---

CHAPTER 1

Survey of local authority practices and initiatives on electoral registration and participation

This section presents the findings of a postal survey of English local authorities designed to investigate current local level practices and initiatives to enhance electoral registration and participation. The survey was conducted during the spring of 1999 and extends and updates a previous study for the Joseph Rowntree Foundation (Rallings, Thrasher and Downe, 1996). Seventy-one per cent of local authorities returned a completed questionnaire, with respondents being broadly representative of the different types of authority - see Appendix 1. Respondents were asked to frame their answers with reference to the compilation of the Register which came into effect on 16th February 1999 and to the most recent local election in their authority.

• The results of the survey are detailed in the following sub-sections:
• Staffing and Finance.
• The Electoral Registration Process.
• Preparing for the Election.
• Advertising the Election.
• Influences on Electoral Participation.
• Change in Practice.
• The Future.
• Conclusion.

Within each set of questions different types of information were elicited. It is important to distinguish between matters of fact and more anecdotal types of evidence. We have included a good deal of material which reflects the views and judgement of local authority staff. We believe that this gives a flavour of what is happening within local authorities and some important pointers to which initiatives may or may not be successful. However, it should be stressed that very few of these interpretations can be subjected to formal testing.

STAFFING AND FINANCE

Local authorities were asked to provide a range of information about how electoral services were staffed and financed.

Staffing
The average number of full time staff who work on electoral matters in a local authority is two. The actual number varies from none in 15 authorities, where the work is covered by part-time staff, to 9 in one of England's largest cities. There is a wide variation in staffing levels between authorities of a similar type and size. For example, in one unitary authority there is one full-time member of staff (no part-time staff) for an electorate of 96,313, whereas in another unitary authority 4 full-time staff serve a smaller electorate of 78,804. In 8.7% of cases the number of full-time staff has increased compared to five years ago, but in over a quarter of cases (25.7%) the number of full-time staff working directly on electoral matters has fallen over the same time period. In several authorities the number of full-time staff cannot be clearly defined as officers who work in the elections unit also have other responsibilities (e.g. Land Charges). The situation in one authority helps to explain:

'In addition to the two full time staff there is one other member of staff who spends a significant amount of time on Registration and Election work at the appropriate times during the year. Two other officers also deal with some election work as necessary'
(North East metropolitan borough)

Use of computers
All responding local authorities reported using a computerised electoral registration system. Nearly three quarters (74%) use a PC based system, while the majority of the remaining authorities (21.6%) use the main-frame computer. Ten years ago the use of computers for administering electoral registration was not so prevalent. This increased use may help to explain some of the fall in the number of full-time staff employed.

Finance
Authorities were asked: 'What is your total expenditure (in the financial year '98/99) on electoral registration? (Please include every item - salaries, overheads, postage etc.)'. The expenditure reported by authorities varied between £26,000 in one small unitary council to £770,910 in one of England's largest cities. In order to take size into account when comparing expenditures, we divided the total expenditure for electoral registration given by authorities in our survey by the electorate according to the 1999 electoral register to produce a figure for the amount of money spent per elector (the budget:elector ratio). Where we did not receive a survey from an authority we used data from the DETR (1998 Budget Register of Electors) and divided that by the 1999 electorate. Where a survey was returned but did not include any information on budgets, we also used the DETR data. Comparisons between the data we compiled from the survey and the data from the DETR show the budget:elector ratio to be a reliable and robust measure.

Table 1 shows that on average more money is spent per elector in the London boroughs (£1.61) than in other types of local authority. However, the mean figure for London hides a great deal of variation between its constituent boroughs. The lowest budget:elector ratio in London is just 35p per elector while £4.07 per elector is spent in another borough. There are also wide variations in the budget:elector ratios in the district councils (from 34p to £3.43).

Table 1:Budget: elector ratios in English local government
	Mean	Min.	Max.
London	£1.61	£0.35	£4.07
Metropolitan	£0.84	£0.28	£1.66
Unitaries	£0.98	£0.34	£1.67
Districts	£1.16	£0.34	£3.43

Within the section of the questionnaire dealing with finance, we asked local authorities: 'In the last five years, have you received any extra resources, in addition to your normal budget, for special registration campaigns?' 15% of authorities had reported receiving extra funding when we asked this question in our Rowntree research (1996). Results from the current survey show that 53 local authorities (21.7%) had received extra resources for a special registration campaign, including fifteen metropolitan boroughs and eight London boroughs. The extra budget varied between £200 for a leaflet distribution - equivalent to 0.5 pence per property, to £92,000 over three years in one London borough. Eighteen local authorities received extra resources in excess of £10,000. These included:

£38,740 - £20,000 for a supplementary canvass in 1997/8 and £18,740 in the same year for a registration awareness and voting campaign in the lead up to the Borough Council and London Referendum elections
(London borough).

£30,000 - Advertising and maps on poll cards
(North East metropolitan borough).

£50,000 - Advertising on buses, radio and newspapers. Targeted OAP Homes, Travel Agents, Estate Agents and Halls of Residences for Absent Voting using posters and leaflets
(Northern unitary authority).

£15,000 - £5,000 annually in 1996/7/8 for a canvassers' incentive bonus payment for achieving a target percentage return of Form As (South East district council).

We have no indication where these resources came from, but elsewhere in the survey several authorities noted that council and/or committee meetings had been held to agree additional funding for registration or 'use your vote' campaigns.

THE ELECTORAL REGISTRATION PROCESS

Electors must complete a Voter Registration Form (Form A) in order to be able to vote at an election taking place during the currency of the Register. The efficient despatch and collection of Form A is the key to an effective electoral registration process. This process of maximising the return of Form A, often through personal visits to households, is called the 'canvass'.

Design of Form A
A considerable number of local authorities have made changes to this form in an attempt to improve rates of return. Nearly 90% of local authorities (88.9%) use forms that are pre-printed with the names and addresses of households as opposed to asking householders to fill in their own details. More than six out of ten authorities (60.7%) that pre-print forms have only made this change in the last five years. Table 2 shows that other changes to the Form A are similarly popular. Around 80% of councils design their own form (79.8%) and have them bar-coded (83.3%). Of the responding authorities who gave a date for the introduction of bar-coding, 78.6% of them have introduced the change in the last five years.

Table 2:Features of the Voter Registration Form (Form A)
	% of auths. which	no of auths. (n)	% introducing feature in last 5 yrs. (auths. giving date only)	no of auths. (n)
Pre-printed	88.9	224	60.7	105
Bar-coded	83.3	210	78.6	136
Specially designed	79.8	201	71.1	123
Colour coded	54.0	136	74.5	82
Hand delivered	78.2	197	22.9	30
Posted	44.0	111	27.7	18

We have a number of different sources of anecdotal evidence regarding the success of each change to Form A. It was suggested that pre-printing Form A appears to encourage residents to return Form A earlier and it reduces the number of queries from the electorate. Within the Elections Office, it makes the registration process more efficient as there are fewer errors with pre-printing and better quality checks on inputting. One authority wrote,

'Pre-printing has resulted in a better response - a large number of people have been deadwooded through more accurate information being received'
(East Midlands district)

Three district councils managed to put a figure on the success of pre-printing forms,

'Introducing pre-printed Form A increased the initial response rate by 5%, thereby reducing the number of reminders'
(East Midlands district)

'Since we pre-printed names on Form A it has increased our first response from 60% to 75%'
(South West district)

'We have a higher initial response due to pre-printed forms. Improvement is about 10%'
(North East district)

Bar-coding Form A also reduces the time spent handling and inputting Form A which in turn reduces costs as fewer temporary workers are employed. Two authorities note other advantages of bar-coding,

'Bar-coding enables the forms to be dealt with more efficiently reducing the number of reminders being sent out to those who have not responded'
(South West district)

'Allows early identification of low returning polling districts, thereby allowing us to take early action to stimulate response'
(South East district)

Box 1:Evidence of the success of making design changes to Form A.

In some councils there has been a modest improvement from a high base: 'We have had an increase in returns of 1.5% over the last two years despite predictions of a population decline across the city' (Their Form A is pre-printed, specially designed, bar coded and posted) (South West district) 'Return of Form As has risen to 95.46% from approximately 93%' (The form is specially designed, bar coded and pre-printed) (North West district) In other authorities substantial increases in the rate of return of Form As have been achieved: 'There has been an annual increase in the percentage of Form A returned from less than 70% in 1987 to over 90% in 1999' (Their form is specially designed, colour coded, bar-coded and pre-printed) (London borough) 'In 1991, the response rate was 83%, in 1998 the figure was 96%' (All the initiatives were introduced between 1991 and 1994) (London borough) 'Increased percentage return of Form A (85% to 98.71% borough wide)' (West Midlands district) 'Since 1993 registration has increased by at least 7% on returns' (The form is now specially designed, bar-coded and pre-printed) (North East metropolitan borough) 'The percentage return has increased by approximately 5-8%' (The form is now specially designed, bar-coded and pre-printed) (1997) (North metropolitan borough)

A number of 'other' changes have been made to the Form A to improve its effect and appearance. One borough council includes a 'Helpline' number on the form which has helped with queries. Another respondent wrote that the colour of their Form A is changed every two years to get electors to notice the form. White envelopes for the form are used in one authority which has increased the response rate through the canvass. Around half a dozen authorities reported that they include the council logo on Form A. It was claimed that this was designed to encourage completion by residents who may recognise the logo as denoting an official communication and/or to help the authority to promote its corporate identity. A handful of authorities mentioned that their Form A has achieved the Crystal Mark from the Plain English Campaign. For example, one London borough wrote that they,

'Received the Crystal Mark for clarity in 1997. No direct evidence of effect except that the electorate has increased every year from 1991'

Best practice concerning design changes made to Form A and their impact on the electorate was summed up by one East Midlands district,

'We try something slightly different every year to bring the form to the attention of electors'

Delivery of Form A
More than three-quarters (78.2%) of authorities use a system of hand-delivery for their Form As, while more than four in ten (44%) use the post. A number of local authorities, therefore, use both methods of delivery. For instance, in one South West district the first reminder is sent by post while the second reminder is hand-delivered.

