As in 2018, Returning Officers and their staff who were responsible for the May elections ran their pilots successfully. There were no significant administrative issues in any pilot on 2 May.
Polling station staff told us that they were satisfied with how polling day went and were confident that they could manage the process of people showing voter identification at a future election.
Additional staffing and training across all pilots was needed to run the pilot.
The IT-enabled pilots worked well, but any Great Britain-wide roll out would need a significant level of resourcing to ensure a smooth implementation.
Returning Officers (ROs) and their staff were responsible for running the elections and the pilot processes. We have considered how the requirements of the pilots affected their ability to do so.
For these pilots, as in 2018, ROs were responsible for delivering the public awareness activity. This was funded by the UK Government but delivered by each local authority.
Significant time and cost, in each pilot, was devoted to making sure voters were aware of the ID requirement. ROs used information gathered through equality impact assessments to help identify the most effective ways to communicate the requirements to different groups of people in their areas.
This is likely to be different if the requirement to show ID were rolled out nationally, when the Electoral Commission would assume responsibility for a national public awareness campaign.
However, in that event it is likely that ROs would still carry out smaller scale public awareness activities locally, as they do currently to encourage people to register to vote.
Issuing local ID/replacement poll cards
In each pilot, ROs ensured the provision of a free and universally available acceptable ID. In the photo only and mixed model, electors could apply for a local identity document (until 5pm on the day before polling day).
In the mixed and poll card model they could apply for a replacement poll card (up to 5pm and 9pm on polling day respectively).
The 2019 pilot scheme offered a limited test of the process and cost of issuing local ID to electors as there was limited take up of this route.
In the photo only pilots, Woking issued 24 and Pendle issued 70 photo ID cards. Feedback from Pendle indicated that additional staffing was used to manage these applications and that they were not evenly spread over time; most applications came in and needed to be processed in the two weeks before polling day.
However, in the mixed model pilots very few local identity documents were applied for or issued; Broxtowe issued two and Braintree issued one.
This lower take-up is likely to be partly a result of the wider ID requirement which offered greater choice for voters, compared to the photo only pilots, and partly because electors could also apply for a replacement poll card (between 4% and 9% of electors used a poll card as a form of non-photo ID).
The number of replacement poll cards ranged from one in Broxtowe to 300 in Derby. Replacement poll cards were also issued to 69 electors in Watford and 40-50 in Mid Sussex. We have heard no significant concerns about this process beyond the additional costs of keeping the council building open for electors until 9pm on polling day.
Overall, the feedback suggests this was a manageable process if appropriately resourced. However, some ROs and their staff did express concerns about the challenge presented by the potential for a much greater volume of either local ID applications or replacement poll cards ahead of a UK general election, particularly with a high volume of applications coming close to the deadlines.
These challenges could also have a significant impact on those people who apply for replacement poll cards close to the deadline, and their ability to show ID on polling day.
Staffing and training
ROs in each of the pilots made changes to staffing in order to deliver their pilot, although this varied significantly. The approaches ranged from a limited increase in central elections team resources (to support public awareness work or issue local ID) to deploying additional polling station staff on 2 May. For example:
- In Craven, the RO employed one additional presiding officer (out of a total of 30), seven additional poll clerks (out of 45) and three additional polling station inspectors (out of six)
- In Broxtowe, the RO added 12 poll clerks (out of a total of 103), two additional polling station inspectors (out of a total of eight) and two further staff in the central elections team
- In Watford, the RO (as they did in 2018) added two polling station inspectors (out of a total of six)
- In Pendle, the RO added two polling station inspectors (out of a total of seven) and three temporary, part-time staff in the central elections team
There were no consistent patterns of how staffing was changed across the different pilot models; mostly changes were made based on the individual RO’s assessment of the risk and issues locally.
There was also no overall consensus on whether any increase in staffing at all would be required to administer voter identification at future elections with higher turnout. Some ROs expected that they would need additional staff in polling stations but others felt that it would be manageable at existing levels. Some areas that used additional staffing on 2 May said they would not anticipate using it again, even at elections with higher turnout.
More and longer training sessions were required to support staff in delivering the pilots. To manage the additional training, some areas replaced online training with face-to-face sessions while others reduced the number of attendees per session.
Our survey found that polling station staff thought that the training was delivered well across the pilot scheme. The majority of staff agreed that their training prepared them well for polling day and that the instructions on the types of ID that could be accepted were clear.
Feedback from ROs again suggests that this was a manageable increase in work but highlighted that there was an additional resource requirement (e.g. in training fees for longer sessions).
This variation in evidence from the pilot scheme means that, in the event of Great Britain-wide implementation, the UK Government would need to carry out a further, careful assessment of what the resourcing impact would be for ROs.
Feedback from ROs and their staff indicate that they delivered their pilot on 2 May without significant problems. Across the pilot scheme, polling station staff were satisfied with how polling day went.
In photo pilots 80% of staff were very satisfied, increasing to 81% in poll card pilots and 86% in mixed model pilots. This also reflects the experiences of Electoral Commission staff observers, who largely saw the process working well.
A high proportion of all polling station staff agreed that asking voters to prove their identity had little or no impact on their work, although the extent to which they agreed varied across the pilots.
Do you agree or disagree that asking voters to prove their identity had little or no impact on your work on polling day?
Staff in poll card pilots were more likely to agree that it had little or no impact, while staff in photo and mixed pilots were more likely to disagree than those in poll card pilots. This could be due to the quantity of the different acceptable identification types in these pilots or the additional work involved in recording information for the evaluation in the non-IT enabled pilots.
The majority of polling station staff said that they would be confident in replicating the process at another election, with almost three quarters across the pilots saying they were very confident.
Mid Sussex and Watford both used IT in the polling stations to scan QR codes on electors’ poll cards. In both, the system worked well and there were no notable issues on polling day related to the IT.
Electoral Commission staff who observed in these pilots also saw this on polling day, with staff appearing confident in using the tablets to scan poll cards and being able to resolve any issues.
However, the planning and setting up of these systems did require a significant amount of time and resource commitment from the electoral administration teams, the software suppliers and Cabinet Office. The commitment of time and resource largely stems from the development of a system to provide the necessary level of assurance and resilience to ensure no problems arise on polling day, such as ensuring there were no issues as a result of connectivity problems in polling stations.
ROs raised some concerns about whether the level of individual support provided for the pilots by the software supplier would be possible if a system was being used across Great Britain.
We have not had access to final data on the overall cost of delivering this element of the pilots, which is likely to have been significant. As an indication, IT equipment costs for similar activity at the 2018 pilots were estimated at £332 - £659 per polling station, though anecdotal evidence suggests that relative savings were made for 2019.
While the pilot costs do not necessarily translate easily into a cost if implemented across Great Britain, it is highly likely that a notable level of resourcing would be required to ensure a smooth implementation.