Guidance

Elections Act New Burdens funding methodology: overseas electors, postal and proxy voting revisions, online absent vote applications and EU citizens’ voting and candidacy rights

Updated 8 December 2023

A. Introduction

1. The Elections Act 2022 has and will continue to introduce a suite of measures to improve the integrity of the electoral system. To date, policies including Voter Identification (which requires photographic identification to be shown at the polling stations) and Accessibility (which changes accessibility requirements for disabled voters) have already been implemented.

2. Further policies are due to be implemented in the coming months, including:

a. Overseas Electors (OE): removing the 15-year limit on voting rights for British citizens living overseas and extending the registration period for these voters from one year to up to three years.

b. Postal and Proxy Voting Revisions (PPVR) and Online Absent Vote Applications (OAVA): limiting the number of electors for whom someone can be appointed to act for as a proxy to four (of which no more than two may be domestic electors), requiring electors to reapply for a postal vote every three years, introducing identity checking for absent vote applications, and introducing an online service for absent vote applications.

c. EU Citizens’ Voting and Candidacy Rights (EUVCR): The removal of automatic voting and candidacy rights of EU citizens who arrived in the UK from 1 January 2021 onwards (except for citizens of member states who have a bilateral treaty with the UK or longstanding voting arrangements, such as Ireland or Commonwealth countries).

3. The introduction of these policies will place an additional burden on local authorities (LAs) when delivering elections and electoral registration activities, and this is known as a New Burden. A New Burden is defined as ‘any policy or initiative which increases the cost of providing local authority services.’ The New Burden doctrine states that the leading department must ensure that the net cost of all additional new burdens placed on LAs is fully assessed and properly funded.

4. This paper sets out the modelling approach undertaken to allocate New Burdens funding to LAs to support the delivery of changes to elections and electoral registration activities made by the Elections Act.

B. Background

5. Engagement with the sector has been crucial over the last several years in shaping and developing the analysis on the Elections Act. The chart below illustrates the engagement that the department (previously Cabinet Office) has had with the sector in developing the evidence base on the various policies.

Figure 1: Sector engagement

6. The evidence gathered as above has fed into the New Burdens funding model, which is broken down into two stages:

a. Policy cost model – The first stage of the analysis calculates the net cost impact on local authorities, at a national level, of implementing each of the Election Act policies. These impacts are evidenced through various data and research collated over the past few years, such as pilot studies, survey data and desk-based research. The national cost for each policy is then allocated to a LA level, which is explained in more detail in Section C.

b. New Burdens Funding Model – The individual LA level cost outputs from each policy model are then collated into a single funding model to estimate the total costs to be allocated to each LA. The various funding mechanisms (e.g. grant, justification-led bid or hybrid approach) are then applied along with the associated payment timeline (i.e. when the LA is expected to receive the funding). Further information is provided in Section D.

7. The New Burdens funding strategy considers all aspects of how the New Burdens costs will be met, including an assessment of funding mechanisms against evaluation criteria and risks, when payments should be made, how they should be made and to whom, taking account of the divergence in policy and delivery across the UK. Three funding approaches have been considered: up-front grants, retrospective funding via a justification led bid (JLB) and a combination of these approaches (hybrid). Further information on the mechanisms is included in section D.

C. Policy cost models and allocation

8. All unit cost estimates outlined in this paper are for financial year 2023-24 only. They are subject to change for future years, depending on inflation and other supply-side factors.

9. Analysis to identify the costs for Overseas Electors (OE), Online Absent Vote Applications (OAVA), Postal and Proxy Voting Revisions (PPVR) and EU Citizens’ Voting and Candidacy Rights (EUVCR) uses a mixture of government data, survey data provided by LAs, desk-based research and supplier costs.

C.1 Overseas electors

C.1.1 Volumes methodology

10. A systems dynamic model was built using United Nations migration data and Office for National Statistics (ONS) migration data to produce projections for the size of the future OE franchise.