It is argued that hand delivery helps to improve the accuracy of the register,

'Canvassers get to know their area, less likely to miss new properties. Have more control over the canvass timetable'
(South East district)

'Net gain of 1,000 households on database after move to hand-delivery'
(North West metropolitan borough)

and that it is more efficient,

'Hand delivery has resulted in no complaints from residents and we now have the ability to show forms are not 'lost in the post''
(South East district)

'Additional council leaflets can be delivered with the Form A at minimal costs - that's Best Value!'
(North West metropolitan borough)

On the other hand, several rural authorities remarked that postal delivery was the only practicable option, at least in certain parts of the area.

Rate of Return of Form A
We asked four questions concerning the response rate for the return of Form As in 1998/9 - see Table 3. First, local authorities were asked what it was at the first stage of the canvass before any reminders were sent out. The results show that this figure varies widely from 30.4% to 96.9% - in both cases in London boroughs. The mean figure of 63.7% means that an average local authority has then to take additional steps to secure the return of Form As from more than a third of their households. Thirty-four authorities had a response rate at the first stage of the canvass of 50% or under.

Table 3:Percentage rate of return of Form A
	Mean	Min.	Max.
Without reminders	63.7	30.4	96.9
Final response	95.6	68.0	100
Best polling district	99.4	80.0	100
Worst polling district	83.1	22.6	100

Second, we asked local authorities for their final response rate to the canvass in 1999. This figure varied between 68% in one North West metropolitan borough to 100% claimed by 3 district councils. Twenty-four authorities replied that their response rate was 90% or under. Sixty-three authorities had a response of less than 95%. The average figure of 95.6% closely tallies with previous research indicating that the electoral register contains approximately 95% of the eligible electorate (Freeth S (1996) Compiling the Electoral Register 1995, Stationery Office: London).

The third question concerned the best response rate in a polling district in each local authority. The average figure for the response rate in the best polling district was 99.4%. The lowest response rate in this category came from a polling district in a metropolitan borough with a figure of only 80%. A total of 147 local authorities claimed to have at least one polling district with a 100% return. One South East district suggested it had 15 polling districts with a 100% return.

Finally, local authorities provided statistics on their lowest response rate in a polling district. The lowest response rate of all was 22.6% in a polling district in a North East unitary authority. Twenty-three authorities had a polling district with a response rate of less than two-thirds. At the other extreme, two districts claimed that they had a perfect electoral register by having a response rate of 100% in all polling districts: one of them arguing that a form is required from every property irrespective of the source of information. The average figure for the response rate in an authority's worst polling district was 83.1%.

Despite the extensive use of computers to administer electoral registration, 39 authorities (15.5%) were unable to provide information on which of their polling districts had the best/worst returns of Form A.

One clear difference between the 1996 Rowntree research and the current survey was a greater awareness by local authorities of their response rate for the return of Form A compared to other authorities. Comments like the following were received,

'Response rates have been approximately 98% since the introduction...(in 1996 of a Form A which is specially designed, colour-coded, bar-coded and pre-printed), which is higher than the average for metropolitan boroughs and higher than expected for the 'type' of area'
(West Midlands metropolitan borough)

'We have since 1994 had a fairly consistent return rate which exceeds the national average. It is considered that the current methodology within the current legislation is the best practice locally'
(South east unitary authority)

Chasing Up Non-Respondents
The Representation of the People Act (1983) states that, 'With a view to the preparation of registers the registration officer shall: a): have a house-to-house or other sufficient inquiry made as to the persons entitled to be registered'. Table 4 examines some of the ways in which local authorities have tried to improve responses to these inquiries among those who have not returned their Form A.

Table 4:Does your local authority do any of the following to increase the response rate of the canvass?
	% of auths. which	no of auths. (n)	% introducing feature in last 5 yrs. (auths. giving date only)	no of auths. (n)
Use personal canvassers	89.2	224	12.5	18
Mention getting credit	61.0	153	87.6	99
Pay canvassers by results	48.2	121	57.8	52
Telephone non-responders	25.9	65	34.3	12
Use calling cards	25.5	64	60.9	28
Use a prize draw	9.2	23	85.7	18
Prosecute non-responders	3.6	9	75.0	6

Nearly nine in ten authorities (89.2%) use personal canvassers at some stage in the compilation of the electoral register. Using canvassers can have similar advantages to hand-delivering Form A. For instance, it is argued that canvassers are invaluable for picking up new properties. Experienced canvassers are also more likely to complete the forms correctly compared to householders and they can also answer queries which saves election office time. Authorities which make best use of their canvasser resources include those which carry out extensive briefing and training, those which ensure that canvassers are assigned to a single, manageable area, and those which effectively monitor the canvassers' performance.

Several respondents mentioned that the greater use of the electoral register as a check by credit agencies encourages a better response and increased claims. Nearly two-thirds of authorities now use the availability of credit as a carrot/stick for reluctant registrees, with the large majority of them having introduced this method in the last five years. One officer commented,

'People certainly take notice of a letter that says they may find it harder to get a mobile phone or borrow money, more so than a letter that merely says they might lose their right to vote'
(South East district)

Paying canvassers by results was used by nearly half of our responding authorities (48.2%). It is a relatively new method of increasing the accuracy of the register. Table 4 shows that more than half of authorities (57.8%) which gave a date for its introduction have changed their way of paying canvassers in the last five years. One authority was keen to point out that not only are canvassers paid a bonus for good results but that payment is also deducted for poor results.

The idea of using a prize draw is to attract people to return their form and to do so at the earliest opportunity. 9.2% of authorities use a prize draw in an attempt to increase the response rate to the canvass. In some instances it has been introduced for only one year because of the costs involved or the limited perceived success. On the other hand, other authorities intimated that a prize draw will be used for the next canvass.

Just nine authorities replied that they prosecute non-responders. One South East district had prosecuted two households a few years ago and these properties now return the Form A before the reminder stage. Another South East district also prosecutes non-responders in the Magistrates Court. They press for their costs against the offenders, keep records of recurring offenders and pass the names of repeat offenders to the Magistrates for higher penalties. A number of officers mentioned that people are threatened with prosecution but not actually taken to court. The main reasons for this are two-fold. First, the time and cost of preparing for court is prohibitive and second, respondents argued that the courts do not treat it as a serious offence and so impose minimal fines.

Officers were asked to specify what other methods they used to increase the response to the canvass. Although occasional initiatives such as using help desks in supermarkets and railway stations were claimed to have the effect of increasing the registration rate, the majority of responses can be grouped into the category of 'using special letters'. The 'threatening letter' was a popular choice for authorities. The letter usually outlines the legal penalties and the implications for obtaining credit. One authority made the important point that their letter is not threatening but 'polite'. It is intended as a carrot rather than a stick. Another council sends the final Form A and letter by recorded delivery. Respondents reported that the response rate to the final letter varied between 20-36%.

Box 2:Evidence of the success of methods to increase the final rate of return of Form As

Use of personal canvassers 'If a canvasser is not available, an area is posted and the response rate is not as good' (North West unitary authority) 'Personal visits achieve about an additional 8% response. I'm sure this response would be half if another form were sent only by post' (South East district) Using the availability of credit 'In the first year, 2000 additional householders responded to the first reminder which included a note on credit' (South East district) Performance pay for canvassers 'Going to performance related pay and smaller canvass areas has increased the borough response at end of first canvass from 73.5% in 1994 to 81.5% in 1995 and is now 85%' (London borough) 'Since we have paid canvassers by results, they have managed to reduce the numbers on their outstanding lists by half' (South East district) Introducing Prize draws 'The Prize Draw introduction in August 1998 resulted in an additional 13% of forms being returned prior to the production of the first reminder. This resulted in postage savings amounting (with discount) to approximately £5,000 - from these savings prizes totalling £2,400 were paid' (North metropolitan borough) 'The Prize Draw which was done before the first reminder was sent out had the returns up by 4% from the same time the previous year' (North West district) Final 'warning' letters 'At the end of the canvass period all non-responding households are sent a letter informing them that their names do not appear on the draft register. We usually get approximately 400-600 more voters on the draft register' (South East district) 'A letter is sent to non-responding households informing them of their liability to a fine, the possible inability to get credit and the intention of the ERO to remove names of certain electors if no response is received. This was introduced in Autumn 1998 and resulted in an additional 8,000 responses in January 1999' (North metropolitan borough)

Many local authorities claim that they can attribute an improving response rate to the canvass to the introduction of new measures. As one district council in East Anglia put it,

'Response statistics suggest that each initiative has marginally affected the overall district response'

Targeted registration campaigns
Some groups of people can be particularly difficult to register. Local authorities were asked which groups were problematic for their council and what methods they used, if any, to increase the registration rate amongst these people - see Table 5.

