The 2005 General Election: Worst Election Ever

Lewis Baston, June 2005

This report is a revised and much expanded version of our preliminary report ‘Worst Election Ever’ which was published a week after the general election.

The conclusions are an indictment of the First Past the Post system used to elect the House of Commons. The national and local outcomes amply justify our description of 2005 as the ‘Worst Election Ever’, and the indictment is spelt out in detail in what follows. However, we also hope that the research that has gone into this report is of interest regardless of the reader’s views on the issue of electoral reform.

I would like to thank all my colleagues, particularly Christine McCartney and Khadijah Elshayyal (Chapter 5), Stuart Stoner (Chapter 6), Alex Folkes (Chapter 7) and Ken Ritchie (Conclusion). The team who entered election data overnight on 5-6 May (Stuart Stoner, Ryan O’Donnell, Aly Verjee and James Osmond) and on that weary Friday (Christine McCartney, Mark Hayward and Khadijah Elshayyal) also deserve heartfelt recognition. Thanks also to Tom Carpenter of texture for the design work.

Government by a small minority – the worst election ever

Labour won an overall majority of 66 seats, or 55.1 per cent of seats, with 35.2 per cent of the vote.

No majority government in British history has ever rested on a flimsier base of public support – or, more accurately, none has since the extension of the franchise in 1918. In terms of active public consent for government, Britain is almost back in the pre-reform era of rotten boroughs. The table below gives the votes and seats for each of the majority governments in the last century which had less than 45 per cent of the vote at the outset.

The only remotely comparable election is 1922, although Labour in 2005 still polled 3 percentage points worse than the Conservatives did in 1922.1

It is notable that no election since 1970 has produced a government with 45 per cent of the vote, and that the trend in the most recent elections has been to produce significant majorities with ever lower shares of the popular vote. A Commons majority has enormous power, and this power has now been awarded on the basis of only 35.2 per cent of the vote. The case for electoral reform has become stronger with each successive election.

Labour’s share of the vote in 2005 can also be compared unfavourably to the support enjoyed in past elections by losing parties. Attlee’s share of the vote in 1955 when Eden’s Conservatives won a majority of 58, comparable to Blair’s majority in 2005, was an amazing 46.4 per cent. Blair’s winning 35.2 per cent is scarcely higher than Neil Kinnock’s share of the vote in 1992 (34.4 per cent) and less than Jim Callaghan scored in 1979 in his unsuccessful bid for a third Labour term (36.9 per cent).

The government’s level of support among voters is therefore small. But taking the electorate as a whole, the proportion of eligible people who cast a vote to return the government is extremely small – only 21.6 per cent, or 9.6 million out of an electorate of 44.4 million. In terms of votes actually cast for Labour, this is the lowest total of any post-1945 election with the single exception of 1983.

The table shows the cases in the last century when a majority government has been returned with the votes of less than a third of the electorate. Again, the 1922 election (although note that this might be a misleading comparison) and the string of recent elections since October 1974 have seen record low shares of the electorate giving support to a government. Even among this company, 2005 stands out as producing a government with exceptionally few votes. The present electoral system may allow the government to carry on regardless for a 4-5 year term, despite its low poll in 2001 and the withdrawal of enthusiasm signified by the drop in the Labour Party share of the vote of 5.5 percentage points from 2001 to 2005.

The electoral basis of British government also emerges looking shakier than in most other democratic countries. Only Turkey has a majority government with a lower share of the vote. In the table, ‘government share of vote’ is in the most recent principal legislative election, for the body most resembling the UK House of Commons. Note that some countries have a different model of government from others. Systems with an executive presidency (e.g. France, Chile, USA) may not require majority support in the legislature for government to be carried out. In Switzerland it is customary for the largest parties to form a consensus government. In Estonia and Lithuania a relatively small party may be in government but rely on support from shifting parliamentary allies to get its business through. In other situations, as in Sweden, a minority government may have the regular external support of a party which while it does not have ministers in the government is still generally supportive and is given rights of consultation etc. In Sweden the total support for Left and Green parties that regularly vote with the government takes the government bloc comfortably over 50%. In the 1970s the ‘Lib Lab pact’ was a UK parallel to this form of external support.

Other than the rather anomalous three Baltic states, Britain is clearly, after Turkey, bottom of the charts for the electoral legitimacy of the government. The nearest competitors are Canada, Croatia, South Korea and France – but in Canada the government has no overall majority and in Croatia, South Korea and France there is a separate presidential election in which the President has received majority support. In Turkey, there is a proportional system with a 10% threshold for election, which excluded all but two parties in the election.

A very low turnout

Turnout, at 61.3 per cent, recovered a little from the depths it had reached in 2001. But there is small consolation in this being the second worst since 1918. Turnout should have increased even more under the circumstances. The polls and the dynamic of the campaign suggested more of a contest than 2001, when opinion surveys found that a foregone conclusion turned many electors away from voting. There was a large increase in postal voting, which in theory should have helped turnout.The political temperature, in the new circumstances after the 2003 Iraq war, seemed higher than in sleepy June 2001, several months before the attack on the World Trade Center. And yet there was only an improvement of 2 percentage points.

In historical terms, 2005 is clearly in the relegation zone of the turnout league table of elections in the last century.The table below gives the turnout in all the elections in the last century in which turnout has fallen below 75 per cent. 1918 was affected by the war (the armistice came only a month before the poll) and disorganisation and the official turnout probably underestimates participation.

Even in relatively recent elections turnout has been much higher – 77.7 per cent in 1992 for instance. 2005’s turnout, like 2001, was about 60 per cent, in contrast to previous elections such as 1970 when a turnout of just over 70 per cent was regarded as poor. The problem of public disengagement from politics has clearly not been solved. Britain’s turnout also compares unfavourably with most other countries. It is the lowest for a national parliament among the original 15 EU states, although in most postcommunist states in eastern and central Europe there seems to be a low rate of this form of democratic participation. Britain, Canada and the USA are all notable in being English speaking (or in Canada’s case bilingual) democracies with FPTP and particularly low rates of turnout.

Unfair representation of parties

Labour’s control of the 2005 parliament is somewhat less lopsided than its dominance in the parliaments elected in 1997 and 2001. But parliament is still a grossly distorted version of what Britain’s voters chose in 2005.

Labour (and, in Northern Ireland, the Democratic Unionists) were the most successful at translating votes into seats – the First Past the Post (FPTP) electoral system gave a large ‘winner’s bonus’ to the largest single party. Once again, the principal victims of the system were smaller parties with evenly spread support, namely the Liberal Democrats, UKIP and Ulster Unionists.

The Conservatives were slightly under-represented, in that if they had won seats exactly in proportion to their vote they would have won 208.The Liberal Democrats had many fewer seats than their share of the vote among the electorate justified – an exactly proportional distribution would have given them 142. Labour were massively overrepresented: a proportional allocation would have given them 227 seats rather than 355.

Small parties with relatively evenly spread support are penalised under FPTP. UKIP was clearly the fourth party in terms of votes, but it failed to win or even come close in a single seat. Seven parties with fewer votes than UKIP, and two Independents, elected members of parliament. The Green Party stood many fewer candidates, conscious that the system was unfair to them, and many voters were deprived of the chance of choosing them. Small parties with concentrated support, like the SNP and Plaid Cymru, did better. The failure to represent small but significant points of view contrasts the Westminster Parliament with the more proportional bodies in Northern Ireland and Scotland where minorities have a voice. BNP support, as in the European election, was probably too small to elect a candidate under all but the most proportional list forms of PR, with very large regional seats, had another system been used in the general election. In no single area would they have gained enough votes to elect a candidate under the Single Transferable Vote, nor under the Jenkins Report’s system of AV+, in 2005.

The Conservatives improved their position somewhat compared to 2001, when there were 50,347 votes per Conservative MP (and 26,000 per Labour MP). This was because their share of the vote did not increase much, while they won more seats thanks to a more favourable distribution of the vote and Labour votes going to the Liberal Democrats. The Lib Dems were a little less well rewarded for their vote than in 2001, when they had 92,583 votes per MP – their vote rose across the country but did not tip over many extra marginal seats.

Wasted votes

FPTP means that a huge number of votes end up being wasted, in that they do not affect who is elected to Parliament either because they are cast for losing candidates or serve simply to swell the majority of the winning candidate. A majority of the votes in 2005 went to losing candidates, while a narrow majority went to the winners in 2001. The 2005 election saw a slight increase in wasted votes compared to 2001.

Over 70% of votes were either for a losing candidate or were surplus to the winner’s requirements. This accounts for over 19 million votes.

Electoral system bias

The electoral system is still heavily biased in favour of the Labour Party, although slightly less so than in 2001 because the Conservatives won more seats with a barely increased share of the vote. The reasons for the bias are explored below in chapter 9, but the basic statistics are given here.

The extent of bias in the 2005 result is apparent from above, but the full scale of bias is exposed when one tries to estimate the effect of changes in the share of vote.

If the Conservatives had drawn level with Labour, with each party polling the same share of the vote (33.8%), they would have gained an additional 19 seats from Labour and 3 from the Liberal Democrats, and one would flip from Labour to the Lib Dems.2 This would have meant 336 Labour seats to 220 Conservatives (assuming, for all the calculations that follow, that Staffordshire South counts as Conservative), an advantage of 116 seats despite equal numbers of votes.The Labour majority would have been 26 – not very comfortable, but more than the majority of 21 John Major achieved with a lead in share of the vote of 7.5%.

For Labour to lose their overall majority it would require a uniform 2.2% swing to the Conservatives (taking account of ‘collateral damage’ in terms of Labour losses to the Lib Dems), i.e. a Conservative lead of 1.4%.

For the Conservatives to draw level with Labour in terms of seats, overcoming the deficit of 158 seats, they would need a swing of 5.2%, i.e. a lead of 7.5% in share of the vote, if it took place entirely from Conservative gains from Labour. However, they would be helped by collateral effects which would reduce the required swing to only 4.6%, i.e. a vote share lead of 6.3%.

For the Conservatives to win outright with a majority of 2, with 324 seats, they would need 126 gains. If these were all to come from Labour this would require a swing of 8.3%, i.e. a national lead of 13.7%. Taking account of collateral Conservative gains from other parties, the target is a scarcely less daunting 7.3% swing, implying a national lead of 11.7%.

The next election will of course not be fought on the same boundaries, except in Scotland. A further detailed analysis of the boundary changes follows in Chapter 8, but in summary the changes will not amount to a substantial correction to bias – a reduction in the Labour majority from 66 to around 52. The swing required for Labour to lose their overall majority would therefore drop a little, maybe by enough to make it unlikely that Labour could poll fewer votes than the Conservatives and still have a majority. But the prospect of a Conservative majority government is pretty much as remote on the new boundaries as on the old – the swing required for them to accomplish this would be something like 7%, implying a national lead of 11%.

The Conservatives may be suffering from the bias in the electoral system, but the Liberal Democrats are still more gravely disadvantaged. This can be illustrated simply by imagining a uniform change in votes across the whole country that gave the three main parties an equal share of the vote. The table illustrates this outcome.

Minority mandates

Only 34% of MPs were elected with over half the vote in their own constituencies, the lowest proportion in British history.

This represents a marked decline since 2001, when half of MPs could claim support from a majority of their local voters. It is actually the lowest proportion of MPs with majority support ever, exceeding the previous record set in February 1974 (408 MPs out of 635, or 64.3%). At 55, the number of MPs elected with less than 40% of the vote was also substantially up on 2001 (21 MPs). 30 were in England, 13 in Scotland, 7 in Wales and 5 in Northern Ireland; 8 were Conservative, 29 Labour and 7 Lib Dem.

Nine MPs were elected with less than 35% of the vote in their constituency.

Because of the generally low turnout, as in 2001 no MPs polled a majority of the electorate in their own constituency or even came particularly close. Only three polled more than 4 voters out of every 10 registered.

At the other end of the scale, three MPs had less than 20 per cent of the electorate.

Galloway was one of few MPs elected in 2001 to have enjoyed the positive support of fewer than one in five of his electorate (19.5% in Glasgow Kelvin) and he has done it again in 2005.

Marginality and turnout

People are more inclined to vote where their vote makes a difference. Turnout in 2005 was strongly influenced by the degree to which there was a genuine contest in the seat. This table compares the turnout in the 10 seats in England and Wales with the smallest percentage majorities in 2001 and the 10 with the largest percentage majorities in 2001.3

Not only was there an enormous gap between turnout in the safe and marginal seats in 2001, but it actually became wider in 2005. Turnout in the 10 most marginal seats was 17.4 percentage points higher than in the 10 safest seats. Further, nearly a third of the entire rise among the 10 safest seats was because one of them, Blaenau Gwent, suddenly became competitive with the decision of the sitting Assembly Member Peter Law to challenge the new Labour candidate. In some inner city seats turnout remained very low, although it generally lifted a little after the nadir of 2001. The constituency with the lowest turnout was once again Liverpool Riverside (41.5%, compared to only 34.1% in 2001), followed by Manchester Central (42.0%, up from 39.1% in 2001). In 32 constituencies, turnout in 2005 fell below 50 per cent, compared to 67 in the 2001 election – but none at all in 1997.

Split votes and tactical votes

FPTP can make constituency contests a guessing game. During the 2005 election campaign various journalists and websites attempted to advise on how voters might accomplish what they considered a desirable overall result. The results were mixed. For some voters, the choice threw up a dilemma of whether to vote for the party they really preferred, or to cast a vote for the party with the best chance of keeping out the candidate they most disliked. In the now highly marginal Harlow constituency, the result was:

Labour held Harlow, albeit by only 97 votes, even though there was a sharp swing to the Conservatives in particular and the right in general. The right wing vote increased from 37.8% in 2001 to 46.1% in 2005, while the left wing vote fell from 48.8% (or 62.2% including the Lib Dems) in 2001 to 41.4% (54.0% including Lib Dems) in 2005. A large part of the reason was the change in the minor parties standing – if a left-wing minor candidate had stood in 2005, or if one fewer right wing candidate had stood, it is likely that the Conservatives would have won. But the Harlow situation was relatively unusual. More common was the situation in seats such as in Shrewsbury and Atcham where the Conservatives gained scarcely any more of the vote than they did in 2001 but still elected an MP because the Labour vote fragmented.

There was a small movement to Conservative, perhaps from UKIP, but the Conservative was elected thanks to a large swing from Labour to Lib Dem. Another extreme case was the result that unseated Chris Leslie in Shipley:

The Conservatives polled rather under what they managed in 2001, but still gained the seat by default. The fact that Shrewsbury & Atcham and Shipley are written up as Conservative gains (despite no real increase in Conservative support) while Harlow is a Labour hold despite the Conservatives achieving a real swing in their favour is a telling indication of how little FPTP results can have to do with movements of opinion. The pattern of 1997 and 2001 was that of tactical votes coalescing about the candidate most able to defeat the Conservatives. The partial unwinding of this effect has led to an increase in minority winners and smaller majorities. Seats such as Shrewsbury (and others like St Albans) end up changing hands less because of changes in opinion than because of the shifting pattern of tactical voting. The Labour Party spent a large part of the last week of the campaign trying to persuade voters that if they deserted the party for the Lib Dems, they could end up with a Tory government. While the arithmetic employed may have been shaky and there was little chance of such a movement leading to an outright Conservative victory, it did happen in some constituencies such as the ones indicated above. Of the 31 Conservative gains from Labour, only 12 were won under the Conservatives’ own steam (i.e. that the gain in the Conservative share of the vote accompanied by an equivalent fall in the Labour vote) and 19 by default because of Labour losses to the other parties.

Marginals and ultra-marginals

During the election campaign it was prominently reported in the Times (6 April) that the main parties were concentrating their efforts on the 800,000 voters who would make a difference to the result.

These were people living in the marginal seats who were judged on the basis of public opinion research to be likely to vote and also open to persuasion as to whom they might support.

These 800,000 lucky people amounted to fewer than 2 per cent of the entire electorate.

The premium that the FPTP electoral system puts on marginal seats means that votes most definitely do not have the same value. The agenda of these swing voters in marginal seats has a disproportionate impact on election campaigns and on British politics in general, and has encouraged a simplistic and sloganising emphasis on issues such as crime, immigration and interest rates from the parties. Concentration on floating voters in marginal seats, who are not necessarily representative of Britain as a whole, has distorted the entire national debate. The number of voters who matter is likely to be even smaller at the next election, because the overall result was closer, there are a larger number of highly marginal seats, and campaign techniques will no doubt have become even more sophisticated and able to target more precisely. After the election it was noted that there was a larger number of what the press quickly dubbed ‘ultra-marginal’ Labour seats than before. In the 2001-05 parliament 20 Labour MPs (4.9% of the Parliamentary Party) had majorities of less than 5 per cent of the vote. After the 2005 election there were 41 whose seats hung by such a narrow thread (11.5%).

The importance of the ultra-marginal MPs is much greater now than before, because Labour could afford to lose all 20 ultra-marginal MPs and still have a three-figure majority while the loss of 41 MPs would see Labour’s overall majority disappear. These 41 MPs will have a particular sensitivity to their constituents’ opinions and an incentive to offer the highest standards of service.

That is as it should be, and an electoral system which encouraged these standards to be universal would be a definite improvement. However, the government itself has a great incentive to focus its efforts on these constituencies as opposed to other less marginal seats.