11. The main cost model applies the following assumptions evidence by ONS, National Records of Scotland (NRS), and DLUHC data to estimate the volumes of OEs passing through various stages of the electoral registration and voting process:

a. Proportion of OEs making an application to register to vote

i. The percentage of the franchise registered during the years of expected UK Parliamentary elections (2024 and 2028) will be similar to those recorded across 2017 and 2019 (around 17%).

ii. 92% of applications will be made online, based on the percentage of recent Register to Vote applications currently made online. An exception applies to Northern Ireland, as online applications are unavailable to OEs, so 100% of applications are made on paper.

b. Proportion of OEs providing further evidence of their identity and/or connection to a qualifying address

i. 90% of all applicants will apply on the basis of previous registration, and 10% on the basis of previous residence.

ii. 5% of previous residence applicants will have left the UK before age 18.

iii. LAs will request additional documentary evidence from 15% of all applicants when confirming their identity and/or connection to qualifying address. Of these, 67% (10% of total applicants) will not be successfully registered via the documentary evidence process and will be asked to complete an attestation. Of these, 64% (6.4% of total applicants) will not be successfully registered as an OE.

c. Proportion of OEs successfully registering

i. 94% of applications will result in a successful registration.

d. Proportion of OEs who are sent a reminder to renew

i. 66% of OEs are sent a reminder to renew each year, of which 43% do not renew. If the OE does not respond to a renewal reminder a second is sent by the ERO.

e. Proportion of OEs making an absent vote application

i. 71% of new applications will also apply for an absent vote, based on the portion of OEs currently holding absent voting arrangements.

f. Proportion of OEs voting in person and by absent vote

i. 29% of OEs do not have an absent vote arrangement in place

ii. 43% of OEs have a postal vote in place

iii. 19% of OEs have a proxy vote in place

iv. 10% of OEs have a postal proxy in place

C.1.2 Cost modelling methodology

12. Data collected from electoral registration teams within LAs is used to inform estimates of the time required to undertake processes at each stage of the OE journey. A cost for each process is then calculated based on the average hourly wage from the survey data, costs associated with contacting OEs using Notify pricing and voting costs based on the 2017 UK Parliamentary conduct costs as the 2019 costs have not yet been fully settled. Detail is provided here as we are yet to publish the secondary legislation Impact Assessment. The main cost model applies the following assumptions:

a. Stage 1 - Initial application to register

i. On average a register to vote application from an OE will take 5 minutes to process for an online application and 13 minutes for a paper application.

b. Stage 2 - Verification of identity and connections to a qualifying address

i. It will take an ERO on average 48 minutes to verify an applicant’s previous registration and 20 minutes to verify an applicant’s previous residence. The previous registration check assumption considers the automated checks of digital registers as well as manual checks of physical registers.

ii. When further evidence is required, it will take an ERO 8 and 12 minutes to request and process documentary evidence via email and by post respectively, and 9 and 13 minutes to request and process an attestation via email and by post respectively.

iii. The cost of printing and posting a documentary evidence request (two-page) letter is assumed to be £1.13 and an attestation request (three-page) letter is assumed to be £1.21. When information is instead requested via email, we have assumed there is no cost additional to an ERO’s time.

c. Stage 3 - Confirmation / rejection of registration

i. It will take an ERO on average 4 minutes to add an OE to the register and 7 minutes to explain their absent vote (AV) choice to them.

ii. The cost of printing and posting an outcome (two-page) letter is assumed to be £1.13. When the outcome is instead sent via email, we have assumed there is no additional cost above an ERO’s time.

d. Stage 4 - Renewal declaration

i. It will take an ERO on average 11 minutes to send via email and process a declaration to renew, 17 minutes to send via post and process a renewal and reminders and 3 minutes to remove an OE who does not respond from the register.

ii. The cost of printing and posting a renewal (two-page) letter is assumed to be £1.13. When the outcome is instead sent via email, we have assumed there is no additional cost above an ERO’s time.

e. Stage 5 - Absent vote applications

i. On average an ERO will take 6 minutes to process a postal vote application, 9 minutes to process a proxy vote application and 15 minutes to process a postal-proxy vote application.

ii. The cost of printing and posting an outcome (two-page) letter is assumed to be £1.13. When the outcome is instead sent via email, we have assumed there is no additional cost above the ERO’s time.

f. Stage 6 - In-person and proxy voting

i. Where newly enfranchised OEs choose to vote by proxy or in person at a polling station, there is likely to be a small additional burden on polling stations. The additional conduct cost per voter is £0.12.

g. Stage 7 - Postal and postal-proxy voting

i. The additional conduct cost per postal and postal-proxy OE is £4.52. This covers the staff time of preparation, printing, postage, opening and verification of the postal votes of newly enfranchised OEs.

C.1.3 Funding allocation methodology

13. Table 1 outlines the funding mechanism and timing for the OE New Burden funding lines. Further information on the split between the Returning Officer (RO) and Electoral Registration Officer (ERO) costs is explained in Section D.