Table 5: Some groups can be particularly difficult to register. Does your local authority have any special problems registering any of the following groups of people?
	%	n
Houses of Multiple Occupation (HMOs)	31.9	80
Students/Attainers	27.1	68
Armed Services	13.1	33
Homeless people	10.8	27
Travellers	9.2	23
Foreign nationals	7.6	19
Non-native English speakers	7.6	19

Houses of Multiple Occupation
More than three in ten authorities (31.9%) declared that they had special problems with registering people who lived in houses of multiple occupation (HMOs). Authorities gave specific examples of types of HMOs: these included high rise flats and properties with intercoms, occupants of accommodation at a large hospital site, properties occupied by large numbers of asylum seekers, residential homes for the elderly, and sheltered housing accommodation (e.g. for persons at risk).

What do authorities do to improve the registration of people living in HMOs? Respondents mentioned that they contact landlords asking for information on tenancy. This can have some success but often landlords felt that it was not their responsibility to provide such information or they did not have the time to deal with the forms from the council. Examples of good practice included:

'Canvassers with areas containing a higher than average number of HMOs receive a higher rate of pay'
(South East district)

'Obtain pass-key from landlords to each block - marginal increase in returned forms'
(London borough)

'Forms specifically addressed after liaising with the Housing department'
(South West unitary authority)

'We have checked the identities of HMOs with Environmental Health Department and followed up with personal visits where there are no, or apparently inadequate, returns'
(South East unitary authority)

'Separate letter to HMOs with more than eight residents sent in early October to encourage accurate return based on qualifying date e.g. to elderly persons' home'
(South West unitary authority)

'Visits by full-time canvassers to homeless hostels'
(London borough)

Students and attainers
Students and attainers were cited as providing a particular registration problem by just over a quarter of authorities (27.1%). There are a wide range of initiatives used to get students in particular to register. These include sending Form A to students in the first week of October when they arrive at university/college and then reminders at the end of October (North West metropolitan borough); working with student unions to increase awareness (North unitary authority); sending notices/posters for display at Halls of Residences; giving presentations at 'Freshers' Fairs'; sending copies of the draft and final registers to universities (West Midlands metropolitan borough); and getting canvassers to make personal visits to student accommodation (South West district).

Another successful initiative involves liaising with the accommodation officer at the college/university. In some local authorities, the colleges will supply lists of students residing on campus (e.g. in one West Midlands metropolitan borough the local university sends student details to the council on disc), although in other councils officers have to chase university officials to receive this information. One authority mentioned that,

'The universities will not provide the ERO with names or even contact addresses of students in units of accommodation not managed by the universities'
(North metropolitan borough)

The fact that students are eligible to register at both their home and term-time addresses provides a further complication in encouraging them to register.

Similar types of initiatives are carried out in schools to increase the accuracy of the register - especially the use of posters, forms, 18th birthday cards etc. A specific student registration guide is produced in one council (North West unitary authority), another council encourages work experience students into the Elections Unit (East Midlands district), and another mentioned their use of focus groups and workshops involving schoolchildren (London borough).

Some councils described their initiatives as being reasonably successful but others complained that there is often limited or no interest. Examples of such comments included, 'No real evidence at the moment that this has helped' (East Midlands district), 'Total lack of interest' (West Midlands unitary authority), and 'Only two students registered at our roadshow event at (Further Education) College and it was later discovered that they were already on the register' (South West district). However, on the positive side, a council wrote that by making information regarding registration of 16/17 year olds clearer on Form A, the number of attainers registered this year increased by 20% as a result (North West metropolitan borough).

Armed Services
Table 5 shows that the Armed Services are another group perceived by a minority of councils (13.1%) to provide particular registration problems.

Box 3:The problem with Armed Services registration

'We find it very difficult to explain service registration procedures. This means that we have a large number of service persons who live in the area but remain registered elsewhere (and who possibly do not or are unable to vote in that area). There are literally hundreds of service persons registered in our area who have long since ceased to live in the area. Many of these have proxies in far distant parts of the U.K. who, in turn no longer live at the address we have. This causes enormous problems, not to mention a lot of abortive work and expense on our part - in the two polling districts with the highest number of service voters we received only 25% of postal votes back at the May's elections. Whilst we try to regularise things it is not easy to track people down and the service registration staff seem unable to help and do not have the sort of IT which could make it a lot easier. Something needs to be done because not only does it involve all the expense and effort to which I have referred, but it also effectively disenfranchises a large number of service personnel. Another problem is that many service spouses seem to forget if they have ever made a service declaration and are probably registered with ourselves as a civilian and elsewhere in the country as a service spouse' (South East district)

One key to improving the situation seems to be good dialogue between the council and the armed forces. One authority wrote that because of the closer working relationship built up with regiments in the district, large numbers of 'old' service voters have been removed (South East district). Another authority wrote that,

'In the aftermath of the 1992 general election when there were a number of visits to the office by unregistered soldiers from the Garrison, we have established contact with the army authorities to stress soldiers' responsibility to register by declaration. The army property services office also provides details on which property on the army estates were occupied on the 10th October each year so that we can concentrate on those that were occupied'
(South East district)

Homeless people
In the 1996 Rowntree survey, 15% of responding authorities answered that they had a special problem registering homeless people. This figure reduced to 10.8% in this survey. There is more evidence now that authorities are devoting additional time and resources to improve the situation. One of the easiest ways to improve the registration of the homeless is to send information letters to groups representing homeless people or perhaps better still to contact homeless people directly at hostels, 'drop-in' centres and soup kitchens. In one North West district, visits are made to the Homeless Action Centres, which has resulted in 23 homeless people registering. A North West metropolitan borough works with the 'Big Issue' magazine and have 15 homeless people on the register in 'unorthodox' addresses. In a central London borough, a personal canvass of the street homeless is undertaken on the 10th October every year (20 street homeless registered last year). A South East district has conducted a campaign with local homeless organisations to increase the registration of homeless people resulting in 'quite high' rates of registration at the two big hostels in the area. Finally, a South West unitary authority enables the homeless to register at the 'Big Issue' office and HUB (Drop-in) Centre with moderate success and a North West metropolitan borough has used a night shelter as a polling station.

Travellers
Nearly one in ten authorities (9.2%) cite problems in getting travellers onto the electoral register. Two councils who have tried to improve the situation have not met success. They wrote that,

'We have sought the help of the Council's 'Site Warden' to increase registration but at the end of the day, this group does not seek registration'
(East Anglia district)

'We sent personal canvassers to caravan sites - this has been totally unsuccessful - none registered'
(North West district)

Foreign Nationals and Non-Native Speakers
Nineteen authorities (7.6%) separately suggested that they had special problems registering one of two other groups of people, foreign nationals and non-native English speakers. In some authorities this is a very serious problem. A respondent from a London borough wrote that,

'European citizens form 12% of the borough's electorate'

A wide range of methods are used to target these groups of people. Three London boroughs provide us with examples. In one borough canvassers carry translations of information leaflets in 11 different languages; another borough offers doorstep translation for European languages; and a third uses dedicated phone lines from time to time for people who speak foreign languages. Outside of London, officers mentioned that they have targeted mosques and temples with translated leaflets and posters over several years (West Midlands metropolitan borough); issued leaflets in the main ethnic minority languages to accompany Form A; used the local Race Equality Forum to promote a registration campaign amongst ethnic minorities (South West district), and used canvassers from the Asian community who speak Urdu.

General registration campaigns
Local authorities were asked about their methods of electoral registration and if they carried out any registration campaigns to increase registration - see Table 6.

Table 6:Does your local authority do any of the following to increase registration?
	%	n
Work with other departments	56.2	141
Work in a consortia	26.7	67
Have an education programme	25.5	64
Have a 'late claims' campaign	24.3	61
Produce leaflets in foreign languages	17.9	45
Work with community groups	15.9	40
Hold a registration roadshow	11.2	28

The most popular method used by election officers to improve the accuracy of the register involved working with other departments in their local authority. The departments used by the responding local authorities included: Council Tax (92 local authorities), Housing (53 local authorities) and Planning (15 local authorities). A number of authorities used more than one department in order to increase registration. Other departments used included: Community Regeneration, Social Services, Education, Borough Treasurer, Technical Services and Environmental Services (New Properties).

The Council Tax department was used by the greatest number of respondents. It can help by informing the elections office of new properties that have been built so that a Form A can be sent at an appropriate time. It can check and confirm uncertain information in order to save time and it can help resolve name queries so that personally addressed letters can be sent to canvass refusals. One London borough wrote that this latter method was '50% effective'. However, although more than half of our respondents confer with other departments, in other cases

'No other departments are used. This is not possible because of the Data Protection Act'
(East Anglia district)

More than a quarter of responding authorities (26.7%) answered that they work in a consortium with other authorities to try to increase registration. This normally consists of joint publicity such as press releases, posters (e.g. displayed on buses), book-marks and the like. A popular but more expensive form of publicity uses advertisements on the radio. Two respondents who have used the radio wrote that,

'Colleague local authorities all confirm that the local radio consortia campaign is successful'
(South East unitary authority)

'Many electors comment that they are responding to the radio publicity campaign conducted jointly with other local authorities in the radio catchment area'
(South East district)

A similar number of authorities (24.3%) have a 'Late Claims' campaign. The success of such a campaign can be clearly measured in terms of actual numbers on the register. Evidence cited included,

'We often add up to 1,000 names on a monthly basis as late claims'
(London borough)

'Leaflet distributed in March, 7,000 forms returned, 8% of forms valid'
(West Midlands metropolitan borough)

There are a large number of general registration campaigns which are not included in Table 6 above. These vary from simply advertising in local newspapers and using posters in locations such as Doctors' surgeries to some original schemes. One metropolitan borough uses a Tape for blind persons (Talking Newspaper) and video tapes for the deaf using sign language; another authority takes the draft register to a local supermarket for three days for electors to check whether they are included (South East district); while a London borough wrote that inclusion on the electoral register is a qualification for parking permits.

