There is, a weekend after the election, a lot of rage at the number of votes required nationally to gain a seat in Parliament. The figures bandied about are often fundamentally wrong, since you cannot count votes for a party in constituencies where they did not stand. The SNP is the most obvious candidate. Constituencies vary in size around a median value of around 70,000 voters. See here. The SNP gained 56 seats out of 59, collecting 1454436 votes on a 71.1% turnout
It has taken significant time to rearrange the results to allow me to discover what I want to know. I assume that those producing results have an intended structure for their results. I found the results in a spreadsheet but it is not quite in a form I can use; three days of pounding the keyboard later...
Here we go:
What I call the Apathy Party, all the people that did not vote, had majorities in 346 of the 650 constituencies. I noticed. in putting in the data, that generally UKIP came 2nd or 3rd and the Greens and LibDems were 4th and 5th. The typical result has an Independent 6th.
National turnout was 66%, i.e. 30698210 out of 46444321. maximum 81.9% (Renfrewshire East) minimum 46.1% (Manchester Central) StDev 5.6%. Total voting was:
15,746,111 people cast no vote, a mean of 24225 per constituency.
11,334,505 Con contested 647, so a mean 17518 per constituency and 34243 per seat won
9,347,326 Lab contested 632, mean 14790 /const, 40290 / seat won
2,413,229 LibDem contested 632, mean 3818/const, 301654 / seat won
6,867,293 UKIP contested 631, mean 10883/const, 6867293/ seat won
1,454,436 SNP contested 59, mean 24651 /const, 25972 / seat won (lowest)
1,149,810 Green contested 570, mean 2017/const, 1149810/ seat won
If we were to look at votes per seat. then that must count votes cast in seats that were contested, so the SNP can only be counted in seats across Scotland among Scottish voters.
Wales: voters 2,282,297 voted 1,498,433 [65.65%] seats 40: votes & seats by party follow
Con votes 783,864 seats 11 contested 40 71,260 /seat
Lab votes 408,213 seats 25 contested 40 16,328 /seat
Plaid votes 177,080 seats 3 contested 40 59,027 /seat
Green votes 38,344 seats 0 contested 35
UKIP votes 204,360 seats 0 contested 40
LibD votes 97,783 seats 1 contested 40
Wales elected the Labour party, 62.55 seats, 37% vote.
Northern Ireland is from habit quite different, politically.
Voters 1,236,683, voted 718,103 for 18 seats. 12/18 had under 60% turnout, 5 under 55%, only one over 61%. This tells us something about politics in Ulster (I’m not sure what, but something). Labour and the Liberal Democrats contested no seats here.
Didn’t vote won 17 of 18 seats (not Fermanagh, UUP. 72.6% turnout).
DUP votes 180,913 gained 8 seats contesting 16, 25.2% of the Ulster vote
Sinn Fein votes 176,232 gained 4 seats contesting 18, 24.5%
SDLP votes 99,809 gained 3 seats contesting 18, 13.9%
UUP votes 118,282 gained 2 seats contesting 15, 16.5%
UKIP votes 18,324 gained 0 seats contesting 10, 2.6%
Con votes 8,640 gained 0 seats contesting 15, 1.2%
Green votes 6,822 gained 0 seats contesting 5, 1%
Alliance votes 61,556 gained 0 seats contesting 18, 8.6%
Others votes 47,110 gained 0 seats contesting 13 around 6%
This is a good example of what 1st past the post does; the DUP gained four seats more than Sinn Fein with around 1% more of the national vote. The Alliance party gained no seats at all. The region elected the DUP, 44% of the seats on 25% of the vote
Scotland had, as we all now know, almost a whitewash, with 56/59 seats to SNP. Voters 4094784, voted 2910465 for 59 seats. Turnout was 71%, where 2 had under 57% turnout, 9 over 75%, one over 80%.
SNP votes 1,454,436 gained 56 seats, contesting 59
Con votes 434,097 gained 1 seat, contesting 59 - Dumfriesshire, Clydesdale & Tweeddale
Lab votes 707,147 gained 1 seat, contesting 59 - Edinburgh South
LibDem votes 219,675 gained 1 seat, contesting 59 - Orkney & Shetland
UKIP votes 47,078 gained 0 seats, contesting 59
Green votes 47,086 gained 0 seats, contesting 28
Others votes 9,009 gained 0 seats, contesting 37
Didn’t vote 783,864; Not voting ‘won’ only 7 seats in Scotland.
The first past the post system favours a two-party system and is sensitive to small changes in the middle ground. The SNP gained 50% of the vote (the votes cast) and 35.5% of the electorate - as if those who voted for independence voted SNP while the rest voted for something else. Under all proportional representation systems this would still give an absolute majority in Scotland to the SNP. Nationally, the vote was 24% Labour, 15% Conservative, 7.5% LibDem and 3% split between UKIP and Greens.
