Memorandum submitted by The Electoral Commission
1. INTRODUCTION
1.1 The Electoral Commission welcomes the
opportunity to present further evidence to the Northern Ireland
Affairs Committee on the new arrangements for voter registration
in Northern Ireland introduced in September 2002. We believe voter
registration is the lynchpin of the electoral system and changes
in registration and in particular individual registration have
potentially far reaching implications for the electorate and political
parties alike. Consequently, the Commission has been keen to monitor
developments in Northern Ireland in order to learn lessons for
the future.
1.2 Since the Electoral Fraud (Northern
Ireland) Act 2002 became law in May 2002 the Commission has monitored
its impact on voter registration. To date we have published two
reports on the subject, one examining the impact of the legislation
after one year and the other examining how the legislation worked
at the Northern Ireland Assembly election held in November 2003.
Copies of both reports have previously been circulated to the
Committee and executive summaries of both are attached to this
paper.
1.3 More recently we commissioned PricewaterhouseCoopers
to conduct further desk research examining the fluctuations in
the register between September 2003 and May 2004. A summary of
the research find ingsis set out in section three of this paper.
We will be conducting further work in this area when the December
2004 register is published.
2. THE ELECTORAL
COMMISSION
2.1 Following the enactment of the Political
Parties, Elections and Referendums Act 2000 (PPERA), The Electoral
Commission was established on 30 November 2000 as an independent
statutory body covering the whole of the United Kingdom. The Commission
is headed by a Chairman with four other Commissioners. The Chairman
and Commissioners do not have connections to any political party,
nor is the Commission accountable to the Government. It reports
directly to Parliament through a committee chaired by the Speaker
of the House of Commons.
2.2 The Commission is responsible for overseeing
a number of aspects of electoral law, including the registration
of political parties and third parties, monitoring and publication
of significant donations to registered political parties and holders
of elective office, and the regulation of party spending on election
campaigns. The Commission also has a role in advising those involved
in elections on practice and procedure and is required to report
on the administration of every major election. However, unlike
many electoral commissions outside the UK, the Commission does
not have responsibility for maintaining and updating electoral
rolls, employing electoral services staff, or conducting parliamentary
or local elections.
2.3 The Commission aims to:
promote and maintain openness and
transparency in the financial affairs of the UK's political parties
and others involved in elections;
review the administration and law
of elections and encourage good practice;
encourage greater participation in
and increased understanding of the democratic process;
ensure that the Commission is able
to undertake the effeQtive conduct of a referendum;
provide for electoral equality in
each local authority area in England while also reflecting community
identity and interests;
carry out all the Commission's statutory
functions impartially using resources efficiently, effectively
and economically.
2.4 In addition to the UK Headquarters in
London, the Commission has three devolved offices serving Scotland,
Wales and Northern Ireland. The Commission's office in Northern
Ireland has responsibility for delivering the corporate aims in
the context of Northern Ireland. It is also able to ensure that
the Northern Ireland dimension is fully represented in the strategic
thinking and operational planning of the Commission. The focus
of the office is very much on the relationship with political
parties, developing and advising on electoral policy and practice
and raising awareness of electoral matters. It does not have an
operational role in maintaining the register of electors or conducting
elections since this falls to the Electoral Office for Northern
Ireland (EONI) and the Chief Electoral Officer. The Commission
has a statutory duty to report on the administration of elections
held in Northern Ireland (other than local elections) and its
first statutory report as published in April 2004.
2.5 The Commission's Northem Ireland office
is located in Belfast and has five staff. Since its establishment
the Commission has developed strong links with a range of stakeholders,
including the political parties, the Electoral Office for Northern
Ireland (EONI), the Chief Electoral Officer, the Northern Ireland
Office, academics with an interest in electoral and political
matters and the voluntary and community sectors.
2.6 The Commission has also established
an Assembly Parties Panel along the lines of the UK Parliamentary
Parties Panel. The Panel meets on a quarterly basis and is chaired
by the Commissioner with a specific interest in Northern Ireland.
All parties represented in the Northern Ireland Assembly are members
and its minutes are made public. The purpose of the panel is to
facilitate discussion in areas of mutual interest between the
political parties and The Electoral Commission. The Panel acts
as a forum for sharing information on electoral matters and has
been central to developing good working relationships with all
the political parties. The Chief Electoral Office for Northem
Ireland and his senior staff attend meetings of the panel.