There is real difficulty in determining the impact of such schemes. This has led some local authorities to abandon their initiatives as there is no hard evidence of their success. Other authorities have taken a different view on the lack of evidence. Respondents wrote that,

'We have no evidence but are concerned that registration figures would decline if campaigns were not sustained'
(South East district)

'We like to think that the sum total of the registration activity helps to maintain our household response rates at an acceptable level. However, we cannot prove objectively that any particular measure is effective'
(West Midlands metropolitan borough)

Nevertheless, a handful of authorities took a tough line on the relative costs and benefits of campaigns,

'Although the District Council has not made special efforts to increase registration rates, the registration rate is high compared to other areas. The cost of mounting special campaigns for what would be a minimal increase in return rates is thought unwarranted'
(South East district)

Carrying Names Forward on the Register

We asked local authorities about their policy for carrying forward names on the electoral register from year to year - see Table 7.

Table 7:In the absence of a returned Form A, what is your local authority's policy on carrying names forward in the electoral register?
	%	n
Not carried	20.0	49
One year	36.7	90
Two years	24.1	59
Three years	3.3	8
Five years	0.4	1
Indefinitely	15.5	38

Table 7 shows that 36.7% of local authorities carry names forward for one year - a number of respondents commenting that this is in line with the current Home Office guidelines. The decision on how long to carry names forward in the register belongs to the officer with responsibility for the electoral register, usually the Returning Officer or the Electoral Registration Officer. To illustrate this, an officer from a Northern metropolitan borough wrote that,

'The previous electoral services manager carried forward names indefinitely. We now have a new computer system and I do not intend to carry names forward for more than two years'

In one London borough, whose general policy is not to carry names forward in the register, names were carried forward for one year in one of their three parliamentary constituencies. Households in that constituency had been sent only a reminder form rather than being subject to a door to door canvass.

More than 4 in 10 local authorities (43.3%) carry names forward for more than one year, with 15.5% of the total responding authorities carrying names forward indefinitely. Those authorities that gave this answer often included a rider to the effect that, as their rate of return of Form A was high, this was not really an issue.

Some authorities mentioned that carrying names forward is dependent upon whether there is information that a household is resident at a particular address. For example, in one South East district names are carried over for one year if Council Tax/canvasser records confirm the person is resident. Otherwise names are removed.

Anonymous Registration
Another area of difficulty for election officers in producing the final register is anonymous registration. This is a useful mechanism for individuals fleeing violence such as 'battered wives' and people whose job may lead to them being put into dangerous situations e.g. prison officers. We asked local authorities how they deal with the registration of people who wish to remain anonymous - see table 8.

Table 8:How does your local authority register those people who wish to remain anonymous?
	%	n
Use an anagram/pseudonym of their name	44.6	112
No policy on anonymous registration	40.2	101
Use the 'Other Electors' part of the register	18.7	47
Publish the register in name order	0.8	2

Just as there are no rules on how long names should be carried forward in the electoral register, there is also no law on anonymous registration. This results in confusion over what is legally acceptable. For instance, in one South West unitary authority an officer commented that the three options we gave for registering electors anonymously were all illegal - although nearly two thirds of authorities do implement one of these options either formally or informally.

A large number of local authorities answered that they had no policy on anonymous registration, but also ticked one of the three options of registering electors anonymously (this was then coded as having the particular policy they ticked). In effect, the majority of councils do not have a formal policy on anonymous registration. Electors were often told unofficially ('off the record') of the various options open to them and it was left to them which they chose if they wanted to register at all. The options offered to potential voters were, 'To change name by deed poll' (East Midlands district), 'Use their maiden name/mothers maiden name' (South East district), and 'Register with a relative/friend' (East Midlands district). In one South East unitary authority the advice given was to register with a 'favourite relative' in another authority, and so lose their entitlement to vote where they actually lived.

Election officers were on the whole sympathetic to the cases of anonymous registration but felt as if their hands were tied on the issue. Officers are put in a position where they can give a person some options, if they exist in their authority, and then leave it up to the individual whether they wanted to register or not. The interesting point here is that people are not forced to register, even though it is against the law not to do so and the authority knows exactly who these people are! One authority wrote that they,

'Offer to leave names off the register after warning about lost entitlements to vote and possible problems getting credit. A recognised procedure adopted nationally would be helpful'
(North West metropolitan borough)

Other respondents echoed such sentiments,

'The Council has adopted a flexible approach to anonymous registration but it is hoped that legislation will clarify this position'
(North unitary authority)

'Obviously, the law needs to be changed. This authority has grave reservations about breaking the law'
(South East district)

Local Authorities and Electoral Registration
Local authorities were asked if the issue of electoral registration had been discussed by their council or its committees in the last five years. 50.2% of authorities had discussed the issue during that time. It seems that electoral registration has become a more important issue recently, as only a third of authorities replied that it had been discussed in their council or committees when the Rowntree research was conducted in 1996. The majority of the discussion that takes place occurs annually when a report written by officers is presented to members at committee. Two examples of this practice are,

'Annual report to General Purposes Committee with details of the size of register, the response rates, analysis of initiatives taken and proposals for future initiatives'
(North West metropolitan borough)

'Annual report to the Policy and Resources Committee each summer giving performance achievement and inter-borough comparisons. Performance also published approximately annually in our own council newspaper'
(London borough)

Electoral registration is also discussed at council meetings when specific changes to policy and procedure are put forward. Such changes have included, 'a new electoral management system' (East Anglia district), 'the introduction of a three year cut-off for non-responders from the 1999 register' (North West metropolitan borough), 'a new canvass policy (personalisation of forms and payments by results scheme)' (South East district), 'an improved format to publish the register' (South East district) and 'a reduction of the role of canvassers to delivery and collection of Form As only' (East Midlands district).

Often the issue of electoral registration is only discussed when a request is made for extra resources to improve current practice. For example,

'Discussion only about the issue of increasing awareness in ethnic community areas. £5,000 was provided to initiate a campaign'
(South East unitary authority)

'All-party member and officer working party set up in 1996 to monitor how extra money on electoral registration was being used and how effective it was. Also looked at other areas which could be tried i.e. council records, using credit as a carrot/stick'
(North metropolitan borough)

On the other hand, electoral registration is also discussed when a local authority is looking to make cuts.

'Discussed at budget time when the option of removing the personal canvass to save money was considered but not passed'
(East Anglia district)

'Discussed at Policy and Resources Committee with regard to staff re-structuring of three sections. Recommendation was to cut two full-time staff and six temporary staff down to one and a half full-time staff'
(South West district)

Just under half of our responding authorities answered that there had been no formal discussion of electoral registration by the council or its committees in the last five years. In one authority, the officer placed no importance on the issue being discussed at member level.

'No specific recommendations have been sought from the council or its committees, the view being taken that it is up to the ERO to decide the policy, not the council'
(West Midlands district)

PREPARING FOR THE ELECTION

Respondents were asked about arrangements in the lead up to a local election, specifically their policies and practice on absent voting, the design and despatch of poll cards, and the siting of polling places.

Absent Voting

Postal and Proxy Vote Applications
Absent voting is a system whereby an elector votes without actually visiting the polling station. Instead, either they appoint a proxy to vote on their behalf or they vote by post. There are a number of circumstances in which it is permitted to use an absent vote. These include, being ill, having a physical disability, working away from home or being away on holiday when the election takes place.

Local authorities were asked, 'At your last May council elections, how many people were permitted to vote either by post or proxy', and 'How many of these electors actually voted?' The figures in Table 9 show the number of postal and proxy votes applied for in responding authorities and the numbers who actually voted by post. The figure of greatest significance is the % of those granted a postal vote who used it. A turnout of 64.1% is approximately twice that for voting in person at a local election.

Table 9:Postal and Proxy Votes
	Proxy applications	Postal applications	Postal votes
Total	93,048	327,382	177,561
Mean	487	1,544	1,032
as % of applications	-	-	64.1
n	191	212	172

Because elections for all the London and Metropolitan boroughs take place at the same time, it is possible to examine the use of the absent vote in more detail and make comparisons between boroughs. Table 10 shows that there were over 60,000 applications for an absent vote at the 1998 London borough elections in the 24 councils for which we have information. Turnout amongst those who applied for an absent vote in London was 54.9%.

Table 10:Postal and Proxy Votes at the 1998 London borough elections
	Proxy applications	Postal applications	Postal votes
Total	11,955	51,319	28,192
Mean	498	2,138	1,175
as % of applications	-	-	54.9
n	24	24	24

The analysis was repeated for the metropolitan boroughs. The results in Table 11 show that there were over 125,000 applications for an absent vote in 28 metropolitan boroughs. The turnout rate of those who applied for a postal vote was 50.5% compared with an overall turnout in the metropolitan boroughs in 1998 of 24.8%.

Table 11:Postal and Proxy Votes at the 1998 Metropolitan borough elections
	Proxy applications	Postal applications	Postal votes
Total	19,906	106,109	53,533
Mean	710	3,790	1,912
as % of applications	-	-	50.5
n	28	28	28

Table 12 shows that the average percentage of people who apply for an absent vote (postal and proxy applications combined) in London is 1.8%. This figure varies from a low of 1% of the electorate to a high of 4.3%. Five London boroughs each have more than 2% of their electorate making applications for an absent vote. The results for the metropolitan boroughs are similar to London. On average, 1.7% of the electorate apply for an absent vote. The variation between the boroughs with the highest and lowest percentage of their electorate applying for an absent vote is smaller than in London - from 0.6% to 2.9%. There are 7 metropolitan boroughs where more than 2% of the electorate applied for an absent vote at the last election.

