Reply to comment

Elections in London

Project Date: 
Jul 1998 - Jan 2006

In June 2004 citizens voting in London had the chance to cast five simultaneous preferences. As in May 2000, the GLA Mayor and Assembly elections allowed voters to simultaneously register up to four preferences, to differentiate between their evaluations of mayoral candidates and political parties and to signal complex preference structures to politicians. And on the same day in the 2004 European elections Londoners had the additional choice of candidates to represent them in the European Parliament.

The Greater London Elections Study, funded by the ESRC and forming part of the Devolution and Constitutional Change Programme involved an opinion poll (conducted by ICM) of 1,474 respondents in London straight after the elections on June 10th. Respondents who reported voting in the elections were asked to 'vote again', recording their vote on ballot papers similar to those used in the actual election while non-voters were asked to indicate how they would have voted had they gone to the polling station on June 10th. Respondents were also asked a number of questions about their views of parties and mayoral candidates standing in the elections, their reasons for voting or not voting and their views on various London-specific policy issues. The study replicates a similar one carried out by the same research team after the first London elections in 2000.

The project is based at the School of Public Policy, UCL. Principal investigators on the project are Patrick Dunleavy (LSE), Helen Margetts (UCL, now University of Oxford) and Simon Bastow (UCL and LSE). David Rowland works on the project as Research Fellow at the School of Public Policy, UCL and Jennifer van Heerde (also UCL) has been co-opted as a statistical expert and co-author.

Co-funding for the research was obtained from three sources:

  • London Connects funded questions on e-government in London, asking respondents whether they had access to the Internet and if so, whether they had ever used it to obtain information or services from government organizations. A question about whether respondents had used the Internet to look for or buy goods or services was also asked for comparison.
  • The Joseph Rowntree Charitable Trust funded two questions asking respondent's views of political parties, as part of a concurrent project looking at attitudes to the far-right in Britain. The first of these was a like/dislike thermometer using a seven-point scale, in which respondents were asked to say how much they 'like' or 'dislike' all the political parties standing in the London elections, including the BNP and UKIP but also the Greens and Respect. The second question asked respondents which statement more closely reflected their views concerning all the same parties: 'I might vote for them in the future' or 'I could never vote for them in the future'.
  • Further co-funding has now been obtained by the Joseph Rowntree Reform Trust to analyse the electoral performance and attitudes to the far right in London since the late 1990s. This research will take the assembly and mayoral election results at ward level for 2004 and 2000, combined with a sample of the local election results in 1998 and 2002 to analyse the origins and implications of support for far-right parties (specifically the BNP and UKIP) in London. The research will include two focus groups in the east of London where support for these parties was particularly high in the mayoral and assembly elections.