3. PRICEWATERHOUSECOOPERS
UPDATE REPORT
ON THE
NORTHERN IRELAND
ELECTORAL REGISTERS
3.1 In addition to wide-ranging consultations
with a cross-section of stakeholders, the report published by
the Commission in December 2003 drew upon independent research
reports designed to provide the Commission with a better understanding
of how the new registration processes had affected the voter franchise
in Northern Ireland.
3.2 These independent research reports were
commissioned in order to assist The Electoral Commission in meeting
its statutory duty to keep under review and, from time to time,
submit reports to government on electoral law and practice. As
a continuation of this process, and to assist in continuing to
monitor the trends in the electoral register, The Electoral Commission
appointed PricewaterhouseCoopers to undertake further desk research
on the trends and patterns in electoral registration in Northern
Ireland. This is the first report to be prepared in this second
stage of the desk research statistical analysis of the electoral
register. It focuses on:
The September 2003 register, which
was the basis for the November 2003 Northern Ireland Assembly
Elections, the first elections to make use of the new register.
The February 2004 register, which
was the second register compiled under the new system following
an annual canvass.
The May 2004 register, which was
used for the June 2004 European Elections.
3.3 The key issue addressed is the decline
in the Northern lreland electoral register from a peak of 1,098,726
in May 2003 to 1,076,937 by May 2004. The framework within which
this issue is addressed is the registration cycle, comprised of
the annual canvass undertaken in autumn of each year followed
by the rolling registration process for updating the register
in the period between each annual canvass.
3.4 The decline in the Electoral Register
from May 2003 to May 2004 has occurred in the face of an increasing
eligible population. Based on Northern Ireland Statistics and
Research Agency's (NISRA) Mid-Year Population Estimates, the population
aged 18+ is growing at about 0.7% per annum.
3.5 When the change in the register is assessed
from one registration cycle to another, the main finding is that
the evidence points to an emerging downward trend in the electoral
register. While the number of observations is limited, since the
new system has only been in operation for a little over one and
a half years, nonetheless the available "like-for-like"
comparisons indicate that the register is falling by about 1.5-2%
per annum.
3.6 A similar trend can be seen in each
of the 18 parliamentary constituencies, albeit the trend varies
from -1% in a number of rural constituencies and those surrounding
Belfast to -3 to -4% in the urban constituencies of Belfast and
Foyle. There is therefore a clear urban-rural/suburban divide
in the emerging trend in the electoral register.
3.7 The emerging downward trend comprises
two effects:
the rolling registration effect.
3.8 The canvass effect is the more important
of the two. The electoral register is compiled afresh each year,
based on a canvass of the preexisting register. Inevitably, there
will be some element of non-response, leading to a reduction in
the Register. In the most recent annual canvass, the new February
2004 Register contained 2.6% fewer entries than the pre-existing
September 2003 Register. Each of the 18 parliamentary constituencies
also recorded a decline over the same period, as did 82%of the
582 Wards.
3.9 Counteracting the canvass effect, but
only partially, is the rolling registration bounce-back that typically
occurs in the months immediately following the annual canvass.
There is some evidence to suggest that the timing and magnitude
of the bounce-back reflects a "looming election effect",
whereby effort is expended by political parties to encourage people
to register.
3.10 The second important finding to emerge
is that deprivation is only slightly correlated with the downward
trend in the Register. Indeed, when other factors are controlled
for, deprivation is not a significant influence on the trend in
the Register. But this in turn owes much to the fact that the
rolling registration bounce-back tends to be stronger in the more
deprived areas, offsetting the negative correlation that appears
to exist between deprivation and the annual canvass effect.
3.11 The Northern Ireland registration rate
is declining in tandem with the fall in the electoral register
described above. When attainers and those living abroad are netted
out of the Register, the estimated Northern Ireland registration
rate was 84.8% in May 2004, representing a two percentage points
reduction compared to the 87% achieved in September 2003.
3.12 Though constrained by lack of more
up-to-date population data, the findings by parliamentary constituency
and the ward are broadly consistent with the analysis of the emerging
trend in the electoral register. Thus, the registration rate at
parliamentary constituency level has fallen more sharply in urban
areas, notably Belfast and Foyle, than in other areas of Northern
Ireland.
3.13 While there have been shifts, the pattern
of registration rates at ward level in May 2004 showed a considerable
degree of stability when compared with previous registers.
