There is very limited evidence from the May 2018 pilot schemes about whether the identification requirement had a direct