd to you, I’ll address that in a moment. But first, what does that 49% mean here in this scenario?
Well 49% is the percentage of people who thought at least one candidate was worth voting for who also think out of the remaining candidates that Labour is the best one.
So what?
In each round every eligible voter effectively has the opportunity to say ‘this is my favourite of these candidates’ or ‘I don’t care I like/dislike them all the same’. There is nothing special about the first round.
Lets say I wanted to vote for the Green Party as my first and only preference. But then the Green Party withdraw their candidate. I don’t have a second preference so who do I vote for? Well nobody. I don’t have another preference so I just don’t vote.
But what if the Green Party hadn’t withdrawn? Well now I’d put them as my first preference. In this scenario they come last, so we go to round 2. Round 2 is exactly the same as the first round in the previous example. Except some people argue that I’m a voter in this example but not in the previous example. Why?
If we added a box that said ‘None of the above’ for people who didn’t like any candidate, the number of ballots cast would increase and the percentage would change again. That’s interesting survey data, but it doesn’t alter in any way the result of the election. All we are doing is playing with numbers and confusing people.
The Radio 5 BBC mock election makes no sense. They counted the non votes but that’s not how the system works.
As it turns out though, counting the results this way wouldn’t really affect the final result. All that could happen is you would continue to count the votes even after no other candidate could overtake the candidate in the lead. An extra round would just mean everybody got to bed a little later.
The 50% only makes sense in the context of a single round. If a candidate receives a majority in a single round of votes it means that the majority of voters in that round would not prefer any of the other remaining candidates. If they get less than 50% it means there might be a candidate left who is preferred by more voters than the candidate currently in the lead. Anybody who never voted at all or has no preferences left are all saying I don’t care who wins out of this remaining lot.
Let’s say you have two right wing candidates and one left wing candidate. In the current system, if half the voters were right wing and half left wing, the left wing candidate would have an advantage because right wing votes get split between two candidates. AV fixes that bias.
And that’s pretty much all it really does. Personally it allows me to vote for Lib Dems instead of the Conservatives without fear of wasting my vote that could have counted against Labour getting in. If my Labour candidate ever reaches over 50% of the vote in any round, then my Lib Dem vote wouldn’t make any difference anyway. But if they are under 50% my vote can be transferred to the Conservatives and they might win.
The reason changing the system like this makes sense is round two is how I would have voted had the Lib Dems never entered a candidate at all.
(Which is also why the first round using the alternative vote is not the same thing as the result of first past the post. People will vote differently.)
That doesn’t mean I particularly like any of the candidates or think they will do a good job. I cast my vote based on who I prefer compared to whoever else is running. That’s all that really matters. Yes, turnout matters as an indication of how much people think the election matters as a whole, but for determining who wins it only matters who has the MOST votes. First Past the Post declares the winner based on who has the most votes and the Alternative Vote does the same. It does not force a situation where candidates are liked by a majority of the electorate. It simply goes a long way to fixing the split votes problem.
How high the percentage of votes is doesn’t mean anything useful. Getting 52% in the first round does not show the same level of support as getting 52% in the fifth round. And of course, getting 52% with a voter turnout of 10% is very different than with a voter turnout of 90%. In my opinion, focusing on these raw figures out of context does more harm than good and makes the system seem more complicated than it is.
And of course, I’ve engaged in it and probably made it sound even more complicated. But trust me when I say, I can make first past the post sound complicated too! This is why I don’t run or really engage in any political campaigns! Instead I quietly think through the issues and put out a rare detailed post that most people won’t read.
Anyway, whatever the result of this referendum kudos to @theresecoffey MP for engaging in a debate with a stranger over Twitter.