3.14 Unless it is rectified, the downward
trend in the register has the potential of embedding itself structurally
in the registration process. If the register is in decline, then
the number that can be canvassed Will also tend to fall from one
canvass to the next, thus reinforcing the cycle.
3.15 Other findings from the research that
have policy implications include the following:
A consistent finding, both in this
report and in the earlier July 2003 PricewaterhouseCoopers report,
is that the use of a postal canvass is usually associated with
an above-average reduction in the registration rate for the affected
ward. In other contexts, notably survey research, reliance on
postal returns tends to result in lower response rates compared
to more direct methods.
The analysis of attainers on the
Register revealed a registration rate in the region of 20-25%for
persons aged 17 at the conclusion of the annual canvass. Boosting
this registration rate would help to increase the Register, as
well as bringing young people into the system at an early age.
As in the July 2003 PricewaterhouseCoopers
report, deprivation was found not to exert a significant independent
influence on the rate of change in the register. Nonetheless,
this is largely because the rolling registration bounce-back tends
to be greater in the more deprived areas, offsetting an above-average
negative canvass effect. To the extent that this reflects a looming
election effect, this finding would give rise to a concern that
deprivation may yet emerge as a factor in registration process.
4. THE ELECTORAL
COMMISSION PUBLIC
AWARENESS CAMPAIGNS
4.1 Under Section 13 of PPERA, The Electoral
Commission has a duty to promote the public awareness of electoral
and democratic systems and therefore has a responsibility for
ensuring that the electorate in Northern Ireland is aware of and
understands the new registration process and what identification
documents are required for voting at polling stations. In the
context of Northern Ireland our public awareness campaigns to
date have been information-based with a particular emphasis on
the requirements for individual registration and photographic
identification.
4.2 The Commission's first public awareness
campaign in respect of individual registration and photographic
identification commenced in September 2002. It was multi-media
in nature and included:
television advertising;
regional and local newspaper advertising;
a dedicated freephone helpline number;
a dedicated website (www.securevourvote.com);
posters for schools, colleges and
universities;
a household leaflet distributed by
EONI canvassers (available in a range of formats and languages).
4.3 Following the annual canvass a further
campaign promoting rolling registration and the requirement for
photographic identification was conducted using television and
radio advertising, regional and local newspaper advertising and
outdoor poster advertising. The dedicated freephone helpline continued
to operate until the middle of May and was terminated after the
postponement of the 29 May Assembly elections. Altogether the
Commission spent almost £1 million promoting the new arrangements
for electoral registration and identification.
4.4 Tracking research commissioned to test
the effectiveness of the 2002 campaign was positive and a majority
of those interviewed indicated that they found the advertising
engaging and informative. A representative sample of the electorate
was asked for its views on the changes to electoral registration
before and after the Commission's public awareness campaign commenced
in Autumn 2002. Prior to the campaign commencing, 15% were aware
that changes were planned. Following the campaign nearly two thirds
(63%) were aware of the planned changes. By April 2003, 79% were
aware that changes had been made to the electoral registration
and voter identification process. Overall the publicity campaign
was reasonably effective in increasing public awareness of the
new arrangements.
4.5 Our public awareness campaign in respect
of the election commenced soon after the date of the election
was announced in October 2003. It was entitled the "turnout
essentials" campaign and emphasised what documents were permissible
to vote at the Assembly election. Another theme adopted in the
advertising campaign was that of "No picture? No vote".
The campaign was multi-media in nature and comprised:
television advertising;
regional and local newspaper advertising;
a dedicated telephone helpline;
information leaflet drop to 650,000
households;
a dedicated website (www.securevourvote.com).
4.6 This activity was supported by a range
of public relations initiatives aimed at keeping the key messages
of the campaign to the forefront of the electorate's mind in the
run-up to the election. Altogether the Commission's public awareness
campaign cost approximately £400K of which almost 75% was
spent promoting the prescribed forms of photographic identification.
4.7 Tracking research was again used to
gauge the effectiveness of our campaign. Altogether 75% of those
interviewed spontaneously recalled either seeing or hearing advertising
about photographic identification. When asked to comment on the
advertising 67% said they found it informative, 35% that it held
their attention, 32% that is made them think and 30% that it was
humorous.
5. CONCLUSIONS
5.1 The new system for voter registration
in Northern Ireland has resulted in a more robust register, but
with the potential to become less accurate with each annual canvass.
A particular focus is needed on targeting hard to reach groups
during the registration process. The Commission will be reviewing
its strategy and contribution to this process.
|