Table 12:The proportion of the electorate applying for an absent vote in London and the metropolitan boroughs, 1998
	Mean (%)	Min. (%)	Max. (%)
London	1.79	1.0	4.3
Metropolitan	1.65	0.6	2.9

Advertising the Absent Vote Facility
What do authorities do to advertise the facility of absent voting?

Table 13:Does your local authority include information about absent voting
	%	n
On Form A	86.1	216
In local newspapers	71.7	180
In the council newspaper	60.6	152
On posters	54.2	136
On a specially produced leaflet	26.7	67
On a leaflet with Form A	15.5	39

As Table 13 shows, the most popular way of providing information on absent voting is by using Form A. Some 86.1% of responding authorities included details about absent voting on their electoral registration form. Some authorities do not use the Form A to advertise absent voting because they believe that it is more effective to use a leaflet delivered with the form, but others are able to produce evidence of its impact.

Councils were also able to quantify the effect of producing a special leaflet on absent voting. Such a leaflet can be enclosed with Council Tax bills, Form A or with poll cards sent to every household, or can be directed to a certain group of people e.g. holidaymakers.

Box 4:Evidence of the impact of advertising the absent vote facility

Using Form A 'Since putting information on the Form A, the number of absent vote applications has gone up by 20%' (East Anglia district) 'There was an increase of about 300 people on the permanent AV lists after AV requests were put on the Form A' (East Anglia district) 'Introduced details on absent voting on Form A for the last canvass. We received over 700 enquiries which produced 200 new applications for the 1999 election' (North West district) 'The inclusion of a request for an Absent Vote form on the Form A resulted in approximately 200 applications' (West Midlands district) Producing special leaflets 'A leaflet is produced for Travel Agents (alongside other methods of advertising). There has been an increase in postal vote applications - doubled in five years' (North unitary authority) 'The leaflet distributed to every property increased the number of absent vote applications by 100%' (West Midlands district) '20% increase in Absent Votes in 1999 following delivery of a special leaflet with poll cards' (South East district) 'Leaflet delivered in March - approximately 2,600 received, at least 50% accepted' (West Midlands metropolitan borough) 'Leaflet sent to every house in the borough. 19% increase in Absent Vote applications between 1995-99' (East Midlands district) 'The leaflet sent out with Form A resulted in approximately 300 requests for additional forms' (North metropolitan borough) 'The leaflet sent out with Form A each year produces over 500 responses' (South East district) 'Leaflet with poll card sent earlier than deadline. At the European elections the poll cards were unable to be issued until after the AV deadline. Applications reduced by 13%. Numerous complaintes were received from disgruntled and disenfranchised electors' (North metropolitan borough)

Other forms of advertising seem able to raise awareness of the absent vote facility, but their effectiveness is more difficult to measure. In one North West metropolitan borough they use their web-site to advertise, while another borough issues advice in a Home Owners' manual. Other forms of advertising include placing posters on the outside of five local buses (North West district), using the radio (North West district), parish magazines (South West district) and the Council Tax calendar (North district). As one respondent put it,

'We have no real evidence other than passing comments such as "I saw your poster at the Guildhall" or "I read it in the paper"'
(South West district)

Absent Voting in Institutions
We were also interested in whether local authorities make special arrangements for people living in certain institutions to have an absent vote.

Table 14:Does your local authority make special arrangements for people living in the following institutions to have an absent vote?
	%	n
Residential/Nursing homes	74.1	186
Hospitals	21.5	54
Hostels	13.9	35
Universities	13.5	34

Table 14 shows that nearly three quarters of respondents (74.1%) make special arrangements for people living in Residential/Nursing Homes. This normally takes the form of sending letters to these homes. For example,

'Letters are sent to nursing homes prior to the election outlining the legal requirements re postal voting but also providing guidance on the necessity to complete declarations of identity etc. Most residents in OAP homes take up the option of postal and proxy voting. Fewer rejected postal votes at this year's elections compared to four years ago'
(East Midlands district)

Some authorities contact the wardens or residents in the homes personally. One such authority,

'Employ a canvasser to visit homes and explain the absent vote procedure and deal with applications'
(North metropolitan borough)

Other examples of good practice include,

'A letter is sent to all hospitals which includes information on who is entitled to an absent vote and the different categories of absent vote'
(East Midlands district)

'Contact is made with hostels used by the Homeless Families Unit'
(London borough)

We have shown that the initiatives carried out in some authorities have been successful in increasing the use of absent voting. We are, however, always faced by the problem of not knowing how many applicants would have applied for an absent vote anyway without any schemes. While we have focused upon the success of some schemes, other councils have not achieved the same results and note,

'Lack of success on last year's postal vote campaign in that only 6 persons responded'
(North West metropolitan borough)

'Overall application numbers are falling from 2,534 in 1998 to 2,394 in 1999 despite greater publicity'
(North metropolitan borough)

Poll cards

Method of Delivery
How are poll cards delivered in a local authority - are they hand-delivered or posted? Table 15 implies that some local authorities do not use just one method of delivery but use a combination of methods for different areas of their authority.

Table 15:Method of delivering poll cards
	%	n
Hand-delivered	66.9	168
Sent by post	43.3	109

Thirty-three local authorities have changed their method of poll card delivery in the last five years. Seventy per cent of these are now hand-delivering rather than posting their poll cards. The most popular reasons given for this change were that it is cheaper and more efficient. On average, our responding authorities suggested that they save about 7p a card by using hand-delivery rather than posting their cards. The delivery of poll cards was sometimes used as a carrot by the authority for canvassers in that only people who carry out a Form A canvass are allowed to deliver poll cards. The hand-delivery of poll cards was argued to be more efficient than the post because canvassers must deliver by a certain date, therefore, officers know that they have been delivered. Complaints of non-delivery of poll cards can also be taken up with the election office's own staff.

A large number of negative comments were received about the service from the Royal Mail. Examples included,

'Postal strike meant that we needed to organise a hand delivery. It was much cheaper and more efficient'
(North East metropolitan borough)

'Post Office not delivering properly (missing people/households or delivering some many days after others)'
(London borough)

Other authorities, on the other hand, maintained that the post was cheaper, more efficient and required less administration by the elections office.

Timing of Delivery
The timing of the delivery of the poll card is at the discretion of the Returning Officer. The Representation of the People Act (1983) states that, 'The Returning Officer shall as soon as practicable send to electors and their proxies an official poll card'. This inevitably results in a wide range in the timing of delivery. The number of days before the election that an elector receives their poll card varies between 4 days in one North East district to 65 days in one North West metropolitan borough. The average number of days given by our responding authorities was 17 days - an increase of 3 days on the figure in the 1996 Rowntree survey. This barely provides an elector the chance to apply for an absent vote as the deadline for application (other than 'late' ones for illness) is the thirteenth working day before polling day.

In the Rowntree research we found that 6% of respondents sent their poll cards out more than three weeks before the election. We advised authorities in the conclusions of that report to 'despatch poll cards earlier' (1996:40). The results from this survey show that over a quarter of authorities now send their poll cards out more than three weeks before the election. In one metropolitan borough, they continue to send them out very early,

'60 days before the election - this allows sufficient time to make a claim against the register'

There is most definitely a movement towards sending poll cards out at an earlier date. In one West Midlands unitary authority, for example, electors would usually have received the poll card 5 days before, but they are now issued by the 11th of April. Similarly in one London borough, the poll card used to be sent out 7 days before an election. They now send the poll card out 24 days before the election, giving electors sufficient time to make an application. Finally, in one South West district, the cards are normally distributed seven days before the election, but for the European Parliament elections in June 1999 they were sent one month prior to the election. This was reported to 'work much better'. There can, however, be problems with sending out the poll cards at an earlier time. One officer wrote that,

'For this May's elections we sent the poll cards out before the deadlines and are now faced with having to send a 'cancellation' letter to the electors in one ward which is uncontested'
(South East district)

Other, mainly rural authorities also cite the uncertainty of whether there will be contests in every ward as the reason for their not despatching poll cards earlier.

Following on from the timing of the delivery of poll cards, authorities were asked if, upon receiving it 'there is sufficient time for an elector to apply for an absent vote and for your local authority to process that request?' 127 local authorities (51.4%) answered 'yes' and 120 local authorities (48.6%) answered 'no'. A handful of authorities replied both 'yes' and 'no', suggesting that their poll cards are delivered over a number of days around the time of the deadline for absent vote applications.

For authorities that have changed their timing of the poll card delivery to enable people to apply for an absent vote, there are some success stories,

'For the first time this year we decided to send out poll cards before the closing date for postal/proxy applications and we received a good response'
(North West district)

'This year we are delivering early (12-16 April) - reaction so far is good, it has encouraged an increase in postal votes'
(North unitary authority)

Information on Poll Cards
We asked local authorities what information they included on poll cards?

Table 16:Does your poll card include any of the following information?
	%	n
The telephone number of the Elections Office	33.2	83
The closing date for absent vote applications	24.0	60
Information about disabled access to the polling station	3.6	9

Table 16 shows that around a third of authorities (33.2%) include the telephone number of the elections office, and nearly a quarter (24%) include the closing date for absent vote applications. Adding extra pieces of information on the poll card can have an impact on electors.

