The ten pilots in May 2019 have provided more evidence about what it would mean for people to show ID at polling stations in Great Britain, building on the five pilots held in 2018.

More local areas took part this year. There was a wider range of rural areas and urban areas, and they included a better mixture of people: different areas have different populations, with varying socioeconomic profiles.

There were also fewer differences between the pilots using each model, which meant that the evidence from this pilot scheme is more robust. However, as was the case with the five 2018 pilots, the ten areas piloting in 2019 are not fully representative, in socio-demographic terms, of many areas of Great Britain.

Looking at the evidence from these pilots and our previous research, we can say that:

  • A large majority of people already have access to an acceptable form of officially issued photo ID from the lists used in these pilots. These people would not have a problem showing it in a polling station if they had to do so.
  • Allowing only existing forms of officially issued photo ID would not be accessible for everyone. Some groups of people would find it harder than others to show photo ID in a polling station, although this could be mitigated if locally issued photo voter cards were easily available for all.
  • Locally issued ID that includes a photo – like the electoral identity card currently provided in Northern Ireland – would be more secure than locally issued ID without a photo.
  • Asking people to show two pieces of non-photo ID would not necessarily be more secure than showing their poll card. It would be more secure if one piece of ID had to be an official document like a birth certificate, but that would make it less accessible.
  • Asking people to show their poll card would be less secure than a locally issued photo ID. The poll card could be made more secure, for example through changes to the form of the poll card with the addition of printed security features, and by reviewing the process for checking them in polling stations. These would still be accessible for everyone, but could be more complicated for Returning Officers to produce.
  • Using scanners or other technology to check the validity of poll cards in polling stations would be much more complicated and costly for Returning Officers and polling station staff to deliver. These checks would not necessarily add more security than visual checks by polling station staff.
  • The experience of taking part in the pilot scheme appears to have had a positive impact on people’s perception of the security of the polling station process, and on their confidence in it. This varied within each pilot model, across individual local authority areas.
  • Returning Officers and their staff ran the elections successfully in the pilot scheme areas, and there were no significant administrative issues in any pilot area. Polling station staff were satisfied with how polling day went and were confident that they could manage the process of people showing voter identification at future elections.

Introducing a voter ID requirement

The data and findings presented in our evaluation build on the evidence base provided by the 2018 pilot scheme. This evidence further clarifies the way in which a voter ID scheme could be delivered in Great Britain.

However, we are not able to draw definitive conclusions, from these pilots, about how an ID requirement would work in practice, particularly at a national poll with higher levels of turnout or in areas with different socio-demographic profiles not fully represented in the pilot scheme.

If the policy is to be developed further, Government and Parliament should consider carefully the available evidence about the impact of different approaches on the accessibility and security of polling station voting in Great Britain.

This should include evidence from the experience of polling station voters in Northern Ireland, who have been required to show ID since 1985 (including requiring photo ID since 2003), as well as the evidence from the local pilot schemes in both 2018 and 2019.

We have identified three key areas for further consideration:

A photo ID requirement would provide the greatest level of security, but each of the models that have been piloted in 2018 and 2019 would provide some level of improved security compared with the current rules.

Government and Parliament should consider what level of security is proportionate to the risk of personation fraud in polling stations.

While a large majority of people already have access to an acceptable form of photo ID, allowing only existing forms of officially issued photo ID would not be accessible for everyone.

To make sure voting at polling stations remains accessible, there would need to be other options for people who do not already have an acceptable form of photo ID. This could involve providing free of charge locally issued photo ID, as currently provided for electors in Northern Ireland.

Alternatively, it could involve allowing voters to use their poll card – on the current model or a different model – as the primary or secondary route to proving identity, depending on the level of security required.

The pilot scheme has shown that some ID options would be more complicated for Returning Officers and polling station staff to deliver.

The relative security benefits of these options would need to be considered alongside the impact on the administration of election procedures, particularly polling station processes.