Friday, April 23, 2010

Plurality voting (FPTP) and proportionality


It is often claimed that FPTP is not a proportional system. Whereas this has been true in recent United Kingdom general elections, the plurality voting system can in fact produce proportional results. The United States House of Representatives is elected with a high degree of proportionality, but the price of this proportionality is a lack of real choice, only two political parties are represented in the US House.

The same is true of the House of Commons, during the 1950s when the Liberal Party was polling at 5% or lower, UK general elections were fairly proportional. Even in 1970 the Liberals only polled 7.5% of the vote. But there seems to have been a sea change in UK politics in 1974, with the Liberal Party polling 19.3% in the first election of that year, and 18.3% in the second.

The Gallagher Index is a way of measuring the disproportionality of an election. The higher the Gallagher index, the more disproportionate the election. When we plot the Gallagher index for all UK elections since 1945 against the third party (Lib or Lib-Dem) vote share of those elections, we see a clear correlation. Elections with a low third party vote share have a low Gallagher index (these tend to be elections before 1974). Elections with a high Gallagher index have a high third party vote share.



When a third party gains a large proportion of the vote it dramatically affects the disproportionality of a plurality election. This is because the winning party's share of the vote is much reduced, reduced to well below the 50% threshold that will normally give a majority. Or to put it another way, gaining 54.8% of the seats in parliament on a 49.7% vote share (as the Conservatives did in 1955) is significantly more proportional than gaining 55.2% of the seats on a 35.3% vote share (as the Labour Party did in 2005).

Ever since the 1970s the UK electorate have voted in large numbers for the third party (varying between 15% and 22% of the vote). The electoral system is fundamentally biased against third party voters, routinely disenfranchising them. In 2005 it cost 26,860 votes to elect a Labour MP, 44,306 votes to elect a Conservative MP and a whopping 96,482 votes to elect each Liberal-Democrat MP. This means that relative to Labour 4.3 million Liberal-Democrat voters are disenfranchised (or 72%).

It's time for our electoral system to reflect fairly how people cast their votes, and to stop this discrimination against third party voters.

No comments:

Post a Comment