INTRODUCTION
Elections are the prime way in which the political will of a community is expressed. Local elections allow local communities to set the political agenda for their local area. The more people that vote, the greater the democratic legitimacy of the elected local authority and the greater its scope to claim a mandate for its actions.
Notwithstanding their apparent importance, the recent record of participation at English local elections has been poor. Turnout is much lower (sometimes less than half the level) than at British general elections and also compares unfavourably with sub-national elections in other advanced democracies. Moreover, turnout levels have recently shown signs of dropping sharply from an already low base. In 1998 turnout at the local elections in England was just 28.8% compared to 41.5% at the same point in the local electoral cycle four years previously. Declining participation is also an issue internationally, and not solely at the sub-national level.
The Government's modernisation agenda, outlined in the White Paper Modern Local Government - In Touch with the People (TSO 1998), should be seen in this context. One of the key objectives of the Government is to enhance local democracy through both representative and participative means. A study of the nature and scope of non-electoral public participation in local government has already been undertaken for the DETR (Lowndes et al 1998). The Representation of the People Act 2000 provides for the piloting of various new procedures to facilitate the act of voting.
The programme of research reported here is designed to draw together the latest evidence about the relationship between electoral procedures and turnout. As such it forms a crucial baseline for any future studies that might be undertaken to evaluate the impact of local authority pilots of new electoral procedures on levels of participation.
PREVIOUS RESEARCH
A number of recent studies have tried to shed light on trends in local electoral turnout and to explore what might be done to improve it. Miller's analysis of surveys conducted in association with the Widdicombe Committee of Inquiry into the Conduct of Local Authority Business found that a person's age, length of residence, sense of psychological involvement in politics and the strength of their party identification were all critical factors in determining the propensity to vote at local elections (Miller 1988). However, very few people appeared not to vote because of alienation from the democratic process. Instead, those who abstained at one set of local elections could in principle be persuaded to vote in another given an appropriate, and stimulating set of circumstances.
A secondary analysis of local authority election statistics for the Department of the Environment published in 1994 concluded that turnout levels were related to aspects of the socio-economic and political character of an area (Rallings and Thrasher, 1994). Affluent neighbourhoods were more likely to register high turnouts than deprived ones. Vigorous competition between the parties and some uncertainty about the result of the election were also positively associated with turnout. By contrast, in local authorities where one party dominated, turnout tended to be depressed.
A closer examination of the ebbs and flows apparent in local authority turnout records further suggested the kinds of factors that could provide an incentive for voters. The proximity of a general election appeared to galvanise the local electorate - perhaps by way of enhanced party activity and media coverage - and lead to an increase in turnout. Equally, election weariness readily seemed to set in with turnout falling at the local elections immediately following a national contest. More substantive political conflict also appeared to account for the general rise in turnout in 1990 at the height of the controversy over the 'poll tax'. There were also localised and short-lived instances of high participation in Liverpool, Westminster, Bradford, Basildon and Tower Hamlets at different points during the 1990s (Rallings and Thrasher 1997).
These earlier studies were able to identify some of the more clear-cut determinants of local election turnout. Factors such as the socio-economic composition and the nature of political competition within a local authority helped to explain a good proportion of the variance in turnout levels between different authorities. There remained, however, widespread and unexplained differences between them.
This gap encouraged a further line of inquiry. It was hypothesised that the process of conducting local elections, from the way the register is compiled to the encouragement given to electors to use their vote, could have a significant effect on turnout. The proposition was tested by way of a survey of current electoral practice among local authorities funded by the Joseph Rowntree Foundation and published in 1996 (Rallings, Thrasher and Downe 1996). The research highlighted the myriad ways in which local authorities had tried to enhance local electoral turnout, but concluded that such campaigns and initiatives had met with, at best, only very limited success.
RESEARCH OBJECTIVES
This research both builds upon and extends the knowledge gained from these previous studies. It has a number of key objectives:
- To provide an up-to-date picture of local authority initiatives and practices designed to enhance electoral registration and participation;
- To evaluate what effect such practices appear to have on levels of registration and participation;
- To determine what appear to be the key drivers for local election turnout;
- To review electoral practices and turnout from the perspective of other countries with different systems and procedures;
- To determine what lessons, derived from these sources, can usefully be applied by local authorities and central government in order to raise local election turnout.
RESEARCH METHODS AND PRESENTATION
These objectives were pursued through a variety of research methods:
- A postal questionnaire was sent to all local authorities in England with responsibility for compiling the electoral register in the spring of 1999. It asked for responses on a wide range of topics and was designed as an audit of electoral practices and outcomes. The survey achieved a response rate of 71%. The findings of the survey are reported in Chapter 1. More details of response rate and a copy of the survey instrument can be found at Appendix 1;
- A comprehensive and rigorous statistical analysis of the relationship between local election turnout and a wide range of political, structural and socio-economic variables - including some derived from the survey of local authorities - was conducted. Using data at both local authority and ward level, this covered all local elections in England during the period 1973 to 1999. Sophisticated multi-variate techniques were used to produce a model of local election turnout. These are outlined and analysed in Chapter 2;
- Case study research was conducted in 8 local authority areas during the autumn of 1999. Sites for the case studies were selected following the statistical analysis of all authorities. Those chosen appeared to display unexpectedly high or low patterns of turnout for their local authority type. The case studies are described in Chapter 3;
- International comparisons were based on both a secondary review of the relevant literature and an extensive programme of 'internet' research. The latter proved a hugely fertile resource of primary material on a wide range of countries and systems. The findings are set out in Chapter 4.
The final chapter summarises and then draws together these different strands. In particular it concentrates on the policy implications of the research.
Published 3 May 2000
Return to Local and Regional Government Index
Return to Homepage
Web site terms