Table 1: OE costs

Cost line	Funding timing
ERO costs
Registration costs	Hybrid: August 2023, April 2024 and November 2024
Training costs	Grant: August 2023
RO costs
Voting costs	This will go through the Consolidated Fund as the costs relate to a UK Parliamentary General Election

C.1.4 Allocation Mechanism

14. The proportion of OEs in each LA as of March 2020 is used to allocate funding at a LA level. This is the closest data point post UKPGE 2019 at the LA level, which is how our funding is allocated. A more recent data set from December 2020 is available, however, there is a decrease in number of OEs registered across almost all LAs.

15. The main limitation of using this approach to allocate funding is that future OE registrations may not follow historical registration trends of OEs. The risk of over/under funding is mitigated by using a hybrid funding approach. Where upfront grant funding is insufficient, Local Authorities will be able to apply for further funding via the Justification Led Bid Process (JLB) where claims are justifiable and reasonable.

C.2 Online Absent Vote Applications & Postal and Proxy Voter Revisions

C.2.1 Volumes methodology

16. A combination of administrative and survey data is used to estimate the total volume of absent vote applicants under the current and new system once the Election Act measure have been implemented. The volume of electors who opt for an absent vote is calculated using known volumes of postal vote or proxy vote holders per country over the previous 4 years, which is projected using ONS population forecasts. The percentage of electors who will be eligible to vote in planned elections is then calculated using data on which local authorities are due to hold planned elections and their respective electorates. This figure is then applied to the volume of electors opting for an absent vote. Our modelling also assumes that the introduction of the online system will see absent vote applications increase by 15%, and this is applied to our total volumes.

17. Applications will progress through a system of identity checks. The modelling uses electoral registration data from the Electoral Commission’, to create the following assumptions to estimate the volume of applicants, both online and paper, who pass through each stage per financial year:

a. Stage 1 - Initial rejection - Of the paper applications, 10.9% will be initially rejected due to being illegible, compared to all applications submitted online progressing.

b. Stage 2 – Central government data matching - 91.3% of applicants who reach the subsequent central government data matching stage will pass.

c. Stage 3 – Exceptions process – It is assumed that 9.8% will fail the central government data matching process and will need to undergo the exceptions process for further documentary evidence to be requested.

d. Stage 4 – Attestation process - Based on the canvass data, 6.5% (of the 9.8% who go through the exceptions process) will fail and be required to submit an attestation.

e. Stage 5 – Rejected applications - Finally, should the applicant fail attestation, they will fail the application process. With the additional rejected illegible applications, this totals to 16% of paper applications being rejected, and 5.8% of online applications being rejected.

18. The modelling assumes that the application process will not begin until January 2026, however we assume that some (around 10% of the annual postal voter stock until 2025-26, at which point all remaining postal voters will be contacted to reapply) will choose to apply for a postal vote unprompted from January 2024 onwards. An internal survey with local authorities revealed 34.7% of proxy applicants require a second reminder to refresh their signature currently, and that 15.9% of applicants sent a reminder to reapply do not.

C.2.2 Cost modelling methodology

19. A range of data sources are used to estimate the total cost, including survey data, commercial estimates, and data from private beta testing of the online system. We have used internal survey data to estimate the average electoral registration administration staff wage per minute, and this is used in conjunction with the estimated staffing time required to perform each stage of the application process.

20. Where processing times varied due to a difference between online and paper applications, a weighted total cost per stage is calculated based on the relative proportions of applications received by each method. Multiplying this by the volume of applicants, as calculated previously, gives a total cost for this stage. There are also correspondence costs for each stage of an application, namely the cost of printing, postage, and an envelope for a one-, two- or three-page A4 letter sent within the UK. The size of the letter is dependent on the type of correspondence required for each stage. The cost of this is based on figures from Gov Notify.

21. The staff unit cost and correspondence unit cost are combined for each stage to give a total unit cost, and then multiplied by the volumes to give an estimated cost per stage of the application. These are then combined to result in several cost lines. More information can be found in the published Impact Assessment.

C.2.3 Funding allocation methodology

22. Table 2 outlines the funding mechanism and timing for the OAVA/PPVR New Burden funding lines. Further information on the split between the Returning Officer (RO) and Electoral Registration Officer (ERO) costs is explained in Section D.