Dividing England into its 9 regions, as uk.gov does, I note that these are uneven in size. The North-East is half the size of Eastern and a third of the SouthEast. ‘Eastern’ is a label inconsistent with the others.
East Midlands Voters 3,350,769 voted 230402 for 46 seats. Turnout 66%: 3 under 60%, 7 over 71%
Con votes 969,379 won 32/46 seats,43.5% vote
Lab votes 705,676 won 14/46 seats, 31.6% vote
Lib votes 124,039 won 0/46 seats, 5.6% vote
UKIP votes 351,777 won 0/46 seats, 15.8% vote
Green votes 66,239 won 0/39 seats, 3% vote
Other votes 9,968 won 0/24 seats.
Didn’t vote 1,120,367 which ‘won’ 21/46 seats
Eastern Voters 4,364,656 voted 2,948,622 for 58 seats. Turnout 68%:
Con votes 1,445,946 won 52/58 seats, 49% vote
Lab votes 649,320 won 4/58., 22% vote
Lib votes 243,191 won 1/58., 8.2% vote
UKIP votes 473,600 won 1/57, 16.1% vote
Green votes 116,274 won 0/54, 3.9% vote
Other votes 15,201 won no seats, contesting 48.
Didn’t vote 1,416,034 which ‘won' 21/58 seats
London Voters 5,401,616 voted 3536291 for 73 seats. Turnout 66%:
Con votes 1,233,378 won 27/73 seats, 34.9% vote
Lab votes 1,545,110 won 45/73 seats, 43.7% vote
Lib votes 272,544 won 1/73 seats, 7.7% vote
UKIP votes 286,981 won 0/73 seats, 8.1% vote
Green votes 171,652 won 0/73 seats, 4.9% vote
Other votes 26,268 won 0/56 seats.
Didn’t vote 1,865,325 ‘won' 31/73 seats
North East Voters 1,941,841 voted 1188183 for 29 seats. Turnout 61%:
Con votes 300,883 won 3/29 seats, 25.3% vote
Lab votes 557,100 won 26/29 seats, 46.9% vote
Lib votes 771,25 won 0/29 seats, 6.5 vote
UKIP votes 198,823 won 0/29 seats, 16.7% vote
Green votes 43,051 won 0/28 seats, 3.6% vote
Other votes 11,201 won 0/15 seats
Didn’t vote 756,658 ‘won' 18/29 seats
North West Voters 5,259,569 voted 3364055 for 75 seats. Turnout 664:
Con votes 1,050,124 won 22/75 seats and 31.2% of the regional vote
Lab votes 1,502,047 won 51/75 seats, 44.6% vote
Lib votes 217,338 won 2/75 seats, 6.5% vote
UKIP votes 459,071 won 0/75 seats, 13.6% vote
Green votes 107,889 won 0/75 seats, 3.2% vote
Other votes 24,926 won 0/75 seats
Didn’t vote 1,895,514 ‘won' 18/75 seats
South East Voters 6,419,548 voted 4,394,357 for 84 seats. Turnout 68%:
Con votes 2,268,973 won 79/84 seats including the Speaker, with 51.6% of the regional vote
Lab votes 804,774 won 4/84 seats, 18.3% vote
Lib votes 413,587 won 0/84 seats, 9.4% vote
UKIP votes 646,959 won 0/84 seats, 14.7% vote
Green votes 226,381 won 1/82 seats - Caroline Lucas at Brighton Pavilion, on 5.2% of the vote
Other votes 32,179 won 0/52 seats
Didn’t vote 2,025,191 ‘won' 22/84 seats
South West Voters 4,076,494 voted 2,836,213 for 55 seats. Turnout 70%: one over 75% 4 under 65%. Good turnout reduced the apathy vote to 19/55 seats. Like Midlands below, divided only between the two major parties, 64.2% of votes.
Con votes 1,319,987 won 51/55 seats, with 46.5% of the regional vote
Lab votes 501,684 won 4/55 seats 17.7% vote
Lib votes 428,873 won 0/55 seats 15.1% vote
UKIP votes 384,546 won 0/55 seats 13.6% vote
Green votes 168,130 won 0/54 5.9% vote seats but close in Bristol West
Other votes 32,871 won 0/35 seats, but Independent 3rd in Devon East.
Didn’t votes 1,240,281 won 19/55 seats
West Midlands Voters 4,140,587 voted 2,628,943 for 59 seats. Turnout 63%: 5 under 55%, 5 over 71%. Con & Lab polled 74.6% of the vote.
Con votes 1,098,113 won 34/59 seats, 41.8% of regional vote
Lab votes 865,067 won 25/59 seats 32.9% of regional vote
Lib votes 145,009 won 0/59 seats 5.5% vote
UKIP votes 412,770 won 0/59 seats 15.7% vote
Green votes 85,653 won 0/59 seats 3.3% vote
Other votes 22,331 won 0/32 seats
Didn’t vote 1,511,644 ‘won' 40/59 seats
Yorkshire & Humberside Voters 3,875,477 voted 2,444,143 for 54 seats. Turnout 63%: 2 under 55%, 6 over 70% (Nick Clegg’s LibDem seat topped this region at 75%). Didn’t vote won 42/54 seats
Con votes 796,792 won 19/54 seats, 32.6% vote
Lab votes 956,837 won 33/54 seats 39.1% vote
Lib votes 174,065 won 2/54 seats 7.1% vote
UKIP votes 382,995 won 0/53 seats15.7% vote
Green votes 86,471 won 0/47 seats
Other votes 38,055 won 0/54 seats
Didn’t vote 1,431,334 ‘won’ 42/54 seats
Across England we elected the Conservatives with 41% of the vote against 31% Labour, 14% UKIP, 8% Lib Dem, 4% Green; the not-red&blue vote amounted to 28%.
What is not discussed here and which perhaps should be is the failure of referenda to produce results. Personally I blame the question posed every time. I do not understand why there must be only one question.
For example, on voting systems: do you want the system changed? Of the several possible Proportional Representation systems, do you have a preference? Would you like to indicate that preference 1,2,3,,, (of course it would have to apply some PR, wouldn’t it?).
On Europe: do you want the UK out of the EU? Do you want the UK to use the Euro? Do you agree with the stated objective of steadily increasing harmonisation? Do you want the UK to have the right to set some law in disagreement with EU government? and so on.
Just because these results would have some internal inconsistency does not make the questioning invalid. By asking more questions we, the general public, get to express more of an opinion. I am sick of the way an answer is spun to say something the answer did not say and I would like for those spun answers to have been specifically dealt with. If that means we have to have focus groups establishing what the questions should be, then I’m all for it. Let’s move politics into the 21st century and have people participating in an informed (not, please, an uninformed) way. That means that among the questions there are checks that the respondent is aware what they’re doing. That is heading to census material - well now, there’s an idea not properly used. This is, I see, beset with political correctness - and that must be dealt with, too.
DJS first draft published 20150512
the anniversary of an earthquake in Chengdu, 2008.
Numbers changed to 000,000 format 20191203, having tried to read this.
Following me posting this, the Guardian came up with this on 31st May. I imply a connection only accidentally. Connected, also, is a report form the Electoral Reform Society published today. Points made include a lot of repetition:
• This was the most disproportionate result in our election history
• 5 million votes for Green & UKIP resulted in two seats gained.
• Labour vote share increased but lost loads of seats
• Conservative majority on minority vote (well, the system is supposed to do that)
• Lib Dem 8% of vote and eight seats retained
• SNP 50% vote, damn near whitewash
• FPTP is exaggerating the political differences of the different regions and nations of the UK, leaving many citizens unrepresented.
The electoral reform society report is clearly biased towards change and must be read with that bias in mind.
The reformers want a system that produces seats in proportion to the national vote or to the regional vote. In a climate of multiple parties this is fair and just. It makes no guarantee that any MP behaves well at a local level.
I read a lot of complaint that the FPTP system has failed but nothing pointing to its strengths, nor to the weaknesses of the possible alternatives. The 2011 referendum asked us to support AV over FPTP when it could have asked whether we want reform. I continue to fail to understand why only one question is posed, as if the electorate is criminally stupid.
We may have returned to single party government for this Parliament but this does not mean a return to stability. Although it is early days, the government may well struggle to pass all its legislation with such a small majority and the Fixed Term Parliaments Act could leave it in the strange position of being unable to pass its legislative programme and yet prevented from calling another election.
The report agrees with my figures (of course it does, they’re figures). Some presentations are different, such as:
The number of MPs elected on less than 40% of votes doubled between 2005 and 2010 (55 to 111). In this election that trend went into reverse, with a mere 50 candidates elected on less than 40% of the vote. Yet some 331 of 650 MPs were elected without an absolute majority. Eight MPs won on less than 35% of votes cast, and one broke the record for the lowest winning share of the vote in UK electoral history, with 24.5%.
Worth reading is the result if cast under other voting systems, namely STV, List PR, AV and AV+. None of the tables download properly, not even if I download the pdf. You may be able to read these, but I can’t.
DJS20150601
I may need to write about different voting systems.
http://electionresources.org/uk/house.php?election=2015 summary results
I built my excel table and generated my figures from this source: http://electionresources.org/uk/2015.html United Kingdom 2015 General Election Constituency Results Data File (830 KB).
