In the English elections held on the same day as ours, the BBC estimated Labour’s share of the vote as 38%. You’ll have to ask the BBC why they didn’t bother coming up with an estimation for Wales. I guess we’re pretty small beer compared to where the real political clout is.
But it got me thinking. Because of the cack-handed nature of local democracy in Wales (a mixture of single- and multi-member wards with the undemocratic first-past-the-post system) it’d be very time-consuming to do on a Wales-wide basis. But it can be done, and relatively simply, on the basis of a number of individual wards.
The mechanism I’ll use is to take the proportion of people voting for every candidate in each ward, summing the percentages for multiple candidates from the same party. Then I’ll add up all the proportions for every ward to come up with a single number for each party, and divide the whole lot so it adds up to 100, giving me a pretty good proxy for the share of the vote.
Some parties will cry foul. After all, this approach benefits those parties that put up a full slate of candidates (and I’ve belatedly realised that the Conservatives were the only party to do so – Labour only fielded one candidate in two-member Sully). But you can hardly claim a share of votes that weren’t cast, so I think my approach is valid.
You can see my workings here.
So here it is, the share of the vote of the different parties in Penarth/Sully in this election.
- Conservatives – 34%
- Green – 2%
- Independent – 6%
- Labour – 41%
- Lib Dem – 1%
- Plaid – 11%
- UKIP – 4%
Before looking at what this means for the parties involved, there’s something else of tangential interest in the calculations. The total available vote in the calculation per ward is 200. Yet the most cast in any ward was 192 in Cornerswell and St. Augustine’s, with 191 in Plymouth and 188 in Stanwell. In Sully just 178 were cast, which means that rather a lot of people in Sully only cast one vote. This can be a way of tactical voting (by voting for only one candidate you reduce the chance of one of your non-preferred candidates getting elected), or it could just be a local dynamic. But let’s tie it in to the turnout in Sully, which at 44% was not only the highest in the area, but was the one ward to buck the trend of turnout increasing in less deprived areas. That all suggests that a substantial number of people voted for one candidate who would not otherwise have voted. I can only guess that the intrigue of two candidates from outside the usual suspects is the reason for this anomaly.
Referring back to the BBC website, Labour and the Conservatives did 3% better than their share of the vote in England. The big loser, of course, were the Lib Dems, who got 15% less than their England vote. It’s worth noting that they also got just half as much as the Green vote, despite standing in double the wards. Plaid meanwhile got 11% more than they did in England 😉
So the Conservatives can feel pretty aggrieved with the way this election panned out. Nailing just two seats with 34% of the vote must be galling. And Labour will be cock-a-hoop with nobbling six seats on just 41% of the vote.
Mind you, that’s the way things go with the outdated, undemocratic voting system we know and love as first-past-the-post.
Let’s fast-forward to our by-election, just for fun, and see how these results compare with the results from Penarth and Cardiff South in 2010. If we make the ridiculous assumption that Penarth is a microcosm of the whole constituency then we’d see that:
- Plaid polled 7% better in 2012 than in 2010
- Conservatives polled 6% better in 2012
- Independent polled 4% better in 2012
- Labour polled 2% better in 2012
- Green polled 1% better in 2012
- UKIP polled 1% better in 2012
- Lib Dems polled 21% worse in 2012
This is actually a bit more of an interesting exercise than I’d initially imagined, because with the exception of the Lib Dems, the polling in Penarth’s 2012 local authority election was within single figures of the 2010 constituency result. Now I think that the likelihood of the Lib Dems polling 1% in the by-election is about as strong as Plaid tripling their vote. Not impossible, but highly unlikely. But it’ll be interesting to use these results to come up with some polling predictions before the election.