'Since the closing date for absent voting applications has appeared on the poll card a greater number of applications have been received'
(North metropolitan borough)

'Included a message about AV and telephone numbers of the office. We have been inundated with telephone calls for application forms re postal/proxy voting claims'
(West Midlands metropolitan borough)

Respondents suggested they included other items of information on the poll card, apart from those mentioned in Table 15. Some poll cards have the address of the elections office, an
e-mail address for the department, relevant notes about car parking, change of polling place etc., details of the results hotline, and a slogan saying 'You don't need the card to vote'.

Separate leaflets are also used by some councils which negate the need to put information on the poll card. One West Midlands district sends an information letter to all properties at the beginning of March providing detailed information re elections. This enabled them to deal with queries and Absent Vote requests well in advance of the election. A South East unitary authority includes the same information on the council tax leaflet, which is delivered at the end of March. Only one council, a South West district, mentioned that their leaflet includes details of candidates. Finally, in one London borough, the information covered in Table 16 is included on the polling station maps delivered separately with poll cards to each household.

The responses in Table 16 imply that a large majority of authorities do not include any additional information on their poll cards. The reasons proffered for this are not financial restrictions, but that they believed it was not permissable to alter the appearance of the poll card. Three comments make this point below.

'No - none of the above are permitted by the legislation'
(East Anglia district)

'None - poll cards have to follow a prescribed wording and layout'
(East Midlands district)

'No - the view is taken that the poll card format is prescribed by law with the slight exception in that a map was included when a polling venue had been changed'
(West Midlands district)

The Representation of the People Act (1983) states in relation to the poll card that authorities 'may use a form to like effect'. The changes made to the poll card by some authorities and the comments directly above show that this phrase has been interpreted in different ways.

Form of Poll Cards
We found in the Rowntree research the first sign of changes being made to poll cards. This time we again asked local authorities if they had implemented any of a number of possible changes - see Table 17.

Table 17:Have you made any of the following changes to the poll card?
	% of auths. which	no of auths.	% introducing feature in 1999	no of auths.
Pre-printed	88.9	224	60.7	105
Changed its text	23.1	58	58.3	28
Altered its size	20.3	51	63.0	29
Included a map	12.0	30	80.8	21
Changed its colour	7.2	18	68.8	11
Included a How to Vote leaflet with the poll card	3.6	9	80.0	4
Used a second reminder poll card	1.6	4	50.0	1
Referred to a landmark near the polling station	0.8	2	0.0	0

In total, about a quarter of councils report having made changes. When a change is made to the poll card, quite often a number are introdued at the same time. In one North metropolitan borough, for example, they changed the size, colour, text and included a map on their poll card - all for the first time in 1999.

Changing the text on the poll card is the most popular single change made by authorities (23.1%). The typeface can been increased to aid the partially sighted (London borough), the clarity of the text can been improved to make the poll card easier to read (North West district) and laser printers can be used to provide a better quality, bolder text. We do not know the effect of these changes, but officers suggested that the cards look better than in the past.

20.3% of responding councils have changed the size of the poll cards. We did not ask about the relative size of the cards but we assume the cards were enlarged to 'A5' size. Two London boroughs commented on the effect of this change. One wrote that the change has resulted in a smaller number of phone calls about non-receipt of poll cards, while another reported that their customer survey in 1998 indicated approval of the doubling of poll card size to A5. Only one council, a North West district, reported negative comments on the larger poll cards.

Of the thirty authorities that have maps on their poll cards, some have only used them when the location of a polling station is changed. Once again, comments from two further London boroughs demonstrate the advantages of including maps on cards.

'Fewer electors now go the wrong polling station or phone up to ask the location'

'The number of phone calls asking for locations for polling stations has drastically reduced since we included a map for the first time'

Changing the content or design of poll cards is unlikely in itself to lead to any significant improvement in turnout. In yet another London borough, the officer pointed out that three changes have been made to the poll card since 1994: the size, text and a map. The turnout in 1994 was 50%, while in 1998 it fell - along with that in all London boroughs - to 36%. Nevertheless, the majority of authorities which have introduced changes reported receiving positive feedback on them from those electors who did go along to the polling stations.

Polling Places/Stations

Location of polling places
We asked local authorities about the location of their polling places and the degree to which they relied on different types of venue - see Table 18.

Table 18:Location of polling places and frequency of use
	% of auths. that use	Mean %	Max. %
Schools or Colleges	99	29.5	90
Church halls	99	18.2	60
Leisure centres	40	1.7	46
Community/Youth centres	91	15.1	70
Village halls	77	26.0	99
Mobiles	62	3.9	44

Virtually all authorities use schools or colleges as places for people to cast their vote. On average, 29.5% of the polling places in an authority are schools and colleges, with one North West district using them in 9 out of 10 cases. Church halls are used equally frequently, but make up a lesser proportion of the overall polling place stock. Village halls are common polling places in rural areas, with one East Anglia district using them almost exclusively. The proportion of authorities using mobile polling stations has risen, from half in 1996, to more than 60% now. However, it is still the case that for most authorities they comprise only a small proportion - on average 3.9% - of total polling places.

We asked local authorities to name 'other' locations of polling places. The most popular 'other' location for a polling place was a library, which was mentioned by 36 authorities. Twelve local authorities reported using supermarkets as locations for polling stations. Some of the more unusual locations listed included public houses, private houses, a boathouse, hotels, temples, farm barns, a car showroom, and swimming baths.

There does not seem to be much evidence of experiments relating to polling place location. A number of authorities have said, however, that they are considering using supermarkets for future elections. The list of polling places above shows that local authorities are forced into using whatever buildings are available. There is simply no choice in some areas, authorities have to make the best of what is available, or if nothing is available, use a mobile unit.

'In the past we have used a portable cabin for a number of years - hiring and transport costs are very high'
(South East district)

'We have 14 locations where no location whatever exists and therefore have to use mobile units'
(South West unitary authority)

'The number of mobiles reflects nothing suitable in that area. We had to use 23 in May 1999'
(North metropolitan borough)

In one authority, the situation of finding suitable premises has proven to be an even harder task because the Council has a commitment that no elector should have to walk more then 800 metres to a polling station (South East unitary authority).

We asked local authorities whether they have problems finding suitable locations to use as polling stations. It was not surprising to find that nearly three quarters of respondents (73.3%) replied that they did have problems. The two most common problems revolved around the use of schools and the problem of there being no suitable buildings on new housing estates.

Schools seem to be becoming less co-operative in allowing their buildings to be used as polling places. One of the main reasons for this is the issue of security. Authorities mentioned that in view of the Dunblane tragedy they are concerned about having a polling station on their site. Schools are also reluctant to close for the day because of the legal requirement to provide a specified number of hours education. In one North metropolitan borough, schools are required to set an Inservice Training Day for polling day, but there has been adverse reaction as the day clashes with examinations and national tests. The problem with schools is summed up by a quote from one officer who wrote,

'We have grave difficulties with schools. The paradox is that although they are cheap to use, well known to the locality and usually accessible, schools (and often parents) are uneasy about having to close - most do close either because they have no spare capacity, or security concerns, or both'
(South East district)

Another significant problem for authorities is that often no communal facilities are built on new housing estates and/or that previously used village halls have closed. This puts additional pressure on them to find alternative locations, or use mobile polling stations. When a hall can be found, there are other problems for the authority in trying to book the location for a single day,

'Owners of private accommodation across the city prefer to let the premises to regular users'
(West Midlands metropolitan borough)

'Many halls are now used for private concerns (nurseries, crèches, training purposes) and there is greater reluctance to release them for polling day'
(West Midlands district)

'We have approached some halls who do not want to give permission to use them as they would need to cancel regular users/playgroups'
(South East district)

Facilities at Polling Places
Not only are there problems finding suitable locations for polling places, but the survey suggests there are considerable difficulties with those polling places currently in use - see Table 19.

Table 19:Type and frequency of problems with polling places
	% auths. that have this problem	Mean %	Max. % (Auth)
Not accessible to wheelchairs	78.1	14.1	90
High cost	64.5	7.0	60
No parking	58.1	7.7	80
Poor facilities	53.8	5.5	50

More than three quarters of authorities (78.1%) report having some polling places that are not accessible for wheelchairs. In some authorities, as many as 90% of polling places are claimed to be inaccessible, though the average is for about one in 7 to be inaccessible. The problem of access can arise as much in old buildings in remote rural areas as in urban areas. Significant numbers of councils also said that polling places were a 'high cost' item - often with specific reference to the hire of mobile units - and had poor or no car parking facilities. The general category of 'poor facilities' covered a range of complaints - often from the perspective of polling officers rather than voters - from inadequate heating to lack of toilet facilities.

Another aspect of the provision of facilities relates to what local authorities themselves provide at their polling stations. For example, we found in the Rowntree research that two thirds of responding authorities provided special facilities to facilitate the physically disabled voting at election time (either ramps or disabled polling booths). The results from this survey show that the use of ramps and booths has increased - see Table 20.

Table 20:Does your local authority provide any of the following at your polling stations?
	%	n
Ramps	74.8	187
Disabled polling booths	72.0	180
Facilities for the partially sighted	31.2	78
Translations of voting instructions	6.4	16

Table 20 shows that around three quarters of authorities ticked that they used ramps (74.8%) and disabled polling booths (72%) at some, but rarely all, of their polling places. Facilities for the partially sighted were less common, but included having white highlights around the ballot box slot (South West unitary authority), A3 enlargement of ballot paper/ Braille copies of ballot paper (London borough), large print notices (London borough), a thick 'border' on the ballot paper (South West unitary authority), use of handrails and non-slip carpet (West Midlands metropolitan borough) and having all steps marked with high visibility tape and asking Presiding Officers to ensure good lighting (North West district).

