Electoral Reform and Socialism

Gordon Brown's speech on reforming the electoral system for parliamentary elections is overdue, but at least it nails down the government's plans after 13 years of ducking and diving on this issue. Unfortunately, the alternative vote system favoured by Brown may be a slight improvement over the current first-past-the-post system (also known as single member plurality, or SMP), but it is still far away from reflecting the democratic will of the voters. While Dave looks at Brown's motives for making this announcement now, here I will discuss some characteristics of the SMP and Alternative Vote systems, the proportional alternatives and why socialists (whether in the Labour party or not) should favour the introduction of PR.

Those who defend the current set up for Westminster elections (generally speaking, those who have most to lose in the two main parties) flag up three arguments in defence of what we have. Firstly, it's a simple system: the candidate who polls most votes wins. Second, SMP is more likely to turn out stable governments than other systems. As anyone who's a member of a small party will tell you, SMP presents a massive electoral hurdle. Effectively minor parties are locked out of the system. On the other hand, as defenders of SMP point out, this means governments do not have to rely on coalition partners who only command a minority vote. In West and later unified Germany between 1949 and 1998, the liberal Free Democratic Party was only out of government for seven years. The price of their cooperation would have met compromise on some of their senior partners' programme. Thirdly, parliament is made up of representatives elected by constituencies. This ties MPs to certain localities meaning, in theory, they have to adequately represent the interests of their locality to be returned in future elections.

In practice SMP has had a deleterious effect on politics in Britain. It has led to the creation of safe seats, meaning elections are decided by a comparatively small number of marginal seats. These tend to be identified with a socially conservative, relatively affluent, aspirational and upwardly mobile 'Middle England' by party strategists; and policies are designed to appeal to this strata. The likes of the Mail and Express are taken to be their authentic voice. In other words, the outcomes of liberal democracy in Britain is determined by a small minority.

Second the local link is overstated. While it sounds nice in theory, unless one has a rebellious Labour or Tory MP, in the vast majority of cases MPs will vote along the lines dictated by central office and enforced by the whips. Even parliamentary questions (where an MP can raise constituency-based issues) have been undermined by party discipline. You just have to watch an installment of Prime Minister's Questions to see how loyal backbench MPs effectively waste time by asking Brown if he agrees with them that the latest government initiative is the best thing since sliced bread.

Brown's proposals for reform (which will be put to a referendum should Labour win the next election) is couched very much in terms of the continuity it has with the present system. On alternative vote, he says:

The alternative vote system has the advantage of maintaining the benefit of a strong constituency link; allowing MPs to be not simply policy makers, but also community leaders, community organisers, and the strongest champions for neighbourhoods they know and love. But if the people decide to back the alternative vote, it also offers voters increased choice with the chance to express preferences for as many of the candidates as they wish. It means that each elected MP will have the chance to be elected with much broader support from their constituency, not just those who picked them as their first choice. In short it offers a system where the British people can, if they so choose, be more confident that their MP truly represents them, while at the same time remaining directly accountable to them.

Alternative vote was first introduced in Australia in 1918 for House of Representatives elections. Candidates are ranked by voters in order of preference. If one candidate has an absolute majority they win the contest. But if they do not the candidate with the least votes tends to be eliminated and their second preferences are transferred. This process is repeated until a winner with 50%+ emerges (there are variations on the theme - some elections are run that eliminate all but the top two candidates and have their second choices divided among them). While this is fairer and more adequately reflects the political mood of a constituency, it does not address the problem of marginal seats. In fact, the Australian experience has shown it strengthens the already-dominant parties because it favours their existing geographical concentrations of support. Alternative vote, though a slight improvement over SMP, will continue to replicate its problems and do little to combat politics' declining legitimacy.

What's the alternative? Socialists favour the extension of democracy in capitalist societies for a number of reasons. The more thoroughgoing the democratisation of the state, the harder it is for the state to be used as an unambiguous instrument of capital. The greater the habits of democracy are ingrained in a population, the more active so-called civil society is and the less likely wide sections of the working class fall into apathy and despondency. And the more the state comes under democratic control, the greater the pressure there is to extend democracy in the private dictatorships that run the commanding heights of the economy.