  • As in 2000, we found that voters responded to their expanded electoral choice in a sophisticated way. For example, for the assembly our poll suggests that between 40 and 43 per cent of voters split their tickets, considerably higher than the 33 per cent we estimated in 2000 (Dunleavy, Margetts and Bastow, 2002). This finding suggests that the extent of ticket splitting in London was not a one-off consequence of Livingstone's expulsion from the Labour party, but a permanent feature of voting behaviour in London which has survived Livingstone's re-entry to the party.
  • There is reason to believe that the London political system has reached a mature state, as in Scotland, forming part of the emerging pattern of British politics in transition under the impact of co-existence between plurality rule and a range of more proportional electoral systems, with 5.8 effective parties in the Assembly election, up from 4.7 in 2000 and far beyond our predictions of 2.7 in 1997 (Dunleavy and Margetts, 1998a). The underlying multi-party patterns of British voters' alignments have emerged as strongly in London as they have in Scotland and Wales, indicating that there is no sense in which the results should be discounted or dismissed as the product of 'second-order' elections.
  • We found higher than expected levels of potential support for the far-right in London. We found that 23 per cent of respondents to our poll said that they 'might vote' for the BNP in the future, corroborating the JRRT State of the Nation Poll which found that 24 per cent of Londoners from a national sample responded in a similar way. Although the BNP was easily the most unpopular party, more than a quarter of respondents had voted or said they would vote for the BNP in the European, mayoral or London Assembly elections or said they identified with it, or said they might vote for it in the future. Among 18- to 24-year-olds, nearly 10 per cent had given one of their preferences to the BNP in 2004.
  • In general, more Londoners appear to think that various aspects of London life have improved than worsened since the mayor and assembly were introduced. On a range of policy issues, we asked whether respondents thought things were 'a lot better', 'a little better', 'about the same', 'a little worse' or 'a lot worse' since the last London elections in 2000. Of those who expressed an opinion, results were as follows, with over half of respondents considering that traveling on buses and traffic congestion had improved since 2000:
      Worse % The same % Better % Total %
    Traveling on Buses 19 26 55 100
    Traffic Congestion 25 25 51 100
    Tackling Crime 26 37 37 100
    Built Environment 24 49 27 100
    Litter on the Streets 36 40 24 100
  • Voters' reasons for voting in the London elections appeared to be concerned with civic obligation and candidate selection rather than national issues or party loyalty. We asked voters what was most important to them when voting in the elections for the London Assembly and London Mayor, giving a range of possibilities from which they were asked to select the two most important. For comparison, we asked the same or similar questions of voters in the European election. Of those offered to London voters, 'feeling it was a duty to vote' and 'choosing the best people to run London' were clear winners, while those choosing parties or national issues lagged far behind further evidence that voters do not regard these elections as second order elections and a weakening of party identification. Responses across the London and European elections were strongly consistent.
      London Elections European Election
    I felt it was my duty to vote 53.4 55.5
    Choosing the best people to
    run London/represent Britain's
    interests in the EU
    51.4 50.7
    I wanted to support a particular party 26.9 26.9
    These elections were a chance to let the national government know what you think about
    national issues
    24.3 24.5
    I wanted to let the government know my view on the Iraq war 12.0 10.8

For non-voters in the London elections, we asked people about their reasons for not going to vote. By far the most cited reason was forgetting or not having time to go and vote (53 per cent), while all other possible reasons lagged behind at 17 per cent or below. Only 7 per cent cited 'The London mayor and assembly cannot improve things in London' as a reason for not voting.

More detailed findings for 2004 and 2000 can be accessed as follows:

  • Headline findings of the poll were presented at the PSA EPOP conference, University of Oxford, September 10-12th 2004 in a paper entitled 'Political Alignments in the 2004 London Elections', by Patrick Dunleavy, Helen Margetts and Jennifer van Heerde.

     Download: Political Alignments in the 2004 London Elections  (198kb/PDF)

  • Findings relating to the far-right were reported in a further paper to the EPOP conference in a paper entitled 'Latent Support for the Far-Right in British Politics: the BNP and UKIP in the 2004 European and London elections' (download long paper on Far Right, sent by HM to AC in PDF), by Helen Margetts, Peter John and Stuart Weir which was later summarised for an article by the same authors published in the New Statesman on 24th January 2005 (download New Statesman article, HM sent to AC in PDF). Helen Margetts were interviewed on the Today programme on 24th January 2005 (the research was also covered in the Times and the Guardian, 24th -- 25th January).

     Download: Latent Support for the Far-Right in British Politics:
     The BNP and UKIP inthe 2004 European and London Elections

     Download: New Statesman Article (January 2005)  (22kb/PDF)

  • Findings relating to Internet use, e-government, e-democracy and digital inclusion was presented by Helen Margetts to over 50 policy-makers from across London local government and voluntary sectors at a forum at City University organized by London Connects on 30th November 2004.

     Download: Using e-government in government in London  (113kb/PDF)

  • More detailed findings of the poll and results will be presented at the Political Studies Association Conference in Leeds in April 2005 in a paper entitled 'Explaining Voters' Choices, London 2004' by Patrick Dunleavy, Jennifer van Heerde and Helen Margetts


The content of this field is kept private and will not be shown publicly.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd>
  • Lines and paragraphs break automatically.

More information about formatting options

By submitting this form, you accept the Mollom privacy policy.