Table 2: OAVA/PPVR costs

Cost line	Funding timing
ERO costs
Registration costs	Hybrid: August 2023, April 2024 and November 2024
Training costs	Grant: August 2023
RO costs
Voting costs	Hybrid: August 2023, April 2024 and November 2024
Training costs	Grant: August 2023

C.2.4 Allocation mechanism

23. The funding for OAVA/PPVR allocated to local authority level uses the volume of absent voters from the 2019 European Parliamentary Election. The rationale for using this data is that it is the last UK-wide dataset which has a complete breakdown of absent voters by local authority, rather than by constituency.

24. The primary limitation of using this as a method to allocate funding for this policy is that historical breakdowns of absent voters may not necessarily reflect the behaviour of future absent voters, especially following the introduction of the online system. Where upfront grant funding is insufficient, Local Authorities will be able to apply for further funding via the Justification Led Bid Process (JLB) where claims are justifiable and reasonable.

C.3 EU Voting and Candidacy Rights

C.3.1 Volumes modelling methodology

25. A combination of ONS and DLUHC data is used to estimate the volume of already registered EU citizens that will be contacted in relation to the EUVCR franchise change. Contact will occur through two reviews: a preliminary database review and, if required, a correspondence-based review process.

Volumes – database review

26. An estimate of the number of EU citizens that will go through the database review in England, Wales and Northern Ireland is based on data from ONS Population Estimates publication. These figures provide a breakdown of the estimated number of EU citizens from each EU nation there are in each home nation. As Commonwealth and Irish citizens are not affected by the franchise change and will not be contacted, they have been removed from this estimate.

27. The analysis estimates that 2.2 million EU citizens (that is all registered EU citizens) will go through the database review across England, Wales, and Northern Ireland. Of this, we estimate that 1.2 million EU citizens will receive a positive determination at the database review stage.

Volumes – correspondence-based review

28. Using data from Individual Electoral Registration Digital Service (IERDS) showing how many EU201 voters have registered in local authorities since the Implementation Period Completion Date (IPCD) date, we have estimated how many EU20 voters will have registered by the EUVCR implementation date (7th May 2024) by projecting the initial figure forward, assuming a constant rate of change.

29. We estimate that around a million people will go through the correspondence-based review process in total.

C.3.2 Cost modelling methodology

30. The costs of contacting all of the EU20 citizens registering to vote between IPCD and the policy implementation date involve the one-time fixed costs to set-up the One-Time Submission Link (OTSL) functionality and a training allocation alongside costs of communication and staffing that vary depending on the volume going through the correspondence-based review process.

31. A range of data is used to estimate the total cost of the franchise change, including survey data and commercial estimates. We have used internal survey data to estimate the average electoral registration administration staff wage per minute, and this is used in conjunction with the estimated staffing time required to perform each stage of the application process.

32. The staff unit cost and correspondence unit cost is then combined for each stage to give a total unit cost, and then multiplied by the volumes to give an estimated cost per stage of the application.

33. The methodology to calculate the volumes and costs for EUVCR can be found here.

C.3.3 Funding allocation methodology

34. Table 3 below shows the funding mechanism and timing for the EUVCR New Burdens funding lines. Further information on the split between the Returning Officer (RO) and Electoral Registration Officer (ERO) costs is explained in Section D.

Table 3: EUVCR cost lines

Cost Line	Funding timing
ERO costs
Communication costs (including costs of correspondence, time taken for personal contact attempts etc)	Hybrid: April 2024 and November 2024
Staff costs
Training costs	Grant: August 2023

C.3.4 Allocation mechanisms

35. The allocation of grant funding for EUVCR is based on the number of registration applications of EU citizens from the 1st January 2021, using data from the IERDS.

36. It is acknowledged that this may not necessarily reflect the number of EU citizens arriving in the UK (e.g. they could move from one LA to another). To mitigate this, funding has been provided via a hybrid mechanism so LAs can apply for further funding if required. Where upfront grant funding is insufficient, local authorities will be able to apply for further funding via the justification led bid process (JLB) where claims are justifiable and reasonable.

D. Funding strategy

D.1 Funding mechanisms

37. Paragraph 7 outlined the different funding mechanisms that were considered for New Burdens funding. A hybrid approach of up-front grant funding and justification led bids (JLBs) has been selected as the best approach for New Burdens funding as it will ensure funding is distributed in a consistent and transparent manner, whilst also providing upfront access to funding and minimising the workload on EROs.