The aggregate of the majorities in the most marginal 34 Labour constituencies is only 28,789; if 14,367 voters in these constituencies switched from Labour to the nearest competitor, Labour would be in a minority position in the House of Commons. While it is not quite that simple, because boundary changes will complicate the picture, the focus in the next general election may be so restricted that one will look back to the good old days of 2005 when as many as 800,000 voters mattered to the national debate. The Labour ultra-marginals

England

Labour have a majority of 43 (assuming that there is not a spectacular upset in Staffordshire) among the English seats, and 92 seats more than the Conservatives, despite polling slightly over 50,000 fewer votes than the Conservatives. The reason for Labour’s artificial majority is not so much that the Conservatives are under-represented; more that Labour have vastly more seats than their proportional entitlement while the Lib Dems are under-represented. The result in England does pose some problems for the future.

The Conservatives have something of a claim to ‘Speak for England’, but this must be heavily qualified. That claim rests on only 35.7% of those voting, and unless a Conservative making that claim concedes the case for electoral reform it is nothing more than a debating point. Nevertheless, the reduced English majority for Labour (43 rather than the 117 English majority the party enjoyed in 2001) could mean more occasions when Scottish and Welsh Labour MPs are called upon to vote down English rebels on English legislation.

Northern Ireland: Helping the extremes to victory

The main story of the election in Northern Ireland was the Democratic Unionist Party (DUP) landslide against the Ulster Unionist Party (UUP). The UUP was reduced to a single seat, while the DUP ended up with nine, half Northern Ireland’s Westminster MPs. The electoral system magnified the victory of the DUP and the fall of the UUP. The DUP was the luckiest party in UK politics, in that it has the fewest votes per MP of any of them, while the UUP was treated worst by the system of any party that won seats. First Past the Post also, to a much lesser extent, helped Sinn Fein in its contest against the Social Democratic and Labour Party (SDLP).

The DUP’s victory in half the Northern Ireland seats rests on only a little over a third of the votes. While the result ended up roughly proportional on the nationalist side, it gravely distorted the voice with which the unionist electorate of Northern Ireland spoke. For every 2 DUP voters, there was one UUP voter, but this 2:1 ratio of votes became a 9:1 ratio of seats. First Past the Post, not for the first time, has flattened the middle ground of Northern Ireland politics – a distorted election result in February 1974 killed off the Sunningdale agreement. We shall see whether the consequences of this election are as far-reaching. The result in individual seats was sometimes peculiar. The SDLP’s victory in Belfast South was caused by a split in the Unionist vote between the UUP and DUP (who stood for the first time here in a recent Westminster election); a nationalist now represents a majority unionist constituency. This is the reverse of what happened in West Tyrone in the 1997 parliament, when a UUP MP represented a constituency with a heavy nationalist majority split between SDLP and Sinn Fein. There was some evidence in this election of tactical voting among unionists to support threatened SDLP candidates against Sinn Fein in Foyle and South Down. The Northern Ireland local council elections were held on the same day as the Westminster election but under a different electoral system. Local authority elections, and the other non-Westminster elections in Northern Ireland, use the Single Transferable Vote (STV) system. Although there are strong similarities in share of the vote, the results in terms of seats look very different from the Westminster result.

While the trend towards polarisation is still present, it is clearly less pronounced under Single Transferable Vote. The STV results do not magnify the DUP victory over the UUP, the 3:2 share of seats reflecting the share of the votes between the two unionist parties much more accurately. Sinn Fein was the only party to win a smaller share of seats than votes, probably due to hostility from supporters of other parties, making them less likely to transfer their vote to Sinn Fein. What is also clear is that the public used their vote differently in the Single Transferable Vote election. Voters were more inclined to support Alliance and other smaller parties, receiving between them 7.9% of first preference votes and winning 6.6% of seats, knowing that their votes would not be wasted. Indeed, turnout in the local election was 1% higher than the Westminster poll, despite being held on the same day and for what is perceived as a less important public office. Westminster elections are now the only elections in Northern Ireland for which STV is not used – local and Assembly elections since 1973 (with the exception of the 1996 Forum election) and European elections since 1979 have all used STV. The outcome, particularly given the suspension of devolved government and the limited powers of Northern Ireland local authorities, means that a system that does not reflect the votes cast, encourages tactical voting and discourages moderation, gives Northern Ireland its principal political voice. If STV had been used for the Westminster election, Northern Ireland would have sent a more balanced delegation to the Commons, with probably something like 6 seats for the DUP and 4 each for the UUP, Sinn Fein and the SDLP (see Chapter 9 below). The plural views that exist within Unionism would have representation. Even in the absence of electoral reform for the whole of the UK there is a case for harmonising the system for electing Northern Ireland MPs with that used for electing every other representative body in Northern Ireland.

Scotland

The Conservatives suffered even worse treatment by the electoral system in Scotland than they did in England.They polled nearly a sixth of the vote but had only one MP out of 59 to show for their pains. With a very similar share of the vote in the 2003 Scottish Parliament elections (16.6% in the constituency section), Scottish Conservatives gained an appropriate amount of representation (18 seats out of 129).

This was because of the proportional system employed for elections to the Parliament. The Scottish Conservatives suffer because their vote is thinly spread across the country.

There are only 10 constituencies where the party was in second place in 2005 and within 20% of the winner, and only three of those are currently Labour-held seats. The electoral system is likely to continue to under-represent the Conservatives in Scotland and consign them to a minor role in Scottish politics at Westminster.

There have now been, following the wipe-out in 1997, two elections in which over a third of a million Scottish Conservatives have had to make do with one MP. Labour’s domination of Scottish representation at Westminster relies on a little under 40 per cent of the vote. Interestingly, the Liberal Democrats as the second party, with support now quite unevenly distributed across Scotland as a whole, are now doing quite well out of FPTP, getting only just under their proportional entitlement.

Wales

Like Scotland, Wales’s representation in Westminster is dominated by the Labour Party, to a much greater extent than merited by the party’s lead in share of the vote.

The Conservatives were once again severely under-represented, although they managed to gain 3 seats in contrast to being wiped out in 1997 and 2001.

Two particularly narrow gains (in Clwyd West and Preseli Pembrokeshire, by majorities of 133 and 607 respectively) were important in getting the meagre level of representation they won in 2005. It was not that Conservatives were wiped out in Wales for eight years, it was just that their voters (over a quarter of a million of them in 2001) were not represented by anyone of their choice. The Welsh Conservatives, like their Scottish counterparts, suffered from having their vote thinly spread across the country. While they were clearly second in terms of votes, with a three-point lead over the Lib Dems, they were joint third in seats, behind the Lib Dems. In the Welsh Assembly elections, for a comparable share of the vote (19.9% in the constituency section, 19.2% for the list) the Conservatives secured 11 representatives in the Assembly of 60, enough to be an effective opposition.

This was despite winning only one constituency seat; 10 of their 11 AMs are there because of the proportional top-up seats.

Unfair local Representation

There are many examples of areas within Britain where the outcome in terms of seats bears little relation to the votes that were cast.

In East Sussex (including Brighton and Hove) the Labour Party astonishingly managed to win 4 out of the 8 seats in the county despite being third with just over a quarter of the vote (25.4%).The Conservatives were well ahead in votes, with 39.8%, but won only three seats, while the Lib Dems had one seat for 26.2% of the vote.

In Cumbria, the result was extremely unrepresentative. The Conservatives were the most popular party in the county, with 38.2 per cent of the vote, but they have only one MP out of six to show for it. Labour came second, with 34.6%, and won four seats, a majority of the county’s representation.The Lib Dems had 23.1% and one seat.

At least in these areas all parties have at least one local voice in parliament. In some places FPTP can exclude parties with substantial but evenly spread local support
from getting a voice at all, and award monopolies to parties in places where there are significant opposing votes.This makes it difficulty for opposing parties to secure enough local media coverage and maintain adequate organisation to maintain a challenge; the perception that a minority party does not matter in the area is self-reinforcing.

Each main party enjoyed areas where it had a monopoly in seats but far from a monopoly in votes. In Cornwall, the Liberal Democrats won all five parliamentary seats with

44.4 per cent of the vote.The Conservatives with 31.8% and Labour with 15.9% won nothing. In Surrey the Conservatives picked up 11 out of 11 with 50.5%; the Lib Dems with 28.4% and Labour with 16.7% had nothing. In Tyne and Wear Labour won 13 out of 13 with 55.8%, compared to nothing for the Lib Dems and Conservatives on 23.2% and 17.4% respectively.

There are examples of large minorities in several areas with no representation for their parties.

There are other areas in which parties have only won token numbers of seats despite polling large numbers of votes.The Conservatives once again won extremely few seats in the metropolitan counties outside London, a total of 5 out of 124 of these urban seats, unchanged since 2001 and one fewer than in 1997. These 5 MPs were all that the
1.1 million Conservative voters in the metropolitan counties managed to elect. If the number of MPs reflected the proportion of votes cast, there would have been 30 Conservatives representing these regions.

FPTP has not only weakened the voice of the Conservatives in the big cities, it has also weakened the voice of the big cities within the Conservative Party. Its task in reconnecting with these areas is made even more difficult.

Labour have not been penalised to the same extent in any broad category of seat such as the Conservative under- representation in the metropolitan areas. Even in counties where Labour’s share of the vote is low there are often urban seats capable of voting Labour. However, Labour’s rural support is under-represented in the House of Commons despite Labour’s victories in seats such as Stroud and Ynys Mon. In the Eastern region all the Labour seats are predominantly urban, despite well over 100,000 people voting Labour in the rural East Anglian seats.

FPTP makes it more difficult for both main parties to adopt a truly national viewpoint, or even represent all the people who support them. The point has been raised in a surprising context. Iain Duncan Smith has argued that to give the final say over the Conservative leadership to the party’s MPs is to deny Scotland, Wales and the cities much of a voice in the process. This is true under FPTP, but under a reformed electoral system Conservative voters in these areas would have some MPs and not risk being completely ignored when the party takes important decisions.

Women: party representation

The overall number of women in the Commons has risen to a historic high from 119 (out of 659 seats, 18.0%) to 128 (out of a reduced 646 seats). At just under 20%, Britain remains behind most Western European parliaments, which tend to have proportional electoral systems.

With three quarters of all female MPs representing the Labour party, even the current insufficient level of women’s representation in the Commons is potentially dependent only on Labour’s majority. Should another party come to dominate at a future election, unless the proportion of women becomes more uniform across the parties, we could witness a decline in the representation of women.

Of the 119 women in the previous parliament, 83 were successful in seeking re-election, 10 lost their seats and 15 retired. 36 of the female MPs elected on 5th May are therefore new MPs. At the next general election, twenty women will be defending majorities of less than 2,000.

The Liberal Democrats and Labour each increased their number of women by four, while the Conservatives added three female MPs to their ranks. However, the impact of these gains on the composition of the party groups is not uniform.

While the total number of Labour MPs fell from 412 to 355, the number of Labour women rose from 94 to 98. At 27.5%, Labour remains far ahead of the other parties.

The Liberal Democrats increased their female representation by 5%, proportionally a greater increase than Labour, though from a lower starting point. Despite not using quotas to increase the number of women selected, the Fawcett Society calculated that 32% of candidates in key target seats were women. One sitting female MP lost her seat, Sue Doughty losing out to

Conservative Anne Milton in Guildford.The five seats in which a woman candidate won the seat for the Lib Dems saw large swings and gains in Falmouth and Camborne and Solihull were particularly unexpected.The number of Liberal Democrat women in the Commons has doubled since 2001.The party needs to build on this positive trend by ensuring it brings forward more women to contest winnable seats.

When Justine Greening took the Putney seat early in the evening, it seemed that victory had a female face for the Conservatives.Yet despite increasing their number of women MPs by three, the proportion of women in their Commons team was virtually unchanged.The three women who gained seats for the Conservatives all have slim majorities, including a wafer thin 347 for Anne Milton in Guildford. Like the Liberal Democrats, the Conservatives can not afford to leave it to chance if they want to see an increase in their proportion of women in parliament.

The Sex Discrimination (Election Candidates) Act 2002 cleared the way for political parties to use positive discrimination in the selection of election candidates, should they wish to do so. Labour is the only party to use this provision and the impact of the policy of using all- women shortlists in a significant number of safe seats with retiring incumbents has had on the number of women returned to parliament is shown in the table:

In 2005, all of the new Labour women elected in England were selected by all-women shortlists. Positive action became a heated and divisive issue in the Welsh constituency of Blaenau Gwent. Welsh Assembly Member for the constituency Peter Law resigned from the party over its imposition of an all women shortlist on the local Labour party, blocking his candidacy. He took on the official candidate, Maggie Jones and won with 58 per cent of the vote. His victory was probably as much a backlash against central control over local constituency associations as a vote of no confidence in all-women shortlists. Nationally though, the policy helped the party achieve its goal of addressing the gender imbalance in the Parliamentary Labour Party.

Both the Conservatives and the Liberal Democrats witnessed small rises in their number of women elected without using all women shortlists.The Liberal Democrats debated the use of shortlists in 2001 and voted against using them. Instead, a Gender Balance Task Force was formed to provide mentoring, training and financial support to candidates and advice to the party on removing barriers to women’s participation.

The Conservatives supported the right for parties to use quotas by backing the bill, but remain ideologically opposed to using them to select Tory candidates. Changes to candidate selection under the chairmanship of Theresa May were designed to remove all forms of discrimination and professionalise the process.The election results do not suggest that women have so far made significant gains under this new regime.

Women: regional variations

With the boundary changes in Scotland, both the number and the proportion of women returned fell. In the last parliament, 12 out of 72 constituencies (16.7%) were represented by a woman MP. Only 9 of the 59 new Scottish constituencies returned a woman MP in the new Commons, a rate of 15.3%. With fewer seats to go around, women candidates perhaps suffered in tighter competition for selections because retiring MPs were often replaced by incumbents displaced from nearby seats.The proportion of women representing Scotland was already behind England and now slips behind Northern Ireland and Wales.

The number of women returned by Scottish voters to Westminster is in stark contrast to the 40 per cent representation of women in the Scottish Parliament, achieved through proactive measures by the parties and the opportunities provided by a new institution with the absence of incumbents (and incidentally a more proportional electoral system).The SNP did not return any women in their six seats and lost Annabelle Ewing, MP for Perth in the previous parliament but displaced to a less winnable seat (Ochil and South Perthshire) in the boundary changes.

Wales, on the other hand, more than doubled the number of women, from 3 in the last Parliament to 7 out of 40 seats. However, this still only represents 17.5% of Welsh MPs, and again is far behind the 50% rate in the Welsh Assembly.

Northern Ireland again returned three women, with the Ulster Unionist Sylvia Hermon being the only candidate from that party to hold her seat.Though the re-election of three women out of 18 MPs (16.6%) marks a step forward for Northern Ireland (the region did not return a single woman for 30 years prior to that), there was little prospect of more women being elected; in no other constituency did a woman candidate come close. While the proportion of women returned to the Northern Ireland Assembly (elected under the Single Transferable Vote) is only slightly higher, due to multi-member constituencies, 13 of the 18 constituencies are represented by at least one female MLA.

Ethnic minority representation

The 2005 election saw a small increase in the number of Black and Minority Ethnic MPs from 13 to 15. If the make up of MPs was to reflect the population in the country (7.9% BME), there should be 51 MPs from minority ethnic backgrounds, demonstrating that the current level of representation is still seriously inequitable.

Ten sitting candidates successfully defended their seats, one retired and was replaced by another black MP and two sitting MPs were unsuccessful in seeking re-election.

The Conservatives elected their first black MP (Adam Afriyie in Windsor) and an Asian MP (Shailesh Vara in Cambridgeshire North West). Both were selected as candidates in safe seats where the incumbent was retiring, without using positive discrimination. The Liberal Democrats have lost their only ethnic minority MP. Parmijt Singh Gill, who won the Leicester South by-election in 2004, saw the seat pass back to the Labour Party after just

ten months in parliament. The Labour Party has increased its total number of ethnic minority MPs by one to 13.
Three new candidates were elected: Sadiq Khan, Shahid Malik and Dawn Butler, who replaced retiring MP Paul Boateng. Oona King was unsuccessful in her bid to hold her Bethnal Green and Bow seat.

Labour still has the only two black women MPs – Diane Abbot and Dawn Butler.The party will also be justifiably disappointed at the failure of Yasmin Qureshi to re-take Brent East, lost to the Liberal Democrats in a 2003 by- election. Had she been returned, she would have been Britain’s first Muslim woman MP. The policy of all-women shortlists has come under fire from some ethnic minority activists who argue that it does not help ethnic minority MPs to get elected, although there is nothing inherent in the process to disadvantage ethnic minority women.

One third of ethnic minority MPs represent London constituencies. Most others are in large towns and cities, all with minority ethnic populations above the UK average.
The two Conservative ethnic minority MPs, and Labour’s Parmjit Dhanda (Gloucester) and Ashok Kumar (Middlesbrough South and Cleveland East) are exceptions. Scotland has only one Asian MP while Wales has none. It is also worth noting that with two exceptions, each of the 15 BME MPs were elected in safe seats where they were either defending a comfortable majority or replacing a retiring incumbent.The exceptions were Gloucester, where first-time incumbent Parmjit Dhanda achieved a small swing in his favour, and the idiosyncratic Bradford West. Many other BME candidates were put forward in seats where the party had little chance of winning.