This is why socialists should support the introduction of proportional systems - not just because it improves the electoral prospects of small left groups.

Proportional representation comes in all shapes and sizes, but generally aim to express the democratic will of the voting population. For example, the PR system most UK voters will be familiar with is the D'Hondt list PR method, used for the European elections. Here parties present the electorate with a list of candidates to fill a number of available seats. Under the D'Hondt method the list with most votes wins the first seat. Their vote is then divided by two. The next seat goes to whoever now has the highest total, and so on until all the available seats are filled. This following anti-BNP video from the Green's European election campaign explains it simply:

The Sainte-Laguë method works similarly, but divides results by 1.4 and results in less proportional outcomes. The proportionality of a system can and is often distorted by local peculiarities. For instance, the European parliament elects representatives by D'Hondt on a constituency basis, allowing proportionality to be effected by geographical variations in support. One way round this would be to treat a whole country as the only constituency. This is done in Israel, but there is a 1.5% threshold parties have to cross before entering the electoral formula. Elections for The Netherlands' 150 membered House of Representatives is probably the closest to a truly proportional system - it operates with an effective threshold of 0.67% to gain a seat. An alternative to list systems is single transferable vote PR, which is proportional but retains a constituency link. The number of representatives a constituency returns is always greater than one. Therefore voters cast preferences. In the Irish Republic this can be as low as one and as high as the number of candidates. In Australia, STV proceeds by stipulating a minimum number of preferences that need be cast. Seat allocation is then decided by the Droop Quota. This is done by taking the total number of valid votes cast and dividing them by the number of seats plus one, and then adding another plus one to the resulting percentage.

For instance, in an example in his 2001 book, Electoral Systems, David Farrell cites Dublin South's 1997 election results to illustrate the Irish STV system. 57,986 valid votes were cast to determine the representatives for five available seats. These were divided by six (five plus one) giving a quota of 9,665. The first seat went to Fianna Fáil, who had won 9,904 (giving them a 239-strong surplus). None of the other parties crossed the threshold, so in the second round votes were distributed according to the second preferences of voters and seats allocated according to these results. Further rounds maybe necessary involving third and fourth preferences until all the vacancies are filled (the returning officer also has the power to eliminate last placed candidates to ensure their second preferences are transferred.

Here, proportionality is dictated by the size of the constituency. The greater the number of seats, the more likely the outcome will be proportional. STV PR is not without its problems. As defenders of the Westminster model would point out, the multiparty system in the Republic has not been as stable as the British two-and-a-half party system, leading to criticisms around weak coalition governments and instability. Second, STV PR in Ireland has led to heavy localism as TDs compete with each other to secure election next time round. This at least renders the constituency-link critique by Westminster enthusiasts null and void. These two examples of PR are far from problematic, but are much better from a socialist point of view. It is true they are more complex than FPTP but not prohibitively so. Is it really too much to ask voters to put a cross next to a party list or rank order in terms of preference? Not at all.

The constituency link argument is a red herring too. Multi-member constituencies don't have to break geographical links. Many council wards up and down Britain return more than one representative to the council chamber (though often these are not all elected at once). If voters have a problem they want looking into, they have a choice of which member to contact.
The strong government argument does not wash either. Only the UK and Barbados use the Westminster system for general elections. European countries use a mix of different PR systems, and there is no general tendency to electoral chaos (the difficulties of the Belgian and Italian party systems owe more to the specifics of those countries than the mechanics that govern their elections). In fact, for all the 'strong government' arguments used to back up the Westminster model it has, as we have seen, engendered a situation where liberal democracy faces a legitimation crisis in the face of mass "apathy" and narrowly defined politics around what plays well in key marginals. Coalition governments at least offer an opportunity for otherwise systematically excluded minorities to get their policies on the agenda.

As socialists we, to nick a Brownite soundbite, stand with the many and not the few. But our idea of the many - the working class - is very diverse and riven with all kinds of sexual, racial and sectional divisions, contradictions and interests. If we are to speak to, represent, and infuse it with a consciousness of its common interests we need to favour an electoral system that allows their free play. Neither SMP or Alternative Vote does this. 

. Edit: Good post on why electoral reform is a class issue by Stuart White here.