Why FPTP (plurality) system favours Labour

David Cameron likes to say that first past the post (or more accurately the plurality voting system) favours Labour because Conservative seats have, on average, larger populations than Labour seats. To some extent this is true, it's down to demographics. Labour does well in Towns and cities, and these are becoming depopulated over time. Constituencies are drawn in 10 yearly boundary reviews from data collected from the electoral register. That means that the new constituencies that will be used in 2010 are derived from data that's already out of date. It means that the average Conservative constituency contains about 6,000 more people (not voters) than the average Labour constituency. Cameron wants to equalize constituencies and have less MPs. But that's not the whole story, the Electoral Reform Society (ERS) with Conservative Action for Electoral Reform (CAER) concluded that even with equally sized constituencies then Labour would still have gained a majority in 2005.

So what is going on? Well two things according to the ERS and CAER.

Vote Distribution

Labour has a more efficiently distributed vote. In constituencies in the south of England it is the Liberal-Democrats who are the Conservatives main threat, not the Labour Party. (e.g. see Eastbourne and Winchester). The Labour Party does not receive large numbers of votes in these constituencies. Conversely the Conservative Party and the Liberal-Democrats rack up large numbers of votes in the south in constituencies that they don't win. This is particularly true of the Liberal-Democrats. On the other hand in seats in the industrial north it is the Conservatives and Labour who are most in contention. (e.g. see Battley and Spen and Dewsbury). So the Conservatives rack up a lot of votes in the north without electoral succes, while the Labour party don't have the same problem in the south (though the Liberal-Democrats and the Tories do). Essentially the Labour vote is lumpy, it is concentrated in certain towns and cities, whereas the Conservative and Liberal-Democrat votes are thinly spread.

Turnout
According to the ERS and CAER, in 2005 turnout was a very big factor in Labour's majority. The average Conservative constituency was already about 9% bigger than the average Labour constituency. This was not enough to change the election result. But in 2005 the average turnout in Labour constituencies was 58% compared to 65% in Conservative constituences. The effective size difference between Labour and Conservative constituencies was therefore nearly 23%. 47,618 people voted in the average Conservative constituency compared to 38,739 in the average Labour constituency. Labour therefore needed a lower number of votes to win seats.
Clearly the dual problems of vote distribution and turnout cannot be addressed by David Cameron's commitment to equalizing constituency size and decreasing the number of constituencies. In fact having larger constituencies will simply mean that Liberal-Democrat votes are squeezed even more.
Full report: The Conservatives and the electoral system