Whilst both the Liberal Democrats and the Conservatives put forward more BME candidates, fielding 40 and 41 BME candidates respectively, several were contesting seats with a sitting ethnic minority candidate. For example, Brent South saw Dawn Butler, James Allie and Rishi Saha contesting for the same seat. As some of the BME candidates were competing against each other, this placed limits on the number which could be elected.

The Electoral Reform Debate

In this section we examine the role that the issue of electoral reform played during the 2005 election campaign – the positions of the main political parties on electoral and some other constitution reform proposals, and the results of the Electoral Reform Society’s candidate questionnaire which was distributed during the campaign. We also look at the efforts of the electoral reform campaign during the election, particularly the Make My Vote Count campaign, before considering the reactions to the election result and how the campaign might proceed during the 2005 parliament.

Party manifesto commitments

The Conservative Party promised to cut the number of MPs by 20%, and to make the House of Commons ‘more capable of standing up to the executive’ by strengthening select committees and ensuring there was time for ‘proper scrutiny of all legislation’.The party also promised to ensure that exclusively English matters would be decided without the votes of MPs sitting for Scottish constituencies, as well as promising to seek cross-party consensus for a substantially elected House of Lords.The party made no mention of electoral reform – a subsequent statement said: ‘The Conservative Party believes that proportional representation undermines democratic accountability, weakens the link between electors and their representatives, and gives a disproportionate influence to minority parties.’

By contrast, other opposition parties all continued to back calls for electoral reform. The Liberal Democrats reiterated their backing for the Single Transferable Vote system, promising to ‘extend this fair voting system to all local elections in Britain, and to the House of Commons, Scottish Parliament and National Assembly for Wales’. The party also promised to review the European electoral system ‘so people can choose their MEPs personally, rather than just vote by party list’, and to institute a predominantly elected second chamber.The Liberal Democrats additionally argued that ‘at the age people can marry, leave school and start work, they will have the right to vote’.

Likewise, the Green Party reiterated their support for proportional representation in all elections, and for an elected Upper House. The Scottish National Party and Plaid Cymru also supported fairer voting systems at all levels – locally, nationally, and at Westminster.

Meanwhile, considerable interest surrounded the Labour Party’s proposals, after media speculation that the proposed review of electoral systems, as promised in the 2001 manifesto, had been dropped. The 2005 manifesto, however, contained this statement:

Labour remains committed to reviewing the experience of the new electoral systems – introduced for the devolved administrations, the European Parliament and the London Assembly. A referendum remains the right way to agree any change for Westminster.

The manifesto also set out the party’s policy on House of Lords reform, that ‘Labour believes that a reformed Upper Chamber must be effective, legitimate and more representative without challenging the primacy of the House of Commons’. Labour promised that, following a review conducted by a committee of both houses, they would seek to codify the powers of the Lords, and place ‘reasonable limits’ on the time Bills spend in the chamber. Moreover, the Party promised to remove the remaining hereditary peers ‘and allow a free vote on the composition of the House’. No mention was made of reducing the voting age.

Views of candidates

Having noted the platforms on which the parties would be standing, the Electoral Reform Society decided to ascertain the views of individual candidates. During April and May, the ERS sent a questionnaire to over 2000 candidates in all constituencies across the UK. The questionnaire asked:

• Would you support a review of the system of election for the House of Commons (as promised in the Labour manifesto in 2001)?
• If there were to be a referendum on changing the electoral system for the House of Commons, would you personally support a change?
• What is your preferred system for electing the House of Commons?
  a) First Past the Post ‘FPTP’ (as now)
  b) Alternative Vote plus ‘AV+’ (as recommended by Jenkins)
  c) Additional Member System ‘AMS’ (as used in Scotland and Wales)
  d) Single Transferable Vote ‘STV’ (as used in Northern Ireland)
  e) Alternative Vote ‘AV’
• Would you support a predominantly elected replacement for the House of Lords, if the alternative were to be a mainly or wholly appointed House?
• Do you support a reduction in the voting age to 16?

Around 500 questionnaires were returned, a healthy return given the amount of paperwork that an average candidate is required to deal with during the General Election campaign. Encouragingly, over 100 replies were received from candidates who were eventually returned as MPs. Although by no means comprehensive, this survey does provide a snapshot into the views of current MPs on electoral reform.

The questionnaire was completed by MPs roughly in proportion to the party strengths in the House of Commons, although one cannot claim that it is more broadly representative of opinion within each party or the Commons as a whole. It may well have been strongly influenced by self-selection, in that MPs who were most interested in (or having the strongest views for or against) electoral reform would have been most inclined to complete it.

From the questionnaires completed by successful candidates, it was found that:

• 55% were in favour of a review of the system of election for the House of Commons
• 32% were certain to support a change in the voting system, with a further 21% willing to consider their position depending on the system proposed. 30% said they would oppose any change
• When asked to indicate their preferred voting system, 36% said they favoured First Past the Post.
• 44% preferred another system. The most popular alternative was STV, which was favoured by 20% of respondents. 13% chose AV+, as proposed by the Jenkins Commission, as their first preference, and 7% AV.The AMS system, as used in Scotland and Wales, was the least popular of the options, with just 4%.
• 62% were in favour of a predominantly elected House of Lords. Only 17% were definitely opposed.
• Opinion was split on lowering the voting age. 39% favoured a change, whilst 40% were opposed.

The views of Conservative and Liberal Democrat respondents largely reflected their own party’s policies as regards electoral reform. In line with their manifesto commitment, all Liberal Democrats were in favour of a review of the electoral system, and were personally in favour of a change.The vast majority were also in favour of a predominantly elected Lords, and supportive of lowering the voting age.

Amongst the Conservatives, the vast majority were opposed to a review of electoral systems, and by extension favoured First Past the Post as the best system. However, a small number of Conservatives did back the call for a review of electoral systems, admitting that ‘the process of selecting Members of Parliament needs review’.

Conservative opinion was more divided on Lords reform. Half the respondents backed the party line seeking a consensus around a predominantly elected House, yet as many as a quarter voiced their opposition to current Party policy. On lowering the voting age there was far more of a consensus, with the majority opposing any change.

The debate therefore appears to be at its keenest amongst Labour MPs, where the dichotomy of views reflects the ambiguity of the Party’s manifesto commitment.The following trends were identified:

• 60% of Labour respondents explicitly favoured a review of electoral systems, whilst just 16% were actively opposed to any such review.
• 30% of respondents were certain to favour a change, with a further 26% willing to consider their position depending on the options put forward. Only 16% were certain to oppose any change.
• On the other hand there is no clear picture which system Labour MPs favour. Of all the options put forward, First Past the Post has marginally the most backing (22% indicated it as their first preference). However, as many as 46% actively favour another system. The most popular alternative is AV+, which is favoured by 20% of MPs.

Meanwhile, there is a strong movement for Lords reform amongst Labour’s representatives. 56% favour an elected House, with only 18% opposed. Opinion on lowering the voting age is split down the middle, with 34% for and 34% against, with the remainder not expressing a settled view either way.

The electoral reform campaign before the election

The candidate questionnaire campaign was just one element of a wider campaign by reformers to draw attention to the flaws in the electoral system. During the election campaign, the Make Votes Count coalition, including the Electoral Reform Society, set up the Make My Vote Count interactive website, www.makemyvotecount.org.uk, to campaign for a referendum on the voting system. Users had the option to sign an online petition, email constituency candidates to ask their views on electoral reform, download campaign materials, access on-line briefings, and follow the campaign in a daily blog.The site attracted thousands of users over the four weeks of the campaign, and proved a most useful resource for all those following the debate.

The campaign itself highlighted the need for electoral reform. On 6 April, The Times published an article entitled ‘The Hidden Election’, which reported that Labour and the Conservatives were spending millions of pounds on a high- tech campaigning targeting just 800,000 voters in marginal seats who would settle the General Election result.This story created a swarm of publicity in the broadcast and print media, and the Make Votes Count coalition ran a publicity campaign in a number of ‘safe seats’ in Greater London, which served to emphasise that the vast majority of votes in these constituencies were taken for granted. Make Votes Count also facilitated a series of successful street stall campaigns, where in towns and cities across the UK, campaigners ran street ballots, asking passers-by whether they thought their vote would count.

Much of the pre-election media focus was on the concerns over turnout and the security of the voting system, not least with regard to the increased prevalence of postal votes.The electoral reform campaign was at pains to point out that an electoral system where thousands of votes were taken for granted was just as great a cause for concern, and indeed was a major contributory factor to the problems of democratic engagement that have been encountered.

The electoral reform debate in Parliament and the media since 5 May

The General Election itself was one of mixed emotions for the electoral reform movement. A number of active members of the All-Party Parliamentary Group on Electoral Reform lost their seats, and other active supporters, such as Paul Tyler (LD, North Cornwall) retired.

However, the disappointment at losing some valued colleagues was tempered by the knowledge that, in all but one of the above cases, the victorious candidates were themselves electoral reformers. Indeed, with both the Liberal Democrats and the Scottish National Party increasing their representatives, there was a net gain of 11 for the parties backing proportional representation as official policy. Amongst the questionnaire responses, nearly 20% were received from new MPs, of which nearly three in four were either explicitly in support of a review of electoral systems, or at least open to persuasion. A number of other new MPs have also expressed support for the All- Party Parliamentary Group on Electoral Reform, just weeks into the new Parliament. This high proportion of engagement amongst the new intake raises the hope that the pressure for electoral reform will be maintained through the new Parliament.

A big impetus to this came from the General Election result itself. As our analysis has shown, the 2005 General Election has highlighted the iniquities of the First Past the Post system more than ever before. During the Queen’s Speech debate, many MPs have pointed out the disparities in the system. A number of Conservative MPs echoed the statistic, uncovered first by the Electoral Reform Society on 6 May, that the Conservatives won more votes than Labour, and yet won 93 fewer seats. Likewise, parliamentarians in both Houses have pointed out with regularity that Labour won the lowest vote share of any majority government in history. Witness Charles Kennedy in the debate on 17 May:

There is another wider reflection that is not just for Liberal Democrats. It has already been a matter of some discussion within both the Labour party and the Conservative party, and quite rightly in the latter case as the party that out-polled the Government in England at this general election. The Government secured only 36 per cent. of the votes cast and a little over a quarter [sic.] of those able to vote. Surely there is a lesson for us all here, because each and every party now must reflect on the fact that to greater or lesser extents we are competing minorities across the country.

The related issue is, of course, the voting system itself. I mentioned the experience in England of the Conservatives, but despite out-polling Labour in England they have no representation in any of the major English cities outside London. Is that healthy for the geographical representation of the body politic? The result of this election yet again underlines that first-past-the-post voting in this day and age is a redundant system that belongs to a redundant age of British politics. British society is more fluid and party allegiances are much less fixed, and our society as a whole is more aspirational as a result.

Therefore, the time has surely come to revisit—as the Prime Minister promised but reneged on when first elected Prime Minister—the whole issue of voting reform in this country. It is no longer a question, as in days gone by, of just those in the Liberal tradition feeling particularly hard done by; there are now losers in every party and in every part of the country. That is why the duty is surely on us all to look at this issue afresh.

Conservative John Redwood echoed the point:

Labour must feel worried that it is only by a quirk of the distribution of constituency boundaries in the electoral system that it is in such a strong position in the House of Commons, with an overall majority and possible support from some of the leftward-inclined parties on a number of crucial issues… It is extraordinary that a Government with a good majority could have achieved that on only 36 per cent. of the vote, and that that in turn was based on a low turnout of around only 60 per cent.

Yet (see Chapter 9 below), the fundamental problem lies not with the distribution of constituency boundaries, but rather with the electoral system itself – a system not equipped to deal with the pluralist politics that is reflected in the way the UK voted in 2005.

This much was recognised by The Independent when it launched its ‘Campaign for Democracy’ on 10 May, stating that ‘thoroughgoing reform of the electoral system needs to be embarked upon now, before faith in the system is lost entirely. It is urgent that the discussion of a proportional representation system for British general elections be revived.’ Over the past month, The Independent, as well as other newspapers and media sources, have continued the debate. The Independent launched a petition, inviting the Prime Minister ‘to institute urgent reform of our voting system so that the British people are encouraged to believe that their votes count and that the result of a general election is more representative of their wishes.’ So far over 30,000 have signed up. For more information on the campaign please see http://news.independent.co.uk/uk/politics/story.jsp?story=637 826.

In addition, columnists such as Polly Toynbee and Jonathan Freedland in the Guardian, and Peter Riddell of the Times made the case for reform, or at least a serious review of the electoral system.The letters pages of all quality newspapers in the UK engaged with the debate, as did numerous radio and television programmes, most notably BBC Radio 4’s Today Programme. Meanwhile, a Make Votes Count meeting held in the House of Commons after the General Election, when a panel comprising Labour MP Martin Linton, Liberal Democrat peer and chief election strategist Chris Rennard, Polly Toynbee and singer/ political activist Billy Bragg debated the case for reform, attracted nearly 300 attendees. As Toynbee herself observed in The Guardian on 13 May:

But in one vast Commons committee room, shortly after the prime minister had addressed his troops on Wednesday night, a great surge of hundreds packed in to protest at such tyrannical abuse of democracy. How did the prime minister gain such power with a mandate of just 22% of the electorate, the lowest since the Great Reform Act of 1832? They were hanging from window ledges, jam-packed against the walls, squatting in rows behind the platform: ‘Never seen a public meeting like it in the house!’ declared Lord Lipsey, chairing the make-votes- count meeting. Peter Tatchell called for a new Chartist-style uprising. An orotund Tory in pinstripes boomed out that he never expected to see the day when he would stand shoulder to shoulder with Tatchell and cry revolution. No taxation without representation!

The democratic shortcomings that the General Election so clearly revealed have generated a wealth of publicity that has been impossible to ignore. The Prime Minister, when questioned about electoral reform at his monthly press conference on 12 May, stated:

I know the Independent is getting very interested in this debate, but the trouble with any electoral system is that it can lead to different types of results in which people claim are unfair for different types of reasons. But the problem with PR systems is that you can often have a result where a small party actually holds the balance of power, and that is unfair as well. I have also got to give you my frank view, that if this was a straight Labour/Tory fight, if people said right the only way is Labour or Tory, I don’t think there is any doubt what the outcome of the election would have been, and I think if it had been Labour versus Lib Dem it would have been the same, and it might even have been, if it had been Tory versus Lib Dem, they would have gone Tory. I don’t think you can tell, in other words, that the existence of the third party with this amount of vote means that people were actually undecided about the government.

Lord Chancellor Lord Falconer backed up the Prime Minister’s judgment when he told the House of Lords on 23 May that the government ‘have no plans to introduce legislation on changing the current electoral system’.

Yet electoral reform remains on the agenda. On 26 May, Labour peer and Chairman of Make Votes Count, Lord Lipsey, tabled a debate examining ‘the workings of the British electoral system in the 2005 General Election’. In the debate, Lipsey condemned the first past the post system that so allowed the Government’s democratic legitimacy to be called into question, given that barely over a third of votes were actually cast for Labour.

In responding to the debate, the Department for Constitutional Affairs Minister, Baroness Ashton, referred to the review of electoral systems, as promised in the Labour manifesto:

We have an internal piece of work under way which will feed into a new ministerial committee on constitutional affairs, the composition and terms of reference of which we have ensured were made available on the website this morning…

On how people will become involved in and consulted on this exercise. I do not yet have an answer because the new ministerial sub-committee has not yet met. I am not able to pre-empt its deliberations. We have indicated that a referendum would be the way forward and we remain absolutely 100 per cent committed to that.

The issue has remained on the agenda. In a written answer on 7th June, Department for Constitutional Affairs Minister Harriet Harman asserted that:

The Government remain committed to their manifesto commitment to reviewing the experience of the new electoral systems – introduced in the devolved administrations, the European Parliament and the London Assembly. The Government have no plans to change the voting system.

In answer to an oral question tabled by Labour MP John Grogan the same day, Harman added:

We have embarked on a considerable programme of voting system reform, as well as constitutional reform, and it is right that we look at the experience of the new proportional voting system and understand the effect of the electoral reform that we have already achieved.

We are undertaking a review, and… there is a balance to be struck between proportionality and simplicity. We want a fair system where there is a strong connection between the elected representative and their constituents, but in the new voting systems that we have introduced in the devolved Assemblies, and for the Mayor and the European system, there is an element of proportionality and we are reviewing those experiences.

On 14th June, Liberal Democrat MP John Barrett tabled a debate on electoral reform. In responding to the debate, Harman acknowledged that the subject ‘is hugely important’, and that ‘there are many deeply held views, not only in the House but outside… we are not the only ones who feel passionately about the subject and who want to engage in the debate’:

Many people are concerned to be involved in the debate more widely, and you ask whether we are prepared to listen and to respect the views of those who participate in it. The answer must be yes, not least because, as he said, we must take a principled approach to a debate about improving democracy… If there is one thing that the voting public would run a mile from, it is the idea that, having got into power, we would rig the system in our favour and that party political interest would somehow override the principled interests. We must always be vigilant about that.