We asked local authorities if they knew of and/or had adopted any of the recommendations of The Scope Reports (Scope, 1992 and 1997). These cover a 'census' of the problem of disabled access at polling places at the 1992 and 1997 general elections. Nearly six in ten responding authorities (59.4%) answered that they had adopted some of the recommendations from the reports. This figure may be slightly misleading in that some authorities will have made relevant changes in advance of the report. It does show, however, that improvements have been made in this area. However, 32 local authorities (14%) were not aware of the report.

There are two particular ways in which authorities have reacted positively to the Scope reports. They have either carried out their own audit of polling stations using an access officer, or have used local disabled groups to act as consultants. The major barrier for an authority in improving access to polling stations is, however, financial. If there is no money available, changes cannot be made. In one authority, a budget was made available to the Council's access officer to help facilitate schools/halls etc. to become fully accessible. Only those few buildings where it is impossible to achieve (or uneconomic) remain inaccessible (South East district).

Reviewing Polling Places
How often do authorities review the location of polling places/stations so that deficiencies can be identified and changes made? - see Table 21.

Table 21:How often is the location of your polling places/stations reviewed?
	%	n
Ongoing process	79.1	197
Annually	9.2	23
Every 2-5 years	10.0	25
Every 6-10 years	0.8	2

Nearly 4 in 5 authorities claim to keep the location of polling places/stations under constant review. One officer reported that,

'We have a rolling programme of reviewing polling stations. We look at some each year. This year we reviewed the 20 with the lowest turnout'
(North East metropolitan authority)

In one West Midlands district, the situation is,

'ongoing - mainly as a result of comments, complaints from political parties, the public, Presiding Officers etc.'

Box 5:Improving Polling Place Access

Following local authority audits: 'We have carried out a survey of our polling stations and identified where improvements were needed. Some permanent ramps have already been installed as a result' (South West district) 'In 1998 we conducted a survey of polling stations on election day from a disabled access point of view. Buildings have been weighted and a rolling programme is under way governed by limited resources' (North West metropolitan borough) 'We were aware we were 49th out of 50 in the survey. We have made great strides in purchasing ramps for those stations without suitable access' (East Anglia district) 'Following the 1992 report extensive permanent access improvements were made to buildings and some stations relocated - we came top of the 1997 survey. We include disability issues on our polling station checklist' (East Anglia district) 'Disabled booths in every station. Choose buildings with good access where available. Signpost separate access if necessary. Instruct Presiding Officer to give all possible help - taking ballot box outside if necessary' (North metropolitan borough) Following consultation with local disabled groups: 'A member of Scope has toured polling stations on polling day with the Returning Officer. We have changed a number of polling stations this year' (South West district) 'We ask the local branch of Disabled Forum to notify us of any suitable premises available' (South East district)

The Rowntree research found that more than one in ten authorities had not reviewed the location of their polling stations for over ten years. The results in Table 21 indicate that no authority now admits to this practice, although more than 10% of authorities still do not review at least some of their polling places/stations annually.

When the location of polling places/stations is reviewed, who do authorities consult for their opinions on their re-positioning? - see Table 22.

Table 22:Who do you consult on the re-positioning of polling places/stations?
	%	n
Councillors	87.1	217
Parish Councils	43.4	108
Parties/Agents	24.5	61
The general public	14.9	37
No consultation	6.0	15

As might be expected nearly 90% of authorities consult their members while in addition, a quarter of authorities contact political parties or agents directly. Parish Councils are also regularly consulted in those areas where they exist. Fewer than 15% of authorities consult the public on the re-positioning of polling places/stations. When the public are consulted this is done through, for example, a press release in local papers (London borough), citizen panels (South East district) and even a letter to all electors (South East district). Other individuals or groups reported as being consulted include resident associations and other community and disability groups, as well as head teachers in schools who were asked for suggestions of alternative locations and the polling staff themselves.

ADVERTISING THE ELECTION

Local authorities were asked what they did to advertise that a local election was taking place - see Table 23.

Table 23:In addition to delivering the poll card, how does your local authority advertise that a local election is imminent?
	% of auths. which	no of auths.	% introducing feature in last 5 yrs. (auths. giving date only)	no of auths.
Local newspapers	80.4	201	35.9	33
Council newspaper	63.7	160	64.3	54
Your own posters/advertising	61.4	154	65.5	55
Radio	36.7	92	83.7	41
Home Office posters/advertising	29.5	74	24.0	6
Leaflets	23.1	58	90.3	28
A banner at the Town Hall	6.0	15	100	9

A handful of authorities admitted that they did nothing except for the Notice of Poll. The most popular method of advertising a local election was by using a local newspaper. This is one of the cheapest and easiest means of advertising and used by 80.4% of responding councils. Only 33 authorities have introduced this method of advertising in the last five years which indicates that the use of newspapers is not a new phenomenon. A few respondents suggested that local newspapers are utilised to their best effect in their authority. Not only are advertisements placed in newspapers, but press releases on election matters are also issued. These news stories can then be picked up by newspapers to run at no cost to the local authority. One authority highlighted their good practice relationship with the media. They advertise the election,

'By encouraging media interest through issuing a full media pack. By faxing out the details of the candidates with a press release when nominations close. By having information put on a local cable company's Community Channel on television'
(Cambridge)

In the Rowntree research (1996) we asked what kind of media authorities used to advertise the election. One of the options was to use the radio or television to advertise. There were very few authorities which used these methods three years ago, but they are now much more popular. Ninety-two authorities (36.7%) advertise the election on the radio, with many of them having begun this during the past 5 years. As such advertising is expensive, campaigns such as these are nearly always carried out with other authorities in a consortium.

There was evidence from our responding authorities that leaflets and letters about the election are being delivered to every household much more so today than was the case in 1996. The main advantage of this method is that you are virtually guaranteed that the advertising will reach the majority of electors. Examples of authorities that use such a method include,

'We deliver a 'Use Your Vote' leaflet to each household with their poll cards'
(North East metropolitan borough)

'Second reminder poll cards in the wards with the worst records of turnout'
(South West unitary authority)

Other ideas for advertising a forthcoming election included advertising on the rear of buses introduced this year (North metropolitan borough), the council's web-site, putting an advert on beer mats for local pubs and clubs, to attract the youngsters (South East district), a signed video to Deaf Centres and Talking Newspapers (North West metropolitan borough), franking machine logos (South East district), roadside boards, bus adverts, and tags on dustbins (North West district).

Box 6:Evidence of the impact of advertising an election

Negative: '1996 - posters and a radio campaign; 1997 - local newspapers, council newspapers and leaflets. Evidence shows a reducing turnout since these campaigns began' (South West district) 'Turnout was 4% lower this year than last elections in 1995, therefore, it could be argued our advertising campaigns were not successful' (North West district) 'Advertising on buses, cinema advertising, library bookmarks, doctors' surgery bookmarks. The turnout has been the lowest for 40 years despite the increased advertising. It seems factors outside of our control are mainly influential' (East Midlands district) 'The turnout has dropped since the leaflet was used but it could have dropped even more if we had not used it' (South East district) Positive: 'Turnout increased where we had leafleted' (London borough) 'Improved turnout in comparison with other London boroughs in 1998' (London borough) 'The increase in telephone enquiries in the lead up to the elections' (North West district) 'In February 1999, every household was sent a letter confirming their entry in the register and informing them of forthcoming elections. As a result, over 100 claims for inclusion were received and a record number of postal vote applications' (South East district) 'We sent a letter between 15th February and the 1st March to every household showing who was registered, what to do if not, how to get a postal vote, permanent or for May 6th/June 10th. Over 800 calls received resulting in 52 additions to the register. 250 postal votes were added too' (East Midlands district)

One North metropolitan borough in particular reported trying a number of different initiatives. The respondent wrote that,

'For the May 1999 local elections, the Council has appointed an agency to run an advertising campaign in an effort to increase voter turnout. The campaign will include 'supersize' displays on buses, posters on Adshels and public buildings, radio advertisements, beer mats and pub washroom posters, advertising in City Centre pubs and nightclubs and delivery of a multi-lingual leaflet to ethnic minority households - all at a total cost of £42,000'

The turnout in that borough in 1999 was 26.9%. This represented an increase of 2.8% compared with 1998. Across the metropolitan authorities as a whole, the turnout was 1.3% higher in 1999 than in 1998.

Generally, the evidence of the success of advertising the election is hard to come by. Is success to be measured by an increased level of turnout at an election or by a raised level of awareness of the election? It is particularly hard to draw conclusions in what might be termed a 'declining market'. The claim that turnout has fallen by less than it might have done without advertising is less arresting than comparisons of rates of increase.

We were also interested in discovering what plans local authorities had to advertise future local elections, in ways different to those already used. Some of these proposed innovations included:

'Posters outside polling stations one week before election day in 1999. Only to be done in 50% of cases so that effectiveness can be measured against those without posters'
(North West metropolitan borough)

'Different poll card format for European elections - sent out early with absent vote application and information on Proportional Representation'
(South East district)

'Six loudspeaker cars with 'Vote on Thursday' message'
(South West unitary authority)

One South East district was consulting the public on what could be done by

'Considering the outcome of the focus groups to look at alternative ideas'

The survey certainly found support for the idea that more authorities were taking it upon themselves to make voters aware of forthcoming elections, but only a few had paid attention to measuring the effectiveness of such advertising.