38. Cost lines are split into RO and ERO activities. ERO activities relate to registration activities for which the ERO is responsible, for example processing postal vote applications online. RO activities include activities for which the RO is responsible and relate to elections activities, for example, polling station staff training.

39. Table 4 summarises the funding mechanism and rationale for each cost line.

D.2 Funding Timings

40. Payment timings follow from the funding approach. Grants will typically be made earlier in the financial year (April), ahead of May elections. JLB payments will be made later in the financial year, giving enough time for LAs to gather evidence of spend and apply for JLBs following the May elections.

41. The exception to this is Financial Year 23/24 where the grant will be released in August 2023 so LAs have sufficient time to use the funding to prepare for the May 24 elections. It should be noted that payment timings and mechanisms may change and are subject to DLUHC Senior Responsible Officer (SRO) and Ministerial approval.

Table 4: Funding mechanism for each cost line and rationale

Cost Line	Funding mechanism	Rationale
Overseas electors
Registration costs	Hybrid	Total volumes are uncertain so funding will be provided via a hybrid mechanism to ensure all LAs/VJBs have some funding up front to support the additional costs which may arise from OE applications. However, LAs/VJBs will also be able to apply for more funding via the JLB process if the costs are justifiable and reasonable.
Voting costs		This will go through the Consolidated Fund as the costs related to a UK Parliamentary General Election
Training costs	Grant	As per Tranche 1 funding for training costs, this is a predictable amount and will therefore be provided via an upfront grant.
OAVA/PPVR
Registration costs Voting (postal ballots)	Hybrid	There is a degree of uncertainty around volumes who will use the new online system and thus the overall potential impact on absent votes. Funding will therefore be provided via a hybrid mechanism to ensure all LAs have some funding up front, and those who need more can request it via the JLB process in November 2024 where funding is justifiable and reasonable.
Training costs	Grant	As per Tranche 1 funding for training costs, this is a predictable amount and will therefore be provided via an upfront grant.
EUVCR
Communication Staff time	Hybrid	Total volumes are largely known but the method of communication may vary depending on the LA, so a hybrid approach has been chosen to ensure we have flexibility to give further funding to LAs if needed via the JLB process in November 2024 where funding is justifiable and reasonable.
Training costs	Grant	As per Tranche 1 funding for training costs, this is a predictable amount and will therefore be provided via an upfront grant.

D.3 Funding Sources

42. Reserved elections are funded by the Consolidated Fund. As such, costs resulting from the new policies that are incurred by ROs in relation to reserved polls will be met by funding from the Consolidated Fund and not New Burdens funding. Where reserved polls are combined with other polls, the full cost will be split between the Consolidated Fund and New Burdens funding, at the ratio of reserved to other polls, in the usual way.

43. Costs incurred by EROs (registration costs) that result from the Elections Act will be funded as New Burdens only.

D.4 Divergence

44. For LAs in Scotland who are members of a Valuation Joint Board (VJB), funding to meet ERO costs will be paid to the VJB while funding to meet RO costs will be paid directly to the LA. LAs that are not members of a VJB will receive both ERO and RO portions of funding.

45. LAs in England and Wales will receive both ERO and RO portions of funding. Two-tier councils in England will receive all New Burdens funding at the District Council level (including for County Council elections). Payments will be made by the Section 31 grant.

D.4.1 Overseas Electors

46. An additional communication provision is provided for Wales so that communications can be bilingual.

D.4.2 OAVA/PPVR

47. The implementation of OAVA/PPVR will diverge between England (where it applies to all elections), Scotland (where it applies to UK Parliamentary elections) and Wales (where it applies to UK Parliamentary and Police and Crime Commissioner elections). The new proxy limit will apply to all reserved polls in Great Britain and Northern Ireland. It will still be possible for an elector to appoint a proxy on a long-term basis at reserved elections (the 3-year reapplication measure does not apply to proxy voting). Further information on the assumptions around divergence can be found in the OAVA/PPVR Impact Assessment in paragraphs 32-36.

D.4.3 EUVCR

48. The introduction of EUVCR will be applied to local elections, Police and Crime Commissioner (PCC) elections and polls in which individuals can vote or stand based on being on the local government register in England. In Northern Ireland, the changes will apply to local elections and Northern Ireland Assembly elections, and it will only apply to PCC elections in Wales. There will be no impact in Scotland.

49. An additional communication provision is provided for Wales so that communications can be bilingual.