She also agreed that the central point is to identify ‘the best system that commands support, has legitimacy, is seen to be fair and simple, and yields what we want, which is a representative Parliament that can hold to account effective Government’, and that ‘it is only right, having introduced such a large range of new electoral systems, that we should take stock of the effect of those systems and how they have worked’:

That is why we are doing what we are said in our manifesto that we would do: reviewing the experience of the new electoral systems that were introduced for the devolved Administrations. We also said that if we were to recommend a change in the system of election to the House, there would have to be a referendum to agree the right way forward. At the very least, we need to review the effect of the current systems while we discuss issues such as public confidence in politicians, availability of information to citizens and accessibility to the vote.

The government’s review of electoral systems

Initial details of the review of electoral systems have indeed been announced, as part of the remit of the Cabinet Sub-Committee on Electoral Policy (CA(EP)), and the Ministerial Committee on Constitutional Affairs. The terms of reference of the review are, as Baroness Ashton stated, ‘to co-ordinate the Government’s policy on electoral issues and the democratic process; and report
as necessary to the Ministerial Committee on Constitutional Affairs’.

The Sub-Committee will be chaired by the Deputy Prime Minister, John Prescott, and its other members are:

• Alistair Darling, Secretary of State for Scotland
• Geoff Hoon, Lord Privy Seal and Leader of the House of Commons
• Hilary Armstrong, Parliamentary Secretary to the Treasury and Chief Whip
• Charles Clarke, Home Secretary
• Peter Hain, Secretary of State for Northern Ireland and Secretary of State for Wales
• Ian McCartney, Minister without Portfolio
• Lord Falconer, Secretary of State for Constitutional Affairs and Lord Chancellor
• John Hutton, Chancellor of the Duchy of Lancaster (Minister for the Cabinet Office)
• Des Browne, Chief Secretary to the Treasury
• Douglas Alexander, Minister of State, Foreign and Commonwealth Office (Europe)
• Phil Woolas, Minister of State, Office of the Deputy Prime Minister

The terms of reference of the Ministerial Committee itself, to which the Sub-Committee will report, are ‘to consider strategic issues relating to the Government’s constitutional reform policies including House of Lords reform and issues arising from devolution to Scotland, Wales and Northern Ireland’. The Ministerial Committee includes many of the above members as well as:

• Gordon Brown, Chancellor of the Exchequer
• Jack Straw, Foreign Secretary
• David Blunkett, Secretary of State for Work and Pensions
• Margaret Beckett, Secretary of State for Environment, Food and Rural Affairs
• John Reid, Secretary of State for Defence
• Baroness Amos, Leader of the House of Lords and Lord President of the Council
• David Miliband, Minister for Communities and local Government
• Lord Grocott, Chief Whip, Lords
• Lord Goldsmith, Attorney General
• Lynda Clark, Advocate General

The Sub-Committee Chair, John Prescott, has admitted in the past that he has no interest in reform, and the Sub- Committee as a whole is conspicuously lacking in reformers. It therefore does not reflect the plurality of views that exists within the Labour Party. Although some members of the group have expressed interest in other systems such AV from time to time, supporters of change in the Cabinet such as Patricia Hewitt and Alan Johnson have been excluded. We await the first meeting of the Sub- Committee – meanwhile the debate will go on in Parliament and beyond.

The announcement of the members of the Sub- Committee follows the announcement of the new Cabinet. There has been a reshuffle of Ministers in the Department for Constitutional Affairs (DCA), the Government department responsible for electoral law. Lord Falconer remains in post as Secretary of State for Constitutional Affairs and Lord Chancellor. His team now includes Harriet Harman as Minister of State, with responsibilities including House of Lords reform and reform of electoral practice.

Bridget Prentice has been appointed Under-Secretary of State, whilst Baroness Ashton remains as Under-Secretary in the House of Lords. In addition, the Government announced on 23 May that the DCA was to assume further responsibilities. Policy on the conduct of local elections was transferred from the Office of the Deputy Prime Minister with immediate effect, and the DCA will also take on the sponsorship of the Parliamentary Boundary Commissions for England and Wales.

In other reshuffle news, former Defence Secretary Geoff Hoon has been appointed as Leader of the House of Commons, with Nigel Griffiths as his deputy. Baroness Amos remains Leader of the House of Lords.

The Conservatives and the Liberal Democrats have also reshuffled their teams. Oliver Heald remains as Shadow Secretary of State for Constitutional Affairs, whilst Chris Grayling will face Geoff Hoon as Shadow Leader of the Commons. David Heath is the Liberal Democrat Constitutional Affairs Shadow.

House of Lords reform

The prospects for House of Lords reform in the immediate future seem brighter. Both the Conservatives and the Liberal Democrats maintain their manifesto commitment to a predominantly elected second chamber. Meanwhile, Labour had committed itself to setting up a Joint Committee to conduct a review on the powers and composition of the Lords. Statements by the Lord Chancellor and other Government representatives have made clear that there will be no developments on composition until the powers of the Upper House are codified. This commitment, set out in the manifesto, has since been repeated in a Parliamentary Written Statement on 23 May.

The proposals suggested – namely ‘developing alternative forms of scrutiny that complement rather than replicate those of the Commons’, and to ‘explore how the upper chamber might offer a better route for public engagement in scrutiny and policy-making’ – have raised concerns amongst opposition politicians. In particular, the manifesto commitment to ‘legislate to place reasonable time limits on the time bills spend in the second chamber – no longer than 60 sitting days for most bills’ has been interpreted by some as an attempt to emasculate the Upper House’s power. In reality, such a sanction would rarely have needed to be applied in recent times as the vast majority of Bills complete their passage through the House in under 60 days. Nonetheless, a number of peers have resented the attempts to place limits on their powers of scrutiny.

A number of these points were raised in the Queen’s Speech debates. In addition, a number of peers raised objections to the Government’s proposals to remove the hereditary peers.There were concerns that any move to do so without clear agreement on who would replace them would constitute a breach of the ‘Weatherill agreement’, whereby the last 92 hereditary peers were retained in the House of Lords as a longstop until the final composition of the House was agreed.

The debate has also served to demonstrate the wide range of opinions on the best way to achieve progress on composition. Whilst some peers continued to call for a directly elected House, others defended the system of appointments as delivering a level of expertise amongst the House’s members that could not otherwise be achieved.
Still others defended the hereditary concept.

The ongoing debates show that consensus has yet to be achieved on what the future House should look like. On the other hand the prospects for a ‘yes’ vote in the Commons for an all-elected House look brighter. With both of the Opposition parties committed to such an outcome, and a large number of Labour MPs favouring this option, it would seem likely to be the most popular option.

Certainly the results of the Electoral Reform Society’s
survey, indicating 62% backing for a predominantly elected House, would back this up.Yet as the responses of Conservative MPs in particular show, there is not necessarily a consensus even amongst parties with a manifesto commitment to elections. When the Commons last voted on Lords reform in 2003, the majority of MPs
voted for one of the ‘majority elected’ options, and yet they could not agree on any of the options. A consensus seems to be slowly building that a predominantly elected House is the only way to proceed – reformers will need to avoid the mistakes of the past to ensure that this happens.

Scotland – The Arbuthnott Commission

Labour’s manifesto made little mention of the Scottish constitutional situation:

In our first term, we devolved power to Scotland and Wales and restored city-wide government to London. Britain is stronger as a result. In the next Parliament, we will decentralise power further.

However the main point of interest over the coming months concerns the findings of the Arbuthnott Commission on Boundary Differences and Voting Systems, which was set up by Alistair Darling, Secretary of State for Scotland, in July 2004. The Commission was set up after the decision to adopt STV for local elections in Scotland from 2007, which meant that there will be four electoral systems in operation: First Past the Post for House of Commons elections, the Additional Member System for the Scottish Parliament, a closed list PR system for European elections, and STV at local level.

The Commission is chaired by Professor Sir John Arbuthnott, former Principal and Vice-Chancellor of Strathclyde University, and currently Chairman of Greater Glasgow NHS Board.The Commission’s remit is to examine the consequences of having four different systems of voting in Scotland, and different boundaries for Westminster and Scottish Parliament constituencies, for:

• voter participation;
• the relationship between public bodies and authorities in Scotland and MPs and MSPs;
• representation of constituents by different tiers of elected members.

The Commission has been asked to make recommendations on whether these consequences require action to be taken in respect of:

• arrangements between elected representatives, to ensure that constituents and organisations receive the best possible service;
• the pattern of electoral boundaries in Scotland;
• the relationship with other public bodies and authorities in Scotland;
• the method of voting in Scottish Parliament elections.

The Commission launched a consultation document on 18 January, and has received evidence from all the major
parties. Labour has reiterated its ‘fundamental support for First Past the Post as an electoral system’ while at the same time recognising ‘the strong desire stated in the Commission’s consultation to respect the devolution settlement and the need for any alternatives to the current system to retain significant proportionality’. Labour has stressed the importance of the constituency link, but at the same time has pointed to some confusion amongst voters with regard to their two votes under the current AMS system for the Scottish Parliament. They have suggested that the second vote could be abolished. Labour is also concerned about the different roles and responsibilities of ‘constituency’ and ‘list’ MSPs.

The Liberal Democrats continue to back STV as the best system, and have called for its adoption for Scottish Parliamentary elections. The Scottish National Party agreed, stating:

The Party’s backing for STV arises from the fact that it is a system that aims for a broad degree of proportionality, while also maintaining a link between constituents and elected representatives. STV allows, in larger multi-member wards, the opportunity for representation for a range of parties and individuals. Unlike the closed list system used currently, it allows voters to select their favoured candidate from within parties. Additionally, preference voting allows voters to order the favourability they feel towards candidates and parties, in a way that AMS, with its single ‘X’ does not.

The Conservatives set out their fundamental support for First Past the Post, but, like Labour, recognised the importance of proportional representation in the post-devolution settlement. Given this, the party favours retaining the AMS system for Scottish Parliamentary elections.

The consultation period closed on 10 June, and the Arbuthnott Commission hopes to report by the end of the year.

Wales

The Labour manifesto was more specific with regard to Wales:

In Wales we will develop democratic devolution by creating a stronger Assembly with enhanced legislative powers and a reformed structure and electoral system, to make the exercise of Assembly responsibilities clearer and more accountable to the public.

On 15 May, the Government released its ‘Better Governance for Wales’ White Paper. The White Paper has proposed legislation over three key issues:

• Enhanced legislative powers for the Assembly through a streamlining of procedures so that its legislative priorities are secured more quickly and easily.
• The way forward for a formal separation between the assembly and the Welsh Assembly Government to avoid confusion and to improve effective scrutiny.
• A proposal to change the provisions in the Government of Wales Act so that losing candidates in Assembly constituency elections are unable to become Assembly members through the regional list under the Additional Member Voting System.

The Government claims that ‘the White Paper proposes gradual, staged moves which could lead to full legislative powers for the Assembly. Where last year’s report of the Richard Commission envisaged greater powers for the Assembly from 2011, under the White Paper proposals Labour will legislate for streamlined powers by 2007.’ Any change to full law-making powers for the Assembly would only follow a referendum. Plaid Cymru expressed disappointment that the proposals fell short of granting full law-making powers to the Assembly as the Richard Commission had proposed.

The White Paper has also failed to adopt Lord Richard’s recommendation that STV be adopted for Assembly elections, as ‘The Government does not agree that this would be the right way forward’. Instead, Labour have proposed legislation to deal with the so-called ‘Clwyd West question’, whereby losing candidates in constituency contests have gained election as a list member.The Government proposes that candidates be barred from standing both for a constituency and a list seat.

In responding to the White Paper, the Electoral Reform Society expressed disappointment that the Government had proposed to change just a single aspect of the voting system, whilst ignoring Lord Richard’s recommendation for a complete overhaul of the system by adopting STV.

The White Paper is open for consultation until 16 September.

Integrity and Security

The UK entered the 2005 general election with a big cloud hanging over the safety and security of the voting system, particularly the ability of all voters to cast their ballot postally. Just a couple of days before the election was called, a High Court judge, sitting as an Election Commissioner, concluded his judgement in the Birmingham fraud case by saying that the system of postal voting and the administration of it ‘would disgrace a banana republic’. Judge Richard Mawrey had heard evidence which, he said, showed that the system of postal voting was ‘wide open to massive and systematic fraud’.

In the event, the election did not throw up as many cases of abuse as had been feared, but the general tumult over the potential has led to a significant shift in the Government’s thinking in this area and proposals for legislation which will go a long way to cutting down on electoral fraud in UK elections.

Electoral fraud is viewed by most as being a problem confined largely to times past. Bribery and personation were seen as issues of the 19th century rather than Britain in the 21st century. But the issues that have hit the headlines in recent times involved postal votes being changed using liquid paper, fictitious electors being added to the register and voters’ signatures being forged. But such cases have almost always been written off as being one-offs which deserved punishment but showed no fundamental weakness in the system as a whole. Indeed, the Government dismissed claims that the system was open to abuse and that some people were taking advantage of such possibilities as scare-mongering.

Until 2000, the ability to cast a vote by post was limited to those who had good reason for needing one.This changed in 2000 with the passing of the Representation of the People Act which introduced postal voting on demand, which meant that any elector could ask for a postal vote either for a specified election or permanently with no reason given and with no attestation as to need.

At the same time, a programme of alternative voting trials was begun with small scale all-postal ballots in some council areas.This programme continued in 2002, 2003 and culminated in 2004 with all-postal ballots in four entire regions of England for the European Parliamentary and council elections of that year.This expansion of postal voting brought benefits in turnout that were significant but modest.

The increased use of postal voting has, however, been accompanied by a number of warnings about the dangers of loss of secrecy and of fraud from organisations including the Electoral Reform Society. In particular, the Electoral Commission in 2003 called on the Government to strengthen the security around postal voting by changing from household to individual voter registration which would allow electoral registration offices to hold information on individual electors which could be used to reduce the risk of personation.This recommendation was made by the Commission again following the experiences of the elections in June 2004, and was also made by the Select Committee on the ODPM.The Government, however, was not convinced, arguing that a change in the system of registration might cause a decrease in the numbers registered.

A further recommendation of the Commission was that applications for postal votes should be returned directly to electoral registration offices rather than through party offices (then a common practice by all major parties) or other third parties, but this was also rejected.

Pressure for change, however, was dramatically increased by the hearing from February to March of electoral petitions in two wards in Birmingham following the 2004 local elections. In both wards, the Judge found that massive and varied fraud had occurred. A vote rigging ‘factory’ was discovered where ballot packs were ripped open and the votes altered if not for the Labour Party; false witness signatures were found; some ballots had been taken by those not entitled to them; and pressure had been brought to bear on electors by community leaders. With 2500 ballots found to have been cast improperly in each ward, the Judge had no hesitation in overturning the result and ordering a re-run in each case.

Although condemned by all political leaders, the scope for making postal voting more secure before the general election was limited. An emergency meeting of the Government, Electoral Commission and electoral administrators held at the start of the election campaign agreed that letters should be sent to all who held postal

votes reminding them of the fact and that extra warnings should be given on ballot packs reminding voters of the need for secrecy and security. Councils which had experienced the problems of fraud in the past, including Birmingham and Hackney, were among those who took the strongest measures to ensure that it did not happen again.

And so the 2005 general election took place against a backdrop of concern about the safety and security of the voting system. Some observers called for a postponement of the election in order to allow emergency changes to the system to be introduced. One Birmingham councillor – who later went on to be elected as the MP for the Yardley constituency – even launched a High Court bid to have the poll delayed.

Overall, the number of electors signed up to receive postal votes rose by a factor of five or six between the 2001 and 2005 general elections and more than 15% of all votes cast in the election were by post. In constituencies such as Stevenage, more than a third of all electors were registered to vote by post. In Wakefield, the rise was more than 1300%. In some areas, the rises led to concerns that campaigners, in their efforts to boost their favoured party’s chances, would go beyond the permissible in seeking postal votes. Newspapers such as the Times, which had highlighted the potential failings for more than a year, as well as the Guardian and the Evening Standard, all ran significant stories over the course of the campaign. Journalists from both the BBC Six o’clock news and the Evening Standard were reported to the Police for stories they ran in which they had applied for a postal vote in the name of someone else – albeit with their permission and for journalistic purposes only. A running total of the number of postal ballots which were due to be sent out was also carried in a number of publications.

In the event, few challenges have been lodged with the courts and the number of allegations of fraud are quite small. However, one of the challenges is in a Birmingham seat and history has proved that many allegations only come to light some time after the deadline for challenges (just 21 days from the result) has passed.

But perhaps concern over the potential for certain categories of fraud might be nearing the end of its lifespan.

Despite the limited number of instances reported during the General Election, the widespread publicity and continued lobbying by the Electoral Commission, Electoral Reform Society and others has ensured that the Government has broadened its plans for legislation. The proposed Electoral Administration Bill will include proposals for the collection of personal identifiers to help check on fraud as well as a longer period for the checking of postal vote applications – measures that will make personation much more difficult, even if it will give little protection against intimidation and other forms of interference in elections. Some, including the Electoral Reform Society, have suggested that the Bill could be amended to go further, but it is a significant step in making postal voting more secure.

However, the question we cannot answer, at least at present, is whether postal voting had any real impact on voter engagement and turnout.The number of postal vote applications rose by more than 10% of the electorate (although fewer than the usual proportion of those postal votes which were issued were cast), but turnout rose by only 2%. It is possible that postal voting preventing turnout from dropping to a new record low, but it is also possible that the method of voting made very little difference to turnout. A detailed study, comparing similar constituencies with different levels of postal voting and different turnouts, will be needed to test whether the postal voting on demand is worth the hassle.

