The pros and cons of proportional representation in Britain
Denmark’s elections do not normally attract much interest in Britain. Last week’s was an exception; and I propose discussing it – though not for the reason it made the news. For the purposes of my argument it is irrelevant that the losing Prime Minister, Helle Thorning-Schmidt, is married to Stephen Kinnock.
Rather, my point is that Denmark shows how we could do politics differently in Britain - specifically, how to avoid two of the anomalies that our own election produced last month:
- The Conservatives have an absolute majority despite winning only 37% of the UK-wide vote
- Ukip and the Greens together won more than five million votes but have only two MPs, while the SNP won fewer than 1.5million votes and have 56 MPs.
The mismatch between votes and seats is not new: the Liberal Democrats have long complained about it; and Labour won an even bigger majority in 2005 with a slightly smaller share of the overall vote than the Tories achieved this time. However, last month’s contrast between the SNP and Ukip looks especially indefensible.
Now compare this with what happened in Denmark. Ten political parties arranged themselves into two loose alliances (though not always comfortably). The five right-of-centre parties won 52% of the vote and 90 mainland seats, while the five left-of-centre parties won 48% and 85 seats. Four more MPs were elected from Greenland and the Faroe islands; if they side with the Left in the new parliament, the Right will still have an overall majority of one.
Here was an election to delight supporters of proportional representation: large and small parties alike represented in parliament in line with their support; and a new government emerging from parties that, together, command an overall majority of voters. Should we adopt the same system here in Britain?
Spoiler alert: my answer later on in this blog is not a firm ‘yes’ or ‘no’, but depends on what we want an election to achieve.
Let’s start with the way Denmark’s system works. Its mainland is divided into ten multi-member constituencies that, together, choose 135 MPs. This gives a broadly proportional result, except for parties with well under 10% support, which tend to lose out. Forty more MPs are elected nationally, under a top-up formula, to ensure that the overall representation is almost exactly proportional. Thus Ms Thorning-Schmidt’s social democrats, the largest party, won 26.3% of the mainland votes and 26.8% of the seats (47 out of 175), while the smallest party in the new parliament, the Conservative Peoples’ Party, won 3.4% of the vote and 3.4% of the seats (six).
Suppose we scaled this up to a 650-seat House of Commons. We would elect 500 MPs from 37 multi-member constituencies, with a top-up of 150. Applied to last month’s vote-shares, this is how the outcome would have compared with what actually happened:
To remain in government, David Cameron would have needed to combine with Ukip’s Nigel Farage and, possibly, Northern Ireland’s unionists.
Would this be better for Britain than the House of Commons we now have? Here are some of the advantages and disadvantages Advantages:
- A Conservative-Ukip-unionist coalition parties would, just, enjoy the support of a majority of the people who voted in the general election
- For those interested in fairness, one in eight MPs would represent Ukip, and one in twenty would have represented the Greens, in line with their share of the vote
- Ukip would have ministers in a Tory-led coalition government (though not everyone will agree this is a good thing)
- Instead of virtually all of Scotland’s MPs coming from the SNP, the pro-union parties would have almost half of Scotland’s MPs, reflecting the fact that they won almost half of Scotland’s votes.
- In England, the contrast between Tory south and Labour north, greatly exaggerated by first-past-the-post, would greatly diminish. Both main parties would be well-represented at Westminster by all parts of the country.
- The main parties would have to campaign in every corner of Britain, and not just the minority of marginal seats that they are defending or attacking.
- The calculations above assume that a proportional system would have produced the same vote-shares for each party as happened last month. In practice, such a system would produce a wider range of parties (as in Denmark) and so more choice. More people could vote for a party they really believed in.
- Coalitions would be a permanent, rather than occasional, feature of British political life. Governments would often emerge from party horse-trading rather than reflect the specific choice of millions of voters.
- Ukip would have ministers in a Tory-led coalition government (though not everyone will agree this is a bad thing)
- This year, the likely coalition government might be unstable, and either unable to take tough measures – or be forced by Ukip or the unionists to do things that most voters did not vote for. (In Denmark, although the Right enjoys a narrow overall majority, the government itself is likely to be a minority administration, as the Conservative Peoples’ Party says it will not join a formal coalition.)
- An average “constituency” would have well over a million electors. The link would be broken between individual MPs and local communities of fewer than 100,000 people
- The wider choice of parties under a proportional system would mean greater fragmentation, with politicians under greater pressure to appeal to their own party’s narrow, purist base, rather than reach out to the widest possible range of voters.
- Under our present system, voters unhappy with a government’s record generally know how to vote in order to “throw the rascals out”. In a more fragmented, proportional system it can be more difficult for voters to do this, when no big party is ever able to govern on its own and one or more small parties hold the balance of power.
Here’s the bottom line. In parliamentary systems, in which the government emerges from, and is answerable to, the legislature, instead of being directly elected (as in France or the United States), there is no system that fully achieves every reasonable objective. That is, we cannot have a system that can be guaranteed to (a) be completely fair to smaller parties, AND (b) maintain a direct link between each MP and a local community AND (c) generate governments that are specifically chosen by millions of voters rather than politicians haggling after the votes have been cast.
If, despite the clear outcome of the Alternative Vote referendum four years ago, we are to reopen a national conversation about the way we elect our MPs, we should at least start from a recognition that no system is ideal and that our choice depends on our priorities. For example: is fairness more important or less important than a clear choice of government? As with so many decisions in grown-up democracies, our quest is not for perfection but the least bad way forward. I grant that’s not a terribly inspiring conclusion but, to coin a phrase, it’s the inconvenient truth.