Britain has seen widespread concern about the erosion of voter participation. In the 2001 general election, about 26 million people voted - 59.4 per cent of the total electorate.

Historically this was the lowest turnout recorded in British general elections since the "khaki" election of 1918. Equally remarkable, it was also the lowest in any postwar general election in any EU country.

A forthcoming study, Britain Votes 2005, found that of more than 43 million eligible electors, almost 17 million abstained on 5 May 2005. Turnout was 61.2 per cent, up 1.8 per cent on 2001, an extremely modest rise which could be attributed to many factors. Public interest may have been stimulated by strongly emotional issues debated during the campaign, notably the Iraq war and asylum-seekers.

Compared with 2001, a closer outcome was widely predicted by the final opinion polls and discussed by many media commentators. Turnout was about 10 per cent higher in marginal seats: the closer the anticipated outcome in any election, the greater the incentive for electors to cast a ballot and for parties to mobilise their support through get-out-the-vote drives.

The rise could also be attributed to reforms introduced by Parliament, notably the adoption of the rolling register and also the easier access to postal ballots, which were issued to an estimated 12 per cent of the electorate, although fewer used them.

Despite these initiatives, turnout remained well below the 76 per cent average in postwar general elections until 1997.

Can Britain learn from other nations? Electoral participation has always varied a lot among established democracies. During the 1990s, for example, elections in Iceland, Israel, and Sweden saw more than 80 per cent turning out at the ballot box, compared with less than half in the US and Switzerland.

Many factors contribute to these variations but one of the most important factors is the voting system. National legislative elections using proportional representation usually generate higher turnout than elections held under majoritarian systems such as Britain's first-past-the-post. A comparison of elections during the 1990s in 164 nations found that turnout was 70 per cent under all PR systems, 10 per cent higher than in countries using majoritarian systems like ours.

The most plausible explanation emphasises the "wasted vote" syndrome. In majoritarian systems, supporters of minor and fringe parties, such as the Greens, the BNP or Respect, have good reason to feel that their votes will make no difference to who wins in their constituency, still less to the overall composition of government and the policy agenda. This argument is strongest in safe seats where the incumbent is unlikely to be defeated.

In contrast, PR elections with low vote thresholds and with many members elected from each district, such as the party list system used in the Netherlands, increase minor parties' chances, and this increases the incentive for their supporters to participate.

British turnout in European elections remains dismal despite a regional party list system, because of lack of interest in the outcome. Nevertheless, PR for Westminster would provide a fairer system for minor parties with dispersed support, it would probably increase the proportion of women MPs, and it would probably also significantly strengthen British turnout.

Pippa Norris is the McGuire Lecturer in Comparative Politics at Harvard University. She is the co-editor (with Christopher Wlezien) of Britain Votes 2005, Oxford University Press, to be published in August 2005