Boundaries and Bias

Since the evidently unfair election result of 5 May, and even during the campaign, there has been growing awareness of the extent to which the electoral system is biased in favour of the Labour Party. It is now fairly well known that the Conservatives need to achieve a massive lead over Labour in order to win a parliamentary majority, while Labour are capable of winning a slim majority despite polling slightly fewer votes than the Conservatives. In the 2005 election Labour actually polled fewer votes than the Conservatives in England, but won 92 more seats and a clear majority.

Conservative commentators in particular have looked around for an explanation. For example, the Sunday Telegraph of 8 May 2005 referred to:

Those Tory voters wondering why their party polled only 3 per cent less of the national vote than Labour and yet has 159 fewer seats. It is clearly a problem that the Tories polled an average of 42,000 votes for each constituency victory, compared to an average of only 28,000 for Labour. This is a structural injustice in the electoral system that the next boundary review must address without delay.

There is usually a winner’s bonus of some sort in general elections – in 1987, for instance, it took 43,800 votes to elect a Labour MP and 36,600 to elect a Conservative MP, and there was little complaint from the press about the injustice on that occasion. However, it is correct to note that the bias is now on a massive scale. John Redwood in the House of Commons, has also complained that:

We know that Labour owes a lot to the fact that several Labour constituencies have few constituents, so it was much easier for a Labour candidate to be elected than it was for a Conservative or a Liberal Democrat. Labour knows, too, that the forthcoming boundary review will go some way to righting the gross injustice in the current electoral system whereby the average Labour Member of Parliament represents several thousand fewer voters than the average Conservative Member.

It is true that the average Labour constituency has fewer electors than the average Conservative constituency. The average English Labour constituency has 67,671 electors, while the average English Conservative constituency has 73,221 electors.The average English Tory MP has 5,550 more electors than his or her Labour counterpart and serves a seat 8.2% larger. Even so, Redwood’s assumption that this is the prime cause of pro-Labour bias in the electoral system is misplaced.

In 1979 the average English Conservative seat had 69,923 electors – and the average English Labour seat had 61,150. The boundaries were therefore much more biased in 1979 than 2005, a difference of 8,773 electors, making the average English Tory seat 14.3% larger. In the 1970s there were much more extreme differences between the largest oversized Tory seats (such as Bromsgrove and Redditch with 104,375 electors) and depopulated Labour seats (such as Newcastle Central with 23,678 electors, which could fit five times into some other seats). But the system as a whole operated much more fairly between the main parties in 1979 than it did in 2005. This should come as no surprise given the real causes of electoral system bias. To paraphrase that famous sign from the Little Rock campaign war room in 1992: It’s Not the Boundaries, Stupid.

This chapter analyses the effect of the current boundary review on the fortunes of the parties and therefore the extent to which changed boundaries will, or could ever, correct the bias in the electoral system.

The current boundary review

The boundary review started in 2000, and should be finished by the end of 2006.The review in Scotland was implemented in 2005, but the rest should take effect at the next general election.The review works county by county and is conducted by a neutral Boundary Commission, one each for England, Wales, Scotland and Northern Ireland.The first step is to ascertain how many seats the county is entitled to on the basis of its registered electorate.The next step is for the boundary commission to produce provisional recommendations, which change the number of constituencies if appropriate and shuffle the boundaries around so that the constituencies are relatively evenly sized, reflect community ties and where possible are not radically different from the constituencies that existed previously.
There is a period for public consultation and there are usually objections to the provisional recommendations; a local public inquiry conducted by an independent Assistant Commissioner then follows at which the case can be argued.The Assistant Commissioner then produces recommendations, and the Boundary Commission can revise its original findings.There is occasionally a second inquiry.The whole process is time consuming.

In general, boundary reviews help the Conservatives because population drifts out of Labour inner city seats and into the suburbs and counties – each review abolishes some depopulated Labour constituencies and creates some safe new Conservative seats. Will the current review, then, solve the bias problem?

The calculations that follow attempt to measure the political impact of the boundary review. More precise calculations await the completion of the review and the modelling efforts of Rallings and Thrasher, and perhaps others, but while there may be errors on individual seats in what follows, it is unlikely to be too far off.

England

The following counties (and former counties) will gain a seat each:

• Avon
• Cornwall
• Derbyshire
• Devon pEssex
• Hampshire
• Lancashire
• Norfolk
• Northamptonshire
• Warwickshire
• Wiltshire.

The following areas, provided that the provisional findings in the metropolitan areas are not reversed after public inquiries, will lose a seat each:

• London
• Greater Manchester
• Merseyside
• South Yorkshire
• Tyne and Wear
• West Midlands
• West Yorkshire.

This suggests a relatively modest aggregate change – crudely, 11 new seats created and 7 seats abolished, compared to 21 new creations and 16 abolitions in England in the previous round of changes finalised in 1995. London in particular was seriously affected, falling from 84 to 74 seats on that occasion.The disparities in the entitlements for each county and region, which were addressed in 1995, were larger than those the current review seeks to correct.

It is not quite so simple as Labour seats disappearing from the cities and new Conservative seats appearing in the shires either. When a safe Labour seat is abolished, its Labour voters are dispersed to neighbouring seats and this effect can shore up majorities in hitherto marginal seats, or even tip Conservative or Lib Dem held marginals over to Labour.To an even greater extent, a safe new Conservative seat is good news for one aspirant MP but bad news for the party. Forming a new rural seat tends to mean that the seats in the county become smaller than they were, and usually draws the best Conservative territory out of marginals.This happened in the last review in Bedfordshire for instance, where the creation of a sixth seat slimmed Luton North and Bedford constituencies to urban cores that were easier for Labour to win. The same happens this time in Warwickshire, Northamptonshire and Derbyshire. Although the Conservatives get a new safe seat in South Northamptonshire, the knock-on effects transfer at least two of their marginals back over to Labour.

The estimated partisan effect of the English boundary changes that have been made final is therefore a net increase of 7 in the total of Conservative seats, no net change in Labour seats and a net increase of 3 in the total of Lib Dem seats. See Appendix A for detail of the county-by-county changes.

This is not quite the full story, as the changes in the metropolitan areas outside London have not been finalised and it is in these areas that several Labour seats disappear. There are, on the basis of the provisional recommendations, a net 5 Labour seats being abolished and one Lib Dem seat is involved in complex changes which probably entail its notional loss to Labour. If the provisional recommendations are confirmed (and there may be changes, as there were in Merseyside) the net English changes will be Conservatives +7 seats, Labour –5 seats and Lib Dems +2 seats.

Scotland

The revised boundaries took effect from the general election of May 2005 and there will not be further changes between then and the next election.

Wales

The changes in most of Wales are minor, and its total entitlement of 40 seats is unchanged.There are small alterations to the highly marginal Conservative seat of Preseli Pembrokeshire, and more major boundary changes in North Wales.These are more difficult to model than those in England, because political behaviour is less easily discerned from local elections and demographic characteristics.

The current marginal constituency of Conwy is altered in a way that makes it a more difficult seat for Labour, adding a Plaid Cymru voting tract of rural territory (and the small town of Betws-y-Coed) from the former Meirionnydd Nant Conwy constituency. But the Conservatives are unlikely to benefit. Part of the existing Conwy goes into a new seat called Arfon, which will be competitive between Labour and Plaid Cymru.There are smaller changes to the neighbouring seat of Clwyd West, Conservative by 133 votes in 2005, which might imperil their majority.

It is probably safest to assume no changes in partisan representation as a result of the boundary changes in Wales, although the rearrangement in the Conwy area could gain (or lose) Labour a seat versus Plaid Cymru.

The net effect of the boundary changes

The boundary changes do not justify the hopes placed in them by some optimistic Conservatives.

The estimated effect would be to increase the number of Conservative MPs by around 7, reduce Labour by 5 and increase Lib Dems by 2. On the basis of the new boundaries the 2005 result would therefore have been Labour 351, Conservative 205 (including Staffordshire South), Lib Dem 64 and others 30, for a Labour majority of 52 rather than 66.

The boundary changes have some more subtle effects than this, in that they will also make various Labour-held marginals easier or more difficult for the Conservatives to gain for a given national swing. For instance, the Labour seat of Bristol North West, currently 136 on the target list and a seat that the Conservatives can aspire to only if they win an outright majority, gains the best Conservative areas from Bristol West and becomes much more marginal. However, for every case like this there is an example such as Gloucester, which becomes more Labour because it loses a ward from the edge of the city where Labour are weak.

There are three counties where the creation of an extra Conservative seat has a disastrous effect on Conservative chances elsewhere by taking some of the best Conservative territory out of the marginals. It is doubtful whether there has been much overall change in the shape of the battlefield – the mountain remains pretty much as steep as it has been, and the bias in the system is only slightly reduced.

Why is there still so much bias even after the boundary review?

Out of date figures?

A minor contributory reason to the continuing bias is that the review takes a long time and it is obliged to use electorate figures from 2000, which makes the exercise out of date before it starts. If 2005 electorate figures were used, the following additional changes would take place:

Counties gaining an extra seat: Cambridgeshire, Oxfordshire, Somerset Counties losing a seat: Tyne and Wear, West Midlands, London

These are not particularly dramatic, but that should be no surprise. Demographic changes happen slowly, and discussion in the media which assumes vast movements of population from the cities to the suburbs happening all the time is badly misinformed. Population movements were much faster and more dramatic in the period between 1945 and the mid 1970s, when millions were rehoused, new suburbs grew up and the New Towns were built.

In addition, if the Commission were to be stricter about equalising electorates and not make exceptions for special geographical reasons, the Isle of Wight would be divided into two seats and Cumbria would lose a seat – but this would also have been true in 2000. The rules have always allowed departures from arithmetical equality for special geographical reasons. It is not coincidental that the largest constituency (Isle of Wight) and the smallest (Na h- Eileanan an Iar) are both based on islands.

Labour would almost certainly lose three more seats from the reductions in the urban regions, but the gains are more complicated. The new seat in Cambridgeshire would almost certainly help the Conservatives, but in Oxfordshire and Somerset a review might create new Con-Lib Dem marginals depending on how the boundaries are drawn. Even if the most up to date figures were to be used in the boundary review, it would only make a difference of 6 more seats to the Labour majority – i.e. bring it down to 46.

The slow pace of the review is therefore not a significant factor in explaining bias.

Wales?
Wales will have 40 MPs in this parliament and the next parliament.This is more than its registered electorate merits
– if treated like England (and Scotland) it would have only around 33 seats.The reasons for over-representation are complex and result from the rules under which the Boundary Commissions operate. If Wales were to be cut back to 33, not all the casualties would be Labour.There are two under-populated Plaid Cymru seats in the north west, the Lib Dems would probably lose in mid Wales and the marginal Conservative seat of Clwyd West would probably be abolished (though it is probable that there would still be a Conservative seat in Pembrokeshire). Even assuming Labour were to suffer 5 out of 7 losses from any change in Wales, the over-representation of Wales still contributes only a net 3 to the Labour majority.

Gerrymandering?
Complaints are sometimes heard alleging that there was bias in the way the boundaries were determined in the run-up to 1997. The boundary commissioners are entirely neutral and take account only of the factors deemed relevant by law, which of course do not include party advantage.

It is sometimes argued that the bias is because Labour made such a professional job of presenting its case to public inquiries during the last boundary review that a systematic bias was created in the system. It is undoubtedly true that Labour did well at organising its case in those inquiries, but there is nothing wrong with that, and the Conservatives have only themselves to blame for the situation. They were in power at the time, the Major government had accelerated the boundary review, and the same professional approach was open to them as to Labour. In the previous review, leading up to 1983, it was the Conservatives who made the best case to the review.

In any case, there is a limit to how much effect the parties’ efforts can have. The overall allocation of seats to each county is pretty much fixed, and the arguments at inquiries are circumscribed by physical and social geography and political history, and the need to present arguments that address the factors the commissioners are allowed to consider. The biggest general issue of principle is described in boundary jargon in terms of ‘doughnuts’ and ‘sandwiches’. In the situation where a town is too big for a single seat and too small for two of its own, should there be a compact urban seat (a ‘doughnut’) with a bit of the town chipped off and added to a rural seat? Or should the town be split into two seats both of which are part urban and part rural?

Labour tended to argue during the previous review for ‘doughnuts’ and the Conservatives for ‘sandwiches’, for obvious reasons of self-interest in each case. The ‘doughnut’ case is easier to make on the grounds of keeping communities together and ease of internal links, so not surprisingly it tended to prevail. The presumption in the rules for keeping communities together is gradually having the effect of creating more socially homogeneous seats, based on common rural and urban identities, and this social pattern produces political consequences.

Ironically, Labour did so well in the 1997 election that they might have been even better off if the Conservatives had succeeded in creating more sandwiches! For instance, Medway and Chatham & Aylesford, the two classic ‘sandwich’ seats that inherited bits of the compact old Rochester & Chatham seat both fell to Labour, as did the two new Swindon seats in 1997. In both cases, Labour still have both bits of the sandwich even after 2005. Labour’s aim in the 1995 boundary review was to achieve advantage in a situation where the parties’ votes were roughly level, rather than to win additional seats in landslide conditions.

The current round of boundary review inquiries has been more evenly matched as the Conservatives have adopted more professional techniques and employed barristers to represent their interests. But the status quo is an important consideration in the rules for redistribution, and in most counties where there has been no change to the overall number of constituencies Labour have been successful in defending existing boundaries (with the notable exception of Staffordshire).

A final nail in the coffin of the intentional bias argument is found in the authoritative Media Guide to the New Parliamentary Constituencies, published in 1995, by Colin Rallings and Michael Thrasher.This modelled the 1992 election results on the new boundaries. Compared to the ‘real’ results, the new boundaries gave the Conservatives 7 more seats, Labour 2 more, Lib Dems 2 fewer and
Northern Ireland 1 more – a Conservative majority in that parliament of 27 rather than 21.The swing from 1992 required for Labour to win a majority increased from 4.1% to 4.3% with the new boundaries.The new boundaries were clearly not a Labour gerrymander.

The real reasons for electoral bias

As we have seen, boundaries make only a small difference to the degree of bias between Labour and the Conservatives that exists in the electoral system.

There is not a lot that tinkering with the boundaries can accomplish in reducing electoral bias. Differential turnout is an important element of electoral system bias. Turnout in the very safest Labour seats has fallen dramatically in recent elections, and rather less so in marginals and safe Conservative seats.This means that, for a given share of the national vote, it makes it easier for Labour to win seats. Imagine that there had been a full turnout in the 2005 election, and that the proportion of votes cast for each party had been exactly the same as in reality.

The share of the votes would have been Labour 36.3% (rather than 35.2%), Conservative 31.5% (rather than 32.3%) and Lib Dem 21.8% (rather than 22.0%) and the numbers of seats would be exactly the same.

The system ‘thought’ the Labour lead in share of the vote was 4.8%, while the votes cast showed only a 2.9% lead.

There is no way of eliminating this source of bias by changing boundaries. A simple numerical example can illustrate how differential turnout can produce startlingly biased results. Suppose that there are five constituencies with 70,000 electors each, each fought only by the two main parties. Three are safe Labour urban seats, two are safe Conservative rural seats, and in each safe seat the dominant party gets 60 per cent and the smaller party 40 per cent.

Turnout in election A is 80 per cent everywhere, turnout in election B falls to 75 per cent in the safe Conservative seats and 50 per cent in the safe Labour seats.

There is no change in the balance between the parties in any of the seats.

The votes are as follows:

Labour have slipped behind the Conservatives in the overall vote.There appears to have been a national swing of 2.2 per cent to the Conservatives although they are no closer to winning any of the Labour seats, and Labour now have a 3-2 majority in seats despite having slightly less of the vote. If the Conservatives had a swing of 5 per cent in their favour, they would be even further behind in the vote and still gain no more seats.This, in exaggerated terms, is a representation of what has been happening in Britain.This factor is nothing to do with any problems in drawing boundaries – all the constituencies are the right size and there have been no changes in population. Unless the causes of differential turnout are addressed, there is no way to solve this problem within the First Past the Post system. Inefficient vote distribution is another element of bias. This concept can be illustrated by a numerical example. Suppose there are three parties, each with a third of the vote but with that vote distributed differently. Party A gets a third of the vote in every constituency. Party B gets 40 per cent in two thirds of seats and 20 per cent in one third of seats; Party C gets 26.7 per cent in two thirds of seats and 46.7 per cent in the other third of seats. Suppose there are 600 constituencies.

Party A’s vote is therefore distributed very inefficiently from the point of view of winning seats, while B’s is extremely efficient. Even if Party A is actually a little ahead of the others it can still end up with no seats.The Conservatives have a rather less extreme version of what Party A is suffering, while the distributions of Labour and Lib Dem support are milder versions of Party B and Party C in the example. Inefficient vote distribution can arise for different reasons. In the 1950s Labour’s vote was poorly distributed because the party piled up massive majorities in safe working class seats while losing other areas to the Conservatives with relatively small majorities. Hence, the Conservatives won a majority in parliament in 1951 despite polling fewer votes than Labour, and the difference between Labour’s majority of 5 with a 2.6% lead in 1950, and the Conservatives’ majority of 58 on a lead of 3.3% in 1955. Inefficient vote distribution can thus be caused simply by political sociology or geography; in the current position if the Conservatives increase their support markedly in rural areas they will gain relatively few seats because their support is already concentrated in rural areas. Efficient vote distribution can also be caused by tactical voting. The outcome in the example above could be achieved by supporters of Party B and Party C trading tactical votes among themselves to keep out Party A.Tactical voting has been a contributory factor in the strong bias against the Conservatives which has developed in Britain since 1992. If a party has evenly spread support and falls much below a third of the vote in a three party system it can be very harshly treated by the electoral system, as has happened time after time to the Liberals and Liberal Democrats.There is no way to fix this sort of problem by changing boundaries. While there are substantial numbers of Conservative supporters in South Yorkshire, for instance, there is simply no way of corralling enough of them into a single-member constituency to give them a chance of winning a seat, however much one bends the boundaries.The same is true for, say, Lib Dems in East Sussex.The Labour vote being ‘lumpier’ (more efficiently distributed), it does tend to be able to achieve representation where it is locally in a minority although there are cases such as Surrey where it is too thinly and evenly spread.

The solution

It is not possible to solve the real problems of electoral bias while maintaining the First Past the Post electoral system.The Conservatives have proposed to rewrite the rules for redistribution to eliminate the variation in size between constituencies, but the detail of how this might be accomplished is full of pitfalls. It would entail the abandonment of the county-by-county procedure used until now, and the division of large local government wards. It would also involve the end of the special geographical circumstances get-out.There would be some very odd creations, including possibly a constituency spanning the Western Isles and Shetland, and another either end of the Isle of Wight ferry combining bits of Portsmouth and Ryde… To insist on rigid mathematical equality would effectively abandon the requirement that constituencies reflect natural communities, but given that community representation is supposed to be one of the plus points of single member districts this would seem a high price. It would also open the door to flagrant gerrymandering of the sort practised in parts of the United States. Neither would it be an effective method of eliminating bias, because it does not address the main causes of system bias.

A fairer electoral system that translates votes more directly into seats is the only way of tackling bias. A proportional system that represented parties that are local minorities, or have their vote spread evenly rather than concentrated, would be much less subject to bias. A system that encouraged healthy competition between the parties in all areas, not just in a small number of marginal seats, would remove some of the causes of differential turnout and inefficient vote distribution. But any system based on geographical districts can be affected by differential turnout.

Changing parliamentary boundaries can only tweak at the edges of electoral system bias, because unfair boundaries are not the real cause of that bias. Only a reformed electoral system can provide a level playing field between the parties.

A ‘health warning’

In this chapter we compare the possible results of the 2005 election under several different electoral systems, namely:

• Regional list PR
• The Additional Member System (AMS)
• The Alternative Vote (AV)
• AV with a proportional top-up (AV+)
• Single Transferable Vote (STV)

There are many problems in estimating what would have happened in the 2005 election under alternative electoral systems, so any effort along these lines must necessarily be approximate and rather tentative. Among the difficulties are:

• Boundary determination. AV+, AMS and STV all require redrawing the boundaries to produce larger or multi-member constituencies. There are different ways of doing this, and different options may produce slightly different outcomes.
• Preference transfers. AV, AV+ and STV involve people voting in a different way, by ranking preferences 1,2,3… rather than with an X. While there is opinion poll data on where the second preferences of supporters of each party will tend to go, it is necessarily imprecise and will not apply invariably in every seat – a particularly attractive or unattractive local candidate will affect transfers. Also, transfers from minor parties are often significant because these candidates are eliminated first, and these are difficult to estimate.
• Double ballots. AV+ and AMS use two ballot papers, one for the local candidate and one for the list element.
  Experience in Scotland in particular shows that in practice a significant minority of electors vote differently in the two parts of the election, and are more willing to support minor parties in the list vote.The simplifying assumption of using the same party preference for both votes may cut out some of the diversity that a proportional system can generate.
• Multiple candidates. In STV voters are not obliged to vote for the entire party slate, and the simplifying assumption that they will loses some of the subtlety of the system. At the margins, the number of candidates the parties choose to stand, and how evenly the party’s vote splits between candidates, can affect the last seat to be determined in particular.

But by far the biggest problem is that voting behaviour itself would change. People who at present tend to support Labour but live in a Lib Dem-Conservative marginal may have voted tactically in 2005 but under a fairer system would be freed to express their real choice. For this reason, Labour might do rather better in votes as well as seats in some areas (particularly in the South West) while the Lib Dems would pick up in some areas where they are currently weak too. In the relatively few seats where the Conservative vote has been suppressed by tactical voting in Lib Dem-Labour contests, their vote might bounce up as well. As well as this sort of tactical unwind, a proportional system would also encourage people who were considering voting for a minor party such as the Greens to actually do so.

Further, systems that involve lists or larger constituencies will give electors a wider choice of parties than was available in many constituencies – for instance the Green Party stood in only 183 out of 645 constituencies in the FPTP election. Their vote would have been higher had the option been available in the other 462 constituencies.

Having made all these caveats, the approximate notional outcomes would be as follows.

Regional list PR/ large top-up AMS

A reasonable example of a list PR system would be modelled on the constituencies used in the European Parliament elections – the 9 English regions plus Scotland, Wales and Northern Ireland. These would be very large regional constituencies and would give a highly proportional result; Labour’s share of seats compared to national list PR would be a little higher, principally at the expense of the minor parties and to some extent the Conservatives.

The regional distribution would be considerably fairer. Currently, just under half of Conservative MPs represent constituencies in two regions, the South East and the East of England – only just under a third of Conservative voters live in these regions. The table below gives the regional distribution of seats in Great Britain under regional list PR compared to FPTP (in brackets).

Most countries using regional list systems to elect their parliaments use smaller electoral divisions than the English regions; this tends to raise the threshold somewhat against the minor parties. The same effect could be achieved by introducing an artificial threshold, as in Germany and for the elections to the Greater London Assembly, where parties polling below 5% are disbarred from winning list seats.

The regional list results can also serve as an approximation of what would happen under an Additional Member System with a large top-up element – around a 50-50 split between constituency and list members gives almost exactly the same result as a regional list, and a 40% top-up is usually enough. The exception is in cases where one party completely dominates a region (as Labour does in the North East), when the dominant party can sweep nearly all the constituencies and the top-up cannot completely redress the proportions. The principal difference in 2005 between regional list PR and AMS with a large top-up would be that some candidates would win on an individual basis in the constituency section (Richard Taylor, Peter Law, the Speaker) who would not qualify on the basis of share of the regional vote. The seats going to the other parties would be adjusted accordingly. Also, as noted in 9.1 above, people might vote differently on the list ballot and this could favour the Greens and UKIP in particular.

Alternative Vote (AV)

The Alternative Vote is a minimalist form of electoral reform which does not alter the current pattern of single member constituencies but instead introduces preferential voting (1,2,3… rather than X). However, as with PR systems people might make different choices under AV than they do using FPTP. AV is not a proportional system – single member district systems cannot be, except accidentally. In some circumstances it could produce more proportional results (for instance in the 1987 and 1992 elections it would have reduced the Conservative preponderance) but in others it could produce even more disproportional results than FPTP. In the 1997 election feeling was running so strongly against the Conservatives that AV would simply have helped several more Lib Dem and Labour voters swap preferences and defeat Conservatives in seats where the Tories were ahead under FPTP. In 2001 it would also have swollen the Labour majority. In 2005 the evidence from opinion polling suggests that Labour would have once again have had a larger majority under AV than FPTP. John Curtice (Independent 10 May 2005) estimated that the Labour majority would have been 98 rather than 66.

The reason AV would swell the Labour majority is that the second preferences of Lib Dem supporters still tend to favour Labour over Conservative when those are the last two remaining choices, although to a lesser extent than they did in 1997. AV would have helped narrowly defeated Labour MPs such as Chris Leslie in Shipley to hold off the Conservatives. It would also have helped the Lib Dems vis-à-vis the Conservatives because Labour voters would favour them heavily in those circumstances; it would help them also somewhat against Labour.The measurement of each of these effects is imprecise and varies between different opinion polls.The Electoral Reform Society is undertaking some research into second and lower preferences, which will be reported in due course.

Curtice used exit poll data, but using some BBC/NOP aggregate data on second preferences in their late campaign polls produces an estimate of a Labour majority of around 88. The numbers of seats that would give different results from those under FPTP are limited because votes from eliminated candidates do not transfer 100 per cent to their favoured candidate (there are some Labour voters who would give their second preferences to Conservatives over Lib Dems for instance). Some of them do not transfer at all (see Chapter 10 on the mayoral elections for instance). Labour would lose around 8 to the Lib Dems and pick up about 19 from the Conservatives; in addition another 4 seats would go from Conservative to Lib Dem. There would also be changes involving the smaller parties which are more difficult to estimate.

The Conservatives would be the principal losers from AV because they would still attract relatively few second preferences from supporters of other parties. This need not be a permanent feature – if the party manages to recast its appeal it could yet become attractive as a second choice, and the political circumstances of future elections need not replicate those of previous elections.

Alternative Vote with small top-up (AV+)

AV+ is the system devised by the Jenkins Commission which reported in 1998. It is based primarily on the Alternative Vote in single member constituencies, but with a proportional top-up element accounting for around 15- 20% of the membership of the House of Commons.This would reduce but not eliminate the exaggerating qualities of FPTP and AV. It is a system designed to permit single- party majority government in circumstances where one party has a high share of the vote, a decisive margin over its opponent or a high level of general consent as implied through preference transfers. AV+ would have produced Conservative majorities in 1979, 1983 and 1987 and Labour majorities in 1997 and 2001. In the less decisive elections of 1992 and 2005, it would have produced no overall majority.

The outcome of such an election result would presumably be a Labour minority government.

AV+ has a threshold pretty much as high as AV or FPTP for smaller parties because the top-up is not large enough to allow them representation.Top-up seats are allocated for the most part on the basis of one for a smaller county such as Bedfordshire and two for a larger county such as Staffordshire; there are no top-ups of more than two members.The top-up seats therefore tend to go to the major parties that find themselves under-represented in the county.

The constituency template used is based on the illustrative scheme in Annex C of the Jenkins Report, with necessary adaptations to cope with the new parliamentary boundaries in Scotland.

AV+ requires perhaps even more sets of assumptions than any other electoral system to be modelled – on the new boundaries, preference transfers and the relationship between list and constituency voting. It is therefore difficult to be precise about modelling an outcome.

Single Transferable Vote

The Single Transferable Vote involves preferential voting between individual candidates in multi-member constituencies. It gives broad proportionality but unlike other PR systems it is candidate-centred rather than party- centred and gives voters a high degree of control over which individuals are elected.

The larger the number of members per constituency in STV, the more proportional the overall result. This example has assumed constituencies electing between 3 and 5 MPs (85 three-member, 64 four-member and 25 five-member), the same spread as used in the Republic of Ireland for Dáil elections. The constituencies are composed by combining existing Westminster constituencies that best fit the local social geography, provided that every constituency has between 3 and 5 members. For details of the boundary scheme employed see Appendix B.

STV has a high threshold for electing minor party candidates. In a 3-member constituency 25% is needed to guarantee election, and in a 5-member seat the quota is 16.7%. Candidates with less than this can still win election if they attract enough transfers from supporters of other parties as the count progresses.

Summary

Two electoral systems would give Labour a majority in 2005 – FPTP and AV. AV+ would make forming a non-Labour government impossible but stop short of giving Labour an outright majority. For the Conservatives, it seems to matter relatively little which electoral system is used – they get around 200 seats whichever is used – except for AV which would knock them down to about 175. STV is narrowly the most Lib Dem friendly system, but any substantially proportional system would help them up to around 140. The more majoritarian AV and AV+ would help them less, but any alternative system is better for them than FPTP.

Fears that PR would let in large numbers of extreme and splinter parties are exaggerated or misplaced. Even under the most proportional systems there would be comparatively few minor party MPs, and systems such as AV+ and STV certainly do not produce an ungovernable rabble of parties.

STV scores relatively well in aggregate as being proportional to a high degree of accuracy for the two main opposition parties but with a significant winner’s bonus for Labour.This winner’s bonus would almost certainly be larger if the predominant pattern were to be three-member constituencies.

Modelling alternative electoral systems is a highly artificial exercise that cannot give precise results and certainly cannot replicate all the effects of introducing a new electoral system. It is, however, indicative of the properties of alternative electoral systems.

Local Elections of May 2005

For the third successive election, the general election fell on the same day as local authority elections for the ‘shire counties’ in England.These are the upper tier bodies that exist in areas where two-tier local government survived the abolition of the metropolitan counties in 1986 and the local government reorganisations of the 1990s. In most of the counties the largest urban elements have been extracted, such as Luton from Bedfordshire and York from North Yorkshire. There were also local elections in a scattering of unitary authorities (Bristol, Isle of Wight, Stockton-on-Tees) and elections for directly elected mayors in four urban areas.

County councils

The Conservative Party, for the third successive round of county elections since the nadir of 1993 when only Buckinghamshire survived under Tory control, improved its position.The party gained Northamptonshire directly from Labour and Gloucestershire, Oxfordshire, Shropshire, Suffolk and Worcestershire from no overall control.The Lib Dems gained three from no overall control – Cornwall, Devon and Somerset.

Just as the Labour Party nationally managed a majority on a very low share of the vote, Labour councillors in some counties also won majorities with less than 40%, as in Lancashire where 37.5% sufficed for an overall majority.

However, the other parties also benefited from disproportional county election results, for instance the 36.0% of the vote that won the Conservatives 58% of the seats in

Oxfordshire, and the 38.0% that gained the Lib Dems control of Devon. In many other counties a vote share (usually for the Conservatives) in the 40-45% region was translated into an overwhelming victory in seats, for instance:

Only in Buckinghamshire – now shorn of Milton Keynes – did the Conservatives poll over 50% of the vote, but their share of seats was still a highly disproportional 77.2%, which will make the conduct of opposition difficult in the county. Labour replicated this achievement in County Durham.

The county council electoral divisions had been subject to a boundary review since the last round of elections in 2001.The Local Government Act 2000 removed the requirement that county council electoral divisions have one member only, and in the new boundaries that came into force in 2005 several double-member wards have been created.There are three three-member wards, Penzance in Cornwall, Hucknall in Nottinghamshire and Bicester in Oxfordshire.

The purpose of allowing double member wards was primarily to increase the level of coterminosity with district council wards while maintaining an acceptable level of electoral equality.The directives given to the Boundary Committee for England gave a greater emphasis on avoiding cross-cutting district and county ward boundaries than in the past. Most of the double member wards are in urban areas within the shire counties.This apparently technical change has had some political consequences.

In smaller towns there may be enough Labour territory to sustain a single member county ward, but mixed with the rest of the town the Labour presence is submerged. In three-member Bicester, Labour had previously won a single-member ward but the Conservatives won all three seats this time. Labour has lost out similarly in places such as Biggleswade in Bedfordshire and High Wycombe in Buckinghamshire, where the recommendations were very controversial. One Wycombe ward – poetically named Downley, Disraeli, Oakridge and Castlefield – drags in enough rural territory to submerge some rather Labour areas of the town. Labour had 15% of the vote in Buckinghamshire, heavily concentrated in High Wycombe, but elected only two councillors (3.5% of the county, and only 2 out of 9 in the town plus nearby rural areas). The Disraeli area, incidentally, appears ironically to have been one of the Labour elements of the ward.

The creation of these double-member FPTP wards illustrates the difficulty of creating rational single-member districts for three tiers of government (district, county and Westminster). Advocates of FPTP often claim that the link between elected representative and an identifiable community is paramount, but in practice it is increasingly apparent that it is impossible to draw single-member districts that give electoral equality and keep natural communities together. Multi-member constituencies are a desirable rationalisation, but in order to reflect local opinion accurately and reduce the possibility of gerrymandering they would have to be elected on a proportional basis. Multi-member FPTP is liable to produce highly disproportional results because it accentuates the winner take all aspect of FPTP. It is not coincidental that some of the most disproportionate election results have been produced in all-out elections in the London boroughs in 2002 and the metropolitan boroughs in 2004, in both of which three-member FPTP was the norm.

This consideration should be borne in mind if the government moves towards rationalising the calendar of local authority elections with a view to instituting all-out elections on a four-yearly cycle for each authority.

Unitary authorities

The fallow year for most of these councils fell in 2005, with elections in only three – all-out in the Isle of Wight (a former county council), for a third of the seats in Bristol and all-out in Stockton-on-Tees after ward boundary changes.The Conservatives gained Isle of Wight from a coalition of Lib Dems and Independents, the Lib Dems replaced Labour as the largest party in Bristol and Labour narrowly lost overall control of Stockton after gains by localised Independent groupings.

Mayoral elections

Four direct elections took place for mayors – in Doncaster, Hartlepool and Stoke-on-Trent for the second time, and in North Tyneside for the third time. In addition, a referendum was held on the Isle of Wight into whether it should have a directly elected mayor. The proposition was rejected, with a 56.3% vote for ‘No’. No referendum has approved changing to a directly elected mayor since 2002.

In Hartlepool incumbent Independent mayor Stuart Drummond was re-elected. Drummond polled 42.1% of first preferences to 16.4% for his nearest competitor, Carl Richardson of the Labour Party, and was also favoured by the transfer of second preferences.The election was notable for the nearly 80% of the vote going to various sorts of Independent, to only 20.8% for the two candidates representing the main political parties. Labour polled 51.5% of the vote in the Parliamentary contest, so less than a third of that vote also backed the party’s mayoral candidate; of the much smaller Conservative vote (11.5%), fewer than half supported their candidate for mayor.

Drummond was clearly much the most popular candidate; his vote in the second count was a whisker (12.5 votes out of 33,795 cast) over half the total vote in the first ballot (50.04%). However, the cumbersome ‘Supplementary Vote’ (SV) system threw up some problems. Of the 14,041 people who had voted for candidates who were eliminated because they did not come first or second, only 3,865 second preferences were counted. Only 27.5% of potential transferred votes were actually counted in the second round. Seven candidates stood, and only those voters who chose Drummond or Richardson as their second choice had the chance of affecting the final outcome. It seems likely that most of the 72.5% of discarded second preferences would have been for other eliminated candidates, although some may well have been left blank deliberately or from confusion, and some will have voted twice for the same candidate.

In Stoke-on-Trent Labour candidate Mark Meredith defeated incumbent Independent mayor Mike Wolfe, confirming Labour’s local comeback that saw them take back control of the council in 2004.The mayoral election was notable for the close three-way tie for second place – only 435 votes out of 82,950 cast separated the second- placed Conservative candidate from the fourth-placed BNP candidate. Stoke very nearly had a run-off between Labour and the BNP thanks to the elimination rules under SV.The uncertainty about which candidates would be in the run-off (the incumbent being eliminated after coming third) was reflected in the low proportion of effective transfers – only 16,627 transfers were counted, 42.1% of the vote for eliminated candidates.The winning candidate, even after the second count, had only 44.6% of the votes originally cast.

In Doncaster Labour incumbent Martin Winter was re- elected. In the first preference count he polled 36.7%, a clear lead over the nearest challenger, Independent candidate Michael Maye who polled 25.1%. After eliminating the five other candidates, Winter was elected with a considerably reduced lead over Maye; the effective second preferences were 10,004 for Maye and 5,727 for Winter, which reduced Labour’s lead from 12,711 to 8,434. However, only 37.8% of the vote for eliminated candidates actually affected the final count, as second preferences were sometimes not used and sometimes scattered among other eliminated candidates. It is interesting to note that had the second preferences divided in exactly the same proportions but that double the number had been used effectively, Maye would have narrowly won. The winning candidate, even after the second count, only had 42.0% of the votes initially cast.

In North Tyneside Labour also gained the mayoralty, this time from the Conservative incumbent Linda Arkley.The margin was narrow, and the election was notable for being the first time that the distribution of second preferences has changed the result of a mayoral election – Arkley was 1,414 ahead on first preferences but lost to Labour’s John Harrison by 1,002 after second preferences were counted. Of the 15,231 votes cast for the eliminated Lib Dem and National Front candidates, 10,398 (68.3%) made effective second choices.This high figure reflects the relative simplicity of the 4-candidate ballot, and perhaps the marginal nature of the election in which electors would have been aware of which two candidates would be in the run-off and how important their second preferences might be. Even after transfers, the winner still had less than half the votes initially cast – 47.7%.

While SV worked relatively well in North Tyneside, the poor rate of effective transfers in Hartlepool, Stoke and Doncaster suggests that it is poorly adapted to elections with a large choice of candidates and uncertainty as to which candidates might be first and second placed.

It would be relatively easy to move instead to the simpler Alternative Vote (AV) system for mayoral elections. AV would end the confusing double ballot required under SV, and also not require voters to guess which two candidates will be in the run-off and cast (possibly tactical) second preferences for those candidates. Winning candidates would be much more likely to have a genuine majority, as only Drummond in Hartlepool did by a tiny margin on this occasion.

Conclusion

The general election of 2005 was a particularly dramatic illustration of the flaws in First Past the Post:

• It awarded the immense powers of a government with a majority in the House of Commons to a party which enjoyed the support of scarcely one in three of those voting and a little over one in five of the total electorate. These figures are the worst ever.
• It resulted in only one in three MPs having majority support among their local voters, and no MP commanding a majority of the local electorate.This is the lowest proportion of majority mandates ever.
• The British people remain disengaged from electoral politics, as demonstrated by the very low turnout – the second or third worst in history.
• Constituencies changed hands for reasons more to do with tactical voting and split votes than actual changes in opinion.
• Despite the Conservatives gaining the most votes in England, Labour still have a majority of MPs from England.
• The decision of the voters of Northern Ireland, particularly unionist voters, was misrepresented.
• First Past the Post short changes voters in Scotland, Wales and Northern Ireland in Westminster elections compared to the fairer representation given to them by the more proportional systems used for their devolved institutions.

In several regions of England the results were extremely unrepresentative.The worst case is probably the 1.1 million Conservative voters in the metropolitan counties outside London, who elected only 5 MPs.

First Past the Post remains an obstacle to fairer representation of women and ethnic minorities.

But out of all these glaring faults, progress may come from the 2005 election. Notwithstanding Lord Falconer’s Nelsonian declaration that he could see no groundswell for reform, the unfairness was so blatant that it has led to a mood of protest and discontent since the election. It has been made manifest in the massive response to the Independent’s Campaign for Democracy and the new impetus that the electoral reform movement has gained. Superficial quick fixes such as changing parliamentary boundaries, or experimenting with technological changes such as electronic or all-postal voting, can be rapidly

dismissed as means of making elections fairer or solving the public’s disconnection from politics. The problems may be obvious, but so is the answer. What is at fault is the voting system, and only a change to the voting system can provide a solution.

Government spokespeople have maintained that the election gave the electorate the outcome they wanted – a Labour Government rather than a Conservative one, although one with a reduced majority.The evidence certainly suggests that given a straight choice between Labour and the Conservatives, more people would have chosen Labour. But that was not the choice: we no longer have a two-party system, and it could equally well be argued that if the choice had been between Labour and the Liberal Democrats, a majority might have opted for the Liberal Democrats. Claims that the outcome is what the electorate wanted are therefore decidedly weak.

During the 18 years of Conservative government (1979 to 1997) it was often claimed by opposition parties that the Conservatives did not have majority support for some of their more controversial legislation, such as the introduction of the poll tax and the privatisation of British Rail. The Conservatives from 1983 onwards, after all, never had more than 42.4% of the vote. If that was a concern then, surely it must be a much greater concern now that we have a Government with only 35.2% of the vote. It remains to be seen, however, whether the Government will accept that the voting system has not given it a satisfactory mandate and make consideration of electoral reform one of its priorities.

Any movement towards electoral reform will, however, require a decision by MPs and it is unlikely that MPs will vote for a measure which they do not see as being in their personal or party interests.What then are the prospects for change?

We can assume that the Liberal Democrats and the smaller parties will continue to demand electoral reform: they have been the main losers under First Past the Post – they have nothing to lose from any move to a more proportional system and have much to gain.

Although the Conservatives received a share of seats more or less proportionate to their share of the vote, compared to Labour they were hugely disadvantaged by the bias in Labour’s favour resulting from the geographic distribution of the Conservative vote. While Labour’s 35.2% of the votes earned it 55.1% of the seats, the Conservative’s 32.3% of the votes won it only 30.5% of the seats. In present circumstances the Conservatives have a huge hill to climb: from their 2005 position to simply overtake Labour as the largest party they would need a swing of 4.3%, but to achieve an overall majority the swing they would require would be 7.3%.These are very high swings by past historical comparison.The swing to Labour in 1997 was 10%; the largest pro-Conservative swing since the war was 5.2% in 1979.The swings in the last two elections have been 1.8% in 2001 and 3.0% in 2005.

The Conservatives are disadvantaged to some extent by differences in constituency sizes, the average size of the constituencies in which they won being greater than those won by Labour.This, unfortunately, has given many Conservatives the false hope that the Boundary Commission can deliver their salvation. Implementation of the Boundary Commission report would narrow the Labour majority by only about 14, and a more radical equalisation would cut it perhaps by double that margin if they were lucky.The predominant reason for the bias against the Conservatives is the spread of their vote, a problem which cannot be overcome under the present system other than by complex gerrymandering.

The Conservatives therefore have strong reasons for wanting a change in the electoral system.

Labour might think that an electoral system which has been so kind to them is worth keeping, but they would be well-advised to take a closer look at the figures.They have won a third term and may have high hopes of a fourth, but there is always the risk of the swing of the electoral pendulum. Although their 66 majority might seem a comfortable one, it is nevertheless fragile. A swing of only 2% (and perhaps a little less under the new boundaries) would see their overall majority disappear.The time may be drawing near when self-interest as well as principle argue in favour of Labour moving on electoral reform.

Thus the issue for Labour is not just that they might lose seats under a more proportional system – some Labour MPs are likely to lose their seats under First Past the Post and a more proportional system would at least give them a fighting chance. However, they can only implement electoral reform while in power, and if they do not take the
opportunity in their third term they risk entering another 18 years in the wilderness of opposition.

All parties therefore have good reasons for wanting electoral reform, even if the situation is not yet apparent to them.

The case for electoral reform, however, rests not just on the present system’s unfairness to the parties. A more important consideration is fairness to the electorate.That the outcome of the election was a foregone conclusion in at least two thirds of the seats did not give many a sense of involvement in the choice of a government. When 71% of the votes were ‘wasted’ in that they were either cast for losing candidates or merely added to surplus majorities, and when the major parties felt able to focus their campaigns on just 2% of the electorate, it is hardly surprising that many felt little incentive to vote.

International comparisons indicate that a proportional system can increase turnouts: certainly they make more votes count, and the general election results show clearly that where there was perceived to be a real contest turnouts were significantly higher. Labour, for example, had much more incentive to campaign for a few hundred extra votes in its key marginals than to seek a substantially increased vote in its heartland constituencies – its objective was not to maximise its vote but to get the votes it needed where it needed them. A more
proportional system would have required different campaign priorities involving the maximisation of its vote in all parts of the country.

However, it is also apparent that a move towards
proportionality alone will not overcome problems of voter disengagement. Perhaps it is the very nature of our politics that alienates the electorate: for many inter-party debate based on accusations and counter-accusations and political point-scoring rather than open and honest debate is not attractive, but nevertheless our winner-takes-all electoral system is one that promotes negative campaigning. If we were to move to a more proportional system in which after elections politicians might need to reach agreements with other parties they might take a more reasoned view of their opponents during the campaign; if we had a system that involved preference voting, in which success might depend on winning second and lower preferences of supporters of other parties, candidates would be less inclined to adopt a needlessly adversarial approach. The right choice of electoral system could therefore encourage a change of political culture to one in which there is more rational debates on the issues and in which politicians would be encouraged to acknowledge that not all problems have simple solutions.

Labour’s manifesto included a commitment to review the experience of electoral systems and noted that the correct way to change the system for the Commons would be through a referendum.The review is already underway, although the present exercise within the Department of Constitutional Affairs falls far short of the open, transparent process which most electoral reformers want to see.
Moreover, the review is a poor substitute for the referendum on the recommendation of an independent commission which Labour promised in 1997 and many have argued for a new, independent commission, and even a ‘citizens’ assembly’ of the type recently used to recommend an electoral system in British Columbia.
Nevertheless, the government’s current review, whatever its limitations, may provide a focal point for debate.

The general election results clearly illustrate the need for reform, but any discussions on changing the electoral system will need to consider what change to make. Here a useful starting point may be the criteria which the Labour Government in 1997 devised for the Jenkins Commission:

• Achieve broad proportionality
• Fulfil the need for stable government
• Extend voter choice
• Maintain a link between MPs and geographical constituencies.

These criteria are one which most electoral reformers, and indeed many who do not advocate reform, accept.The problems lie in interpreting what they mean and in prioritising their relative importance when, for example, trade-offs are needed between the maintenance of links to geographic constituencies and the level of proportionality to be achieved.

One much discussed system is the Alternative Vote (AV). It is seen as the simplest ‘reform’. It maintains single-member constituencies and the unique MP-constituency link which many MPs, and particularly those opposed to reform, want to keep. This report is not the place for a discussion of the merits and demerits of single-member constituencies, but what is clear from the general election results is that AV would do little to restore the legitimacy of government. It would, if anything, have given Labour an even greater majority. AV would reduce the need for tactical voting as voters could vote for their preferred candidate and know that their votes would not be wasted, and it would overcome the problem of so many MPs not having majority support in their constituencies. But it would not produce fairer representation and all parties would still suffer from ‘electoral deserts’.

A more proportional system is therefore the only answer to the major defects of First Past the Post. However, proportional representation is not a voting system but a whole family of systems, and the parameters of each member can be adjusted to give different degrees of proportionality.Too often the case for PR has foundered not on the need for PR but on the particular form of PR that should be used.

Britain’s experience of list systems in the European Parliament elections has not been a happy one.The ‘closed list’ system used in these elections has been unpopular for the good reason that it does not allow voters to vote for the preferred candidates, only for their preferred parties. Those high on the lists of the major parties are almost guaranteed election, whatever the voters might think of their merits, while excellent candidates near the bottom of lists have no chance of success.The system therefore allows party selectorates huge influence in deciding who becomes an MEP, shifting the accountability of those elected from their electorates to their parties.The problem is compounded by large regions which makes MEPs distant from their electors. List systems could be used with smaller regions, of, say, 4 or 5 existing constituencies, for Westminster elections and ‘open lists’, which allow voters to choose the candidates they want from their parties’ lists, would be a further improvement. But even open lists would not provide all the advantages of the Single Transferable Vote (see below).

Many argue the case for an Additional Member System (AMS) which preserves single-member constituencies, but at the expense of having list members who have little direct accountability to the electorate.This is not an attractive proposition if voter disengagement is considered to be in part a consequence of the perceived distance between electors and their representatives. Constituency contests would still suffer from the prevalence of safe seats and lack of real competition, from high numbers of wasted votes and the need for tactical voting to make votes count. Certainly some of these problems could be overcome by the AV+ system recommended by the Jenkins Commission, but there are serious objections to that system on account of its complexity, requiring voters to vote preferentially for constituency candidates as well as for a party list. It would be especially complex if, as recommended by Jenkins, people are given the option of voting for particular list candidates on their party’s list.

However, both AMS and AV+ face a major practical problem: if the Commons were to consist of both constituency and list members, unless there were to be a significant increase in the number of MPs, the number of constituencies would need to be reduced.The thought of a major redrawing of constituency boundaries is not something that would allow many MPs to sleep easily.

The Single Transferable Vote (STV) also requires a reconfiguring of constituencies (although it is possible to do it relatively easily by combining existing constituencies without the need for drawing completely new boundaries), but other than this change it does not suffer from some of the drawbacks of AMS and AV+. It is relatively simple for the voter to use, requiring him or her only to number individual candidates in order of preference. It retains a strong link to geographical constituencies and gives a strong incentive for all MPs to be attentive to the people who elected them. It is broadly proportional but does not involve a splintering of representation between small
parties. It has been the preferred system of the ERS for many decades and the election of 2005 (and the experience of the flaws of AMS in the Scottish and Welsh elections) only strengthens the case.

Choosing an electoral system requires various trade offs between different objectives. With STV the key decision is over constituency sizes.The larger the multi-member constituencies used, the better the proportionality and the more choice offered to voters, but the larger the areas which must be covered by elected members. Most
supporters of STV advocate constituency sizes of 5 or 6 seats, but even STV with 3-seat constituencies, while not providing the full advantages of the system, would still deliver a broadly proportional result (allowing a greater chance of single-party government) and overcome, or at least amerliorate, the worst problems of First Past the Post.

Of the systems under consideration, however, only STV would combine broad proportionality with preference voting, the combination which we have already argued is most likely to promote a change in our political climate and better voter engagement.

However, while the debate on the most appropriate system for electing our MPs is an important one – we would not want to introduce a system that does not adequately overcome the problems of First Past the Post – the important message of the 2005 general election is that change is urgently needed.The outcome of these elections demands that the Government acts and acts soon.

Labour, since its election in 1997, can be justifiably proud of its record in constitutional reform: it has devolved power to Scotland, Wales and Northern Ireland (although not without expected difficulties in the case of Northern Ireland), introduced broadly proportional systems for devolved levels of government, introduced proportional representation for European Parliament elections (even if an unfortunate choice of system), made a start to the reform of the second chamber and experimented with new methods of voting. But so far it has failed to deliver on its promise to allow the electorate to decide how it will elect its MPs.The outcome of the 2005 general election presents it with a challenge and an opportunity: either it can go down in history as the most reforming government of all times, or the one that missed the opportunity to take British politics into the twenty-first century.

Methodology

There is no precise way of estimating the impact of boundary changes, because parliamentary election results are not available in units smaller than the whole constituency. Preparation of ‘notional’ results for revised constituencies generally involves using local authority election results to model how the areas being transferred would have voted in the last general election. Particularly in areas with non-partisan local elections, it can lead to errors even when undertaken in a highly methodical fashion (for instance, the estimates for results in the new Scottish seats of Gordon in 1992 and Dumfriesshire CT in 2001 were clearly in error). Further, votes may change when people are moved into a seat where the tactical position is very different – for instance people formerly in the Lewes constituency who may have voted Lib Dem there may well have started voting Labour once transferred into marginal Brighton Kemptown in 1997.

These estimates are fairly rough calculations for the most part (although detailed working has been done in some cases e.g. Harlow), and are based as is usual on local election patterns. Where the conclusion is particularly uncertain this has been noted. Although individual estimates might be inaccurate, errors may well balance out overall.

Alternative electoral system estimates

Throughout these estimates, no change in voting behaviour has been assumed even though such changes would be practically guaranteed under a different electoral system.
Because votes for individual and party are indistinguishable under FPTP, they have been assumed to be interchangeable.This is to say that the Labour list vote would be the aggregate of the votes cast for Labour candidates, and that the votes for Labour in a multi- member seat would be the same as the sum of the votes cast for Labour in the component FPTP seats.

Lists and AMS

National and regional list PR (and thus large top-up AMS) estimates are relatively simple to model on these steady- state assumptions. Aggregate votes for a party’s candidates are translated into list votes on a national or regional basis and then translated into seats using largest remainders (for the ‘pure’ model) and D’Hondt (for everything else).

AV

The polling data used in the estimates was BBC/NOP surveys 30 April and 1 May 2005. It suggested the following pattern of second preferences (read down and then across the table).

I additionally made the assumptions that UKIP votes would flow heavily (60%) to Conservative and that Green votes would flow heavily to Lib Dem (80%) or Labour (60%) against the Conservatives and Lib Dem (60%) over Labour. These assumptions, particularly the one relating to UKIP and the non-reallocation of transfers to ‘Other’, probably favour the Conservatives and reduce the effect of AV compared to other sets of assumptions. With less of a UKIP transfer rate and less leaky transfers between Lib Dem and Labour it is easy to envisage how a calculation giving fewer Conservative seats, and adding somewhat to the Labour majority, might be made.

The following 19 Conservative seats under FPTP would be won by Labour under AV: Clwyd West, Croydon Central, Dumfriesshire CT, Preseli Pembrokeshire, Reading East, Scarborough & Whitby, Shipley, Shrewsbury & Atcham, St. Albans (all by comfortable margins) and rather more narrowly Forest of Dean, Gravesham, Hemel Hempstead, Hornchurch, Milton Keynes North East, Putney, Rugby & Kenilworth, Scarborough & Whitby, Wellingborough and Wimbledon

The following 4 Conservative seats under FPTP would be won by the Lib Dems under AV:

Eastbourne, Guildford, Ludlow and Weston-super-Mare.

The following 8 Labour seats under FPTP would be won by the Lib Dems under AV:

Aberdeen South, Edinburgh South, Islington South & Finsbury, Leicester South, Norwich South, Oxford East, Watford, and possibly Edinburgh North & Leith.

The following Labour seat under FPTP would probably be won by the SNP under AV:

Ochil & South Perthshire.

The following Respect seat would probably be won by Labour under AV:

Bethnal Green & Bow.

The following SDLP seat would probably be won by the UUP under AV:

Belfast South.

AV+

The top-up constituencies are the same as those given in Annex C to the original Jenkins Report, except in Scotland where the new Parliamentary boundaries meant they had to be adapted to contain constituencies as follows:

• Edinburgh (6+1): The five Edinburgh city seats plus Linlithgow & EF, Livingston
• Glasgow (6+1): The seven Glasgow city constituencies
• Highlands (5+1): Argyll & Bute, Caithness SER, Inverness NBS, Na h-Eileanan an Iar, Orkney & Shetland, Ross SI pMid Scotland and Fife (7+1): Dunfermline & WF, Falkirk, Fife NE, Glenrothes, Kirkcaldy & Cowdenbeath, Ochil & SP, Perth & NP, Stirling
• North Clyde (6+1): Airdrie & Shotts, Coatbridge CB, Cumbernauld KKE, Dunbartonshire E, Dunbartonshire W, Lanark & HE, Motherwell & Wishaw
• North East (7+2): Aberdeen N, Aberdeen S, Aberdeenshire WK, Angus, Banff & Buchan, Dundee E, Dundee W, Gordon, Moray
• South (7+1): Ayr CC, Ayrshire Central, Berwickshire RS, Dumfries & Galloway, Dumfriesshire CT, East Lothian, Kilmarnock & Loudoun, Midlothian
• South Clyde (6+1): Ayrshire NA, East Kilbride SL, Inverclyde, Paisley & RN, Paisley & RS, Renfrewshire E, Rutherglen & HW

Estimates of the composition of the enlarged constituencies within each AV+ top-up region are contestable in that there are different acceptable schemes for most areas, and also the workings of eliminations and transfers under AV needs to be estimated. Working these through for imaginary constituencies would inject false precision into the exercise, so the calculations are approximate but defensible. In some cases variations do not end up making any difference because they would be cancelled out through the workings of the top-up. For this reason the political complexion of the AV seats and top-up seats is not given in the main text, but for the interested reader they are tabulated below. In summary, the AV seats in Britain would fall something like 292 to Labour, 155 to Conservative, 64 to Lib Dem and 10 to others.The compensating top-ups would be allocated 15 to Labour, 44 to Conservative, 46 to Lib Dems and 2 to others (SNP and Respect).

STV

Constituencies have been designated involving 3-5 members. Boundaries between the English regions have always been respected in drawing up the STV constituencies. Except in two cases (the ex-Somerset portion of Avon and the peculiar case of the Isle of Wight) county boundaries have not been breached, and only in one case (Hammersmith & Fulham) has a London borough been divided.

South West region
Cornwall: entire county of Cornwall
Plymouth: Plymouth Devonport, Drake and Devon SW South Devon:Teignbridge,Torbay, Totnes
North Devon: Devon E, Devon N, Devon W&T, Exeter, Tiverton & Honiton
South Somerset: Bridgwater, Somerton & Frome,Taunton, Yeovil
North Somerset: Bath, Wansdyke, Wells, Weston-SM, Woodspring
Bristol: Bristol E, S, W
South Gloucestershire: Bristol NW, Kingswood, Northavon Mid Gloucestershire: Forest of Dean, Gloucester, Stroud North Gloucestershire: Cheltenham, Cotswold,Tewkesbury Dorset: Dorset MNP, N, S, W
Bournemouth: Bournemouth E, W, Christchurch, Poole South Wiltshire: Devizes, Salisbury, Westbury
North Wiltshire: Swindon N, S, Wiltshire N

South East region
Wight and New Forest: Isle of Wight, New Forest E, W Southampton: Eastleigh, Romsey. Southampton Itchen,Test Mid Hampshire: Fareham, Gosport, Hampshire E, Winchester
Portsmouth: Havant, Portsmouth N, S
North Hampshire: Aldershot, Basingstoke, Hampshire NE, NW
West Berkshire: Newbury, Reading E, W, Wokingham East Berkshire: Bracknell, Maidenhead, Slough, Windsor North West Oxfordshire: Banbury, Wantage, Witney Oxford and Henley: Henley, Oxford E, WAB
North West Sussex: Crawley, Horsham, Sussex Mid South West Sussex: Arundel SD, Bognor RL, Chichester,
Worthing ES, W
Brighton and Lewes: Brighton K, P, Hove, Lewes
East Sussex: Bexhill & Battle, Eastbourne, Hastings & Rye, Wealden
East Kent: Canterbury, Dover, Folkestone & Hythe,Thanet N, S
Mid Kent: Ashford, Faversham & MK, Maidstone & The Weald, Sittingbourne & Sheppey
Medway: Chatham & Aylesford, Gillingham, Gravesham, Medway
West Kent: Dartford, Sevenoaks,Tonbridge & Malling, Tunbridge Wells
North Surrey: Esher & Walton, Runnymede & Weybridge, Spelthorne
West Surrey: Guildford, Surrey Heath, Surrey SW, Woking East Surrey: Epsom & Ewell, Mole Valley, Reigate, Surrey E

London region
London – by boroughs, except for City & West End which consists of the City, Westminster, Kensington & Chelsea and the constituency of Hammersmith & Fulham (not the whole borough).The Ealing & Hounslow seat therefore contains part of north Hammersmith. East End is Newham and Tower Hamlets.

Eastern region
North Bedfordshire: Bedford, Bedfordshire Mid, NE South Bedfordshire: Bedfordshire SW, Luton N, S
West Hertfordshire: Hemel Hempstead, Hertfordshire SW, Watford
Mid Hertfordshire: Hertsmere, Hitchin & Harpenden, St Albans, Welwyn Hatfield
East Hertfordshire: Broxbourne, Hertford and Stortford, Hertfordshire NE, Stevenage
South Cambridgeshire: Cambridge, Cambridgeshire S, SE

North Cambridgeshire: Cambridgeshire NE, NW, Huntingdon, Peterborough
Southend: Castle Point, Rayleigh, Rochford & Southend E, Southend W
South Essex: Basildon, Billericay,Thurrock
West Essex: Brentwood & Ongar, Epping Forest, Harlow Mid Essex: Braintree, Chelmsford W, Maldon CE, Saffron
Walden
North Essex: Colchester, Essex N, Harwich West Suffolk: Bury St Edmunds, Suffolk S, W
East Suffolk: Ipswich, Suffolk Central, Coastal, Waveney North West Norfolk: Norfolk Mid, N, NW, SW
South East Norfolk: Norfolk S, Norwich N, S, Great Yarmouth

East Midlands region
Northampton: Daventry, Northampton N, S
East Northamptonshire: Corby, Kettering, Wellingborough West Leicestershire: Bosworth, Charnwood, Leicestershire
NW, Loughborough
South Leicestershire: Blaby, Harborough, Rutland & Melton Leicester: Leicester E, S, W
Derbyshire Peaks: Amber Valley, Derbyshire W, High Peak Derby: Derby N, S, Derbyshire S, Erewash
Chesterfield: Bolsover, Chesterfield, Derbyshire NE Holland: Boston & Skegness, Grantham & Stamford, Sleaford & NH, South Holland & The Deepings Lindsey: Gainsborough, Lincoln, Louth & Horncastle
North Nottinghamshire: Bassetlaw, Newark, Sherwood West Nottinghamshire: Ashfield, Broxtowe, Mansfield Nottingham: Gedling, Nottingham E, N, S, Rushcliffe

West Midlands region
Hereford and Worcester: Hereford, Leominster, Worcester, Worcestershire W
Worcestershire: Bromsgrove, Redditch, Worcestershire Mid, Wyre Forest
Warwickshire: Nuneaton, Rugby & Kenilworth Stratford, Warwick & Leamington
Coventry: Coventry NE, NW, S
Meriden: Meriden, Solihull, Warwickshire North Birmingham East: Hodge Hill, Sparkbrook SH,Yardley
Birmingham North: Erdington, Ladywood, Perry Barr, Sutton Coldfield
Birmingham South: Edgbaston, Northfield, Selly Oak, Hall Green Walsall: Aldridge-Brownhills, Walsall N, S Wolverhampton: Wolverhampton NE, SE, SW Dudley: Dudley N, S, Halesowen and RR, Stourbridge Sandwell: Warley, West Bromwich E, W
Shropshire: entire county of Shropshire including Telford & Wrekin
South West Staffordshire: Cannock Chase, Stafford, Staffordshire South, Stone
South East Staffordshire: Burton, Lichfield, Tamworth
North Staffordshire: Newcastle-under-Lyme, Staffordshire Moorlands, Stoke-on-Trent Central, N, S

North West region
Wirral: Birkenhead, Wallasey, Wirral S, W East Cheshire: Congleton, Macclesfield, Tatton
West Cheshire: Chester, Crewe, Eddisbury, Ellesmere PN North Cheshire: Halton, Weaver Vale, Warrington N, S
St Helens and Knowsley: Knowsley NSE, S, St Helens N, S Liverpool: Liverpool Garston, Riverside, Walton, Wavertree,
West Derby
Sefton: Bootle, Crosby, Southport
Lancaster and Fylde: Blackpool NF, S, Fylde, Lancaster & Wyre, Morecambe & Lunesdale
South West Lancashire: Chorley, Lancashire West, Preston, Ribble South
South East Lancashire: Blackburn, Hyndburn, Rossendale & Darwen
Mid Lancashire: Burnley, Pendle, Ribble Valley
Manchester: Manchester Blackley, Central, Gorton, Withington Trafford and Wythenshawe: Altrincham & Sale W, Stretford
& Urmston, Wythenshawe & Sale E Stockport: Cheadle, Hazel Grove, Stockport
Tameside: Ashton under Lyne, Denton & Reddish, Stalybridge & Hyde
Wigan: Leigh, Makerfield, Wigan Bolton: Bolton NE, SE, W
Bury and Middleton: Bury N, S, Heywood & Middleton Oldham and Rochdale: Oldham ES, WR, Rochdale Salford: Eccles, Salford, Worsley
Cumberland: Carlisle, Penrith ATB, Workington Westmorland: Barrow & Furness, Copeland, Westmorland
& Lonsdale
Yorkshire and the Humber region
Rotherham: Rotherham, Rother Valley, Wentworth Doncaster: Barnsley EM, Doncaster Central, N, Don Valley

Barnsley and Hillsborough: Barnsley Central, WP, Sheffield Hillsborough
Sheffield: Sheffield Attercliffe, Brightside, Central, Hallam, Heeley
Bradford: Bradford N, S, W, Keighley, Shipley
Calder and Colne: Calder Valley, Colne Valley, Halifax, Huddersfield
Wakefield: Hemsworth, Normanton, Pontefract & Castleford, Wakefield
Dewsbury and Morley: Batley & Spen, Dewsbury, Morley & Rothwell
Leeds East: Elmet, Leeds E, NE
Leeds West: Leeds Central, NW, W, Pudsey
North Lincolnshire: Brigg & Goole, Cleethorpes, Great Grimsby, Scunthorpe
Hull: E, N, WH
East Yorkshire: Beverley & Holderness, Haltemprice & Howden,Yorkshire E
North West Yorkshire: Harrogate & Knaresborough, Richmond, Skipton & Ripon,Vale of York
York and Scarborough: Ryedale, Scarborough & Whitby, Selby,York

North East region
Northumberland: Berwick, Blyth Valley, Hexham, Wansbeck North Tyneside: Newcastle EW, North Tyneside, Tynemouth Newcastle & Gateshead: Blaydon, Gateshead EWW,
Newcastle Central, N, Tyne Bridge
Sunderland: Houghton EWW, Jarrow, South Shields, Sunderland N, S
North Durham: Durham City, N, NW, Easington
South Durham: Bishop Auckland, Darlington, Sedgefield Stockton and Hartlepool: Hartlepool, Stockton N, S Middlesbrough: Middlesbrough, Middlesbrough SCE, Redcar

Wales
Newport and Monmouth: Monmouth, Newport E, W Gwent Valleys: Blaenau Gwent, Islwyn, Torfaen
Cardiff: Cardiff Central, N, SP, W
South Glamorgan: Bridgend, Ogmore, Pontypridd, Vale of Glamorgan
Glamorgan Valleys: Caerphilly, Cynon Valley, Merthyr TR, Rhondda
West Glamorgan: Aberavon, Gower, Neath, Swansea E, W Carmarthen and Pembroke: Carmarthen ED, WSP, Llanelli,
Preseli P

Mid Wales: Brecon & Radnorshire, Ceredigion, Montgomery Gwynedd: Caernarfon, Conwy, Meirionnydd NC,Ynys Mon North West Clwyd: Clwyd W, Delyn, Vale of Clwyd
South East Clwyd: Alyn & Deeside, Clwyd S, Wrexham

Scotland
Dumfries and Ayr: Ayr CC, Ayrshire C, Dumfries G, Dumfriesshire CT
Berwick and East Lothian: Berwickshire RS, East Lothian, Midlothian
Edinburgh: 5 city seats
Falkirk and West Lothian: Falkirk, Linlithgow, Livingston Argyll and Inverclyde: Argyll & Bute, Ayrshire NA, Inverclyde Renfrewshire: Paisley RN, Paisley RS, Renfrewshire E Glasgow: NE, E, Central, S, SW
Dunbartonshire: Cumbernauld KKE, Dunbartonshire E, W, Glasgow N, NE
Kilmarnock & Hamilton: East Kilbride, Kilmarnock L, Lanark HW, Rutherglen HE
North Lanarkshire: Airdrie & Shotts, Coatbridge CB, Motherwell and Wishaw
Fife: Dunfermline, Fife NE, Glenrothes, Kirkcaldy and Cowdenbeath
Perthshire and Stirling: Ochil PS, Perth NP, Stirling Angus: Angus, Dundee E, W
Aberdeenshire: Aberdeen N, S, Aberdeenshire WK, Banff & Buchan, Gordon
West Highlands: Caithness SER, Ross SL, Western Isles East Highlands: Inverness NBS, Moray, Orkney & Shetland

Northern Ireland
Antrim and Lagan Valley: Antrim E, N, S, Lagan Valley Belfast: Belfast E, N, S, W
Down and Armagh: Down N, S, Newry & Armagh, Strangford, Upper Bann
West of the Bann: Fermanagh & ST, Foyle, Londonderry E, Tyrone W, Ulster Mid

Parties with one or two whole quotas in a constituency are assumed to win at least a seat per whole quota.

Parties with 0.9 or more of a quota are awarded a seat for that quota unless it conflicts with the rule about whole quotas.

These rules distribute all seats, or all but one seat, in the vast majority of constituencies.

Allocating the last seat is as much an art as a science. Preference is given to transfer-friendly parties – Lib Dems first, then Labour and Nationalists, then Conservatives (except when there is a UKIP or BNP vote to be transferred). Where one party is clearly ahead once the whole quotas have been knocked out, and it would take an extraordinarily efficient pattern of transfers to overtake that party, that party’s candidate is deemed to win the last seat. In the very fine cases, I have attempted not to make systematic errors.