INFLUENCES ON ELECTORAL PARTICIPATION

Perceptions of turnout levels
Authorities were asked about their perception of the turnout level in their authority compared to like authorities. The same question was posed in the Rowntree research in 1996 when we found that 30% of local authorities claimed their turnout to be either 'good' or 'quite good'. In 1999, the figures are similar. Table 24 shows that 32% of respondents claimed to be either 'good' or 'quite good'. In 1996, just 10 authorities (3% of the total) admitted their turnout rate was 'below average' or 'poor'. Three years later, this figure has increased to 17 cases (6.9%). The actual average levels of turnout among authorities in each category lend some credence to these judgements.

Table 24:What is your perception of the level of turnout in your local elections compared to like authorities? Would you say that it is,
	%	n	Actual ave. turnout
Good	8.1	20	39.2
Quite good	23.9	59	36.4
Average	61.1	151	33.0
Below average	4.5	11	27.1
Poor	2.4	6	22.5

We compare in Table 25 the average turnout among those authorities who thought that their performance was 'good' or, alternatively, 'below average' or 'poor' with the mean for their type of authority in 1998/99.

Table 25.Perceived and actual turnout levels 1998/99
	Ave % turnout - 'good'	n	Ave % turnout' below ave' or 'poor'	n	Ave % turnout in tier
London boroughs ('98)	39.3	4	29.1	2	34.6
Metropolitan boroughs	30.9	1	22.0	7	26.1
Unitaries	38.0	1	24.4	3	31.5
Districts	39.8	14	30.4	5	35.8

The figures in Table 25 suggest that most authorities had a realistic impression of their turnout levels. However, one London borough and 4 district councils claimed to have 'good' turnouts despite polling below average for their tier, and one district council thought it was 'below average' despite having a turnout some 2 percentage points better than the tier mean. More significantly a very large number of authorities were content to report their turnout as 'average' despite falling short, in some cases by a considerable distance, of the overall mean.

The impact of local factors on turnout

Positive influences - ward level
All authorities were asked what local factors they felt had a significant positive impact on local election turnout in certain wards in their local authority. Four main groups of factors were suggested; political reasons, local issues, social make-up of the electorate and 'other' reasons.

The main reason put forward to explain the high levels of turnout in particular wards were 'political'. This term encompasses features such as active political parties, a 'real' contest, marginality, well-known Independent candidates, 'good quality candidates' (South East unitary authority), 'very active councillors who are well supported' (North East unitary authority), 'well-known local candidates' (North West metropolitan borough) and 'the involvement of local councillors in the community' (North unitary authority). One North West metropolitan borough summed up one interpretation,

'turnout was higher in marginal wards; in others, electors felt their vote would make no difference'

Local issues which provoke the concern and interest of the electorate can also have an influence on the level of turnout in particular wards. These include issues such as planning, 'Nimbyism', rural development schemes (West Midlands district), a bypass and a marina (South West district), and speed bumps and local development (South East unitary authority). One respondent commented,

'A highly charged battle between two factions in one ward concerning whether or not a new village hall/community centre should be built, resulted in a higher turnout than elsewhere'
(South East district)

Closely linked to local issues is an active local press. Without the involvement of the local media, it is harder for a local issue to receive publicity and spark debate. The role of the local media is discussed more fully in the analysis of the case studies in Section 3.

The third reason suggested by our respondents for a high level of turnout in certain wards is the social make-up of the electorate. One officer commented that,

'There is a high turnout in the... ward because of a high proportion of elderly people'
(South West district)

Turnout is also perceived to be higher where there is an 'intelligent, professional electorate' (South East district), where there are 'well educated voters with social responsibility' (East Midlands district) and 'where the wards are predominately owner-occupants' (London borough). Other respondents wrote of the importance of a 'voting tradition, particularly in rural wards' (South East unitary authority) and of a 'strong community ethos' (London borough).

Positive influences - local authority level
The second part of this question was to ask respondents if there were any local factors which they thought influenced the level of turnout across their local authority. The responses we received were similar to those above. Political activity was important; local issues were also mentioned, but not deemed to be as influential; and socio-economic variables were also put forward as an explanatory factor.

The main difference in the responses to this question about the turnout at the local authority, as opposed to ward level, concerns the role of the local authority itself. Officers argued that what the authority does and how it is perceived by the electorate can have a positive effect on the level of turnout across the authority. These things include, 'The Local Plan' (West Midlands district), 'publicising the council and its services' (North unitary authority), 'having a good consultation process' (South East unitary authority), 'friendliness of staff' (North unitary authority), 'promotional work by the Voter Services Office' (South East district), 'provision of polling stations close to the electorate (East Midlands district.), disabled access (London borough), 'having polling stations in supermarkets, closer to the electorate and increased publicity (South West district), 'ethnic language details in polling district' (West Midlands metropolitan borough), and 'continually advertise during four weeks leading up to the election using leaflets and the radio' (West Midlands district).

Other factors mentioned by officers, but beyond the direct control of the local authority, included, 'holding elections by thirds (East Midlands district), 'having small polling districts (North West metropolitan borough) and 'holding Parish Council elections on the same day' (East Anglia district).

Less mention was made by our respondents of local issues which affected the whole authority. Nonetheless, it is clear that issues such as 'major road proposals' (South East district), 'bin service and council tax reduction' (North West metropolitan borough), 'development in rural areas' (South East district) and 'Donnygate issues' (North metropolitan borough) have the potential to impact on turnout across an authority in particular years.

Regarding the social make-up of authorities, one factor mentioned by a number of respondents was the age profile of residents. One officer wrote that the authority is, '... a coastal district with many retired persons who tend to take their franchise more seriously' (East Anglia district). Similarly, a rural South West district reported, 'We have an elderly population - they tend to be more interested in local issues'.

Some authorities claimed that it was not only individual wards where a tradition of participation was important. One East Midlands district commented on 'a local culture of turning out to vote', while a North West district highlighted the importance of a 'sense of involvement with local politics'.

Negative influences - ward and local authority levels
Just as the factors above may help to explain the high turnout in certain wards and across the local authority, the lack of these factors may lead to a low level of turnout. Several respondents wrote that low turnout in some wards is the result of no political activity as the ward is safe for a political party. Although parties field candidates in such wards, no leaflets are delivered and no canvassing is done. An officer from one authority explains the situation,

'There is a general feeling of disenchantment with local politicians and local government. Very little campaigning by party activists - many 'abstaining' voters expressed the view that if activists could not 'be bothered' then why should they bother to turn out and vote'
(East Midlands district)

The absence of local issues can also depress the level of turnout. One respondent wrote that,

'Turnout is low in some wards because there is no particular contentious issue to arouse public interest/feelings'
(West Midlands district)

The socio-economic make-up of a ward and authority is believed by our respondents to have an effect upon the level of turnout. As one North East metropolitan borough put it, 'there appears to be some correlation between levels of deprivation and level of turnout'. Other likely indicators of low turnout included, 'bed-sit single person occupations' (East Midlands districts), high density social housing (London borough), 'large student population' (North unitary authority), and 'transient population' (South East unitary authority). One London borough was clear that the high turnover of electors was responsible for the low turnout in some wards. Approximately 22% of its total electorate are new electors each year, with the figure reaching above 25% in two wards. In another London borough, the turnover of electors is 40% annually, which is believed to have a depressing effect on turnout levels.

Other factors believed to have a negative impact on turnout are specific to a particular area, such as wards that have a high percentage of service voters or of short-term seasonal employees.

Many respondents were prepared to accept that the local authority has an important role to play in influencing the level of turnout. One council accepted that it was 'not a high profile local authority' (South East district) and another complained about the 'lack of publicity' of council activities (West Midlands metropolitan borough). A London borough noted that most of its publicity was 'adverse'. Some authorities believed that issues of finance were related to turnout. One South West district asserted that the 'level of council tax increase' influences the numbers of people turning out to vote, while a North unitary authority thought the fact that they had the 'second highest council tax in the country' meant 'the public feel they have no control over the local authority'.

On the other hand, authorities were clear that many important factors lay outside their control: they could do little about the level of activity of political parties or the socio-economic character of the area; one South East district was explicit in blaming low turnout on the 'perceived diminishing role of the local authority'. These are all issues we will subject to further examination in Chapters 2 and 3.

CHANGE IN PRACTICE

Changes to electoral registration practice
Our survey makes clear that changes to electoral registration have become common over the past few years. Drivers for change have included the issuing of more detailed guidance by the Home Office and the Association of Electoral Administrators. Fifty-six local authorities (25.5%) have adopted some of the recommendations of the Rowntree report (1996). Another nineteen authorities mentioned that they had read the report and either some of the recommendations were put into practice before the report, or they are currently under consideration. Authorities have taken and implemented recommendations from each of the chapters on registration, absent voting, poll cards and polling stations, and turnout.

One of the most popular changes was to target attainers and students in a bid to increase registration. There is also a significant amount of evidence to show that authorities are now giving more publicity to the facility of absent voting. Another popular innovation has been a review of the location of polling stations. Some councils have tried to ensure that they are all within walking distance for the electorate, others have assessed their accessibility and yet others have experimented with different locations for polling stations. Changes have also been made to the appearance, content and timing of the delivery of poll cards.

One London borough listed the initiatives they had taken from the Rowntree report and applied since 1996,

'Early start to the canvass, payment by results, mention of credit, consult other departments before deleting non-respondents, joint publicity with adjoining borough(s), publicising absent voting, early despatch of poll cards, application form published in council newspaper, access information on poll card and consultation with disabled groups, more ramps etc.'

Encouraging electoral participation
We were also interested in the schemes and initiatives that local authorities had introduced specifically to encourage local election turnout. Many of these initiatives revolved around advertising. This took place in a variety of arenas including local newspapers, council newspapers and the local authority's web-site. This type of advertising does not have to cost the authority any money. One South West district wrote that their media campaigns