The British single member plurality electoral system (commonly known as “first-past-the-post”) has long been a source of interest to academic analysts. Almost universally they dislike it. They like proportional representation (of one kind or another) whereas the British system is proudly non-proportional. Yet, notwithstanding its disproportionality, all of the past eight general elections (from 1979 to 2010, inclusive) have yielded the result that the biggest party in votes has won the biggest number of seats, the second biggest party in votes has won the second biggest number of seats while the third biggest party in votes has won the third biggest number of seats.
The long-standing grievance against the system has always been its regular over-representation of the two big parties (Conservative and Labour) and the equally regular, indeed grotesque, under-representation of the third party, which used to be called Liberal and is now called Liberal Democrat. However, in more recent times, academics have noticed another bias creeping in. Added to the big party robbery of seats from the Liberal Democrats, the system has shown a bias between the two big parties. The system has most favoured Labour.
The May 2005 general election supplied the statistics to prove the point. For the United Kingdom as a whole, Labour won 35% of the votes and 55% of the seats, an over-representation of a cool 20%. At the same election the other two parties were under-represented, the Conservatives to a minor extent, the Liberal Democrats grossly, as usual. However, it was the result in England which really told the story. The Conservatives won 8,102,537 votes and 193 seats whereas Labour won 8,045,874 votes and 286 seats.
The explanation for this bias lay, substantially, in the differing electoral enrolments. The biggest enrolment was 109,046 in the Isle of Wight (Conservative) and the second biggest was 85,950 in Bethnal Green and Bow (Independent). By contrast two Scottish island sets had enrolments that, by Australian standards, were absurdly small. They were Orkney and Shetland (Liberal Democrat) on 33,048 and the Western Isles seat (which now has an unpronounceable Gaelic name) on 21,576. (The Scottish Nationalists took this seat from Labour in 2005).
Conservative seats predominated at the top end of this list, Labour seats at the bottom end. A partial redistribution occurred in time for the recent general election, increasing the total number of seats in the House of Commons from 646 to 650. The academic estimate was that this would help the Conservative Party against Labour but would not entirely eliminate the bias.
Now that the results are final, I give readers the 2010 statistics. The Conservative Party won 36.1% of the vote and 305 seats. That means it won 46.9% of the seats and is over-represented by 10.8%. Labour won 29% of the vote and 258 seats. So Labour won 39.7% of the seats and is over-represented by 10.7%. These statistics tell us that the bias has, indeed, been reduced. However, it has not been eliminated. The seat share of the biggest party should be exaggerated significantly more than that of the second biggest.
The Liberal Democrats won 23% of the votes and 57 seats. That means they won 8.8% of the seats and are under-represented by 14.2%. All the rest (Greens, British National Party, Scottish and Welsh nationalists and parties in Northern Island) won 11.9% of the vote and 30 seats. So they won 4.6% of the seats and are under-represented by 7.3%. In short, the Liberal Democrats continue to be the big sufferers under the system.
Finally, the British people are to get a referendum for what they call “the alternative vote”. That would mean retaining single member constituencies but the instructions on the ballot paper would change and so too would the method of counting. The instructions would read something like that for the Penrith (NSW) ballot paper for which a by-election will be held on June 19. “Place the number 1 in the square opposite the name of the candidate for whom you desire to give your first preference vote. You may, if you wish, vote for additional candidates by placing consecutive numbers beginning with the number 2 in the squares opposite the names of those additional candidates in the order of your preference for them.”
A British move to what we call “optional preferential voting” and they call “the alternative vote” would not be proportional representation. Nevertheless, I estimate that the Liberal Democrats would have won 75 seats at the recent general election. That would be 18 seats more than they actually won. However, it would only be half the 150 seats they would have won under a strictly proportional system.
Forgive the immodesty–and I certainly do not present myself as any kind of expert on electoral systems to compare to MM–but I discussed this on the Drum on Tuesday. There is a nice graphic showing the UK Electoral Reform Society’s calculations based on a preferential voting system, and a PR-STV system; both assumed the uniformity of electorates, something that is not going to happen anytime soon though the Con-LibDems want to do so. Apparently the inequality is partly related to keeping boundaries the same as local government boundaries which obviously is a serious limitation. I do not recall if this is just a fudge by Labour because they have been a huge beneficiary; or whether there is anything in existing law about it.
abc.net.au/unleashed/stories/s2907567.htm
Oh, Malcolm Mackerras, shame on your Scots name!
The wee Western Islands seat of na h-Eilanean an-Iar is about as simply phonetic as any Celtic name could be. ‘Nuh heil-u-nun an-yar’. The islands of the west…
Pity the poor foreign psephologist wrapping his tongue around ‘Leicchardt’, ‘Berowra’ or ‘Grayndler’.
There’s also the reasoning that Labour supporters turnout at lower rates, but it matters less because they are not so congregated in marginal seats, and this inflates its seat-vote ratio.
If true, it would be an interesting interaction of culture and electoral distributions: under voluntary voting, in a safe Labour or Tory seat where a Labour vote is far from crucial, voting is less likely to be an habitual duty or symbolic act of self-expression for someone who is not middle class.
An elected upper house should test this thesis. Labour will need to turnout supporters in all regions where it hopes to win Senators or People’s Princesses, or whatever they will be called.
Doubtless the Tories will be happy to reduce the number of Scottish seats in particular. Ironic in a way for the Unionist party to have turned against federalism.
The other factor to consider in the UK electoral system is tactical voting, by which I mean many people don’t put a tick by the candidate they favour, but put a tick by the candidate who they think can head off the one they hate the most.
Most of the analysis I have read ignores this fact. So when Malcolm aggregates up all the votes for each party as one would in a truly proportionate system such as New Zealand, he hasn’t considered how these votes might change should the electors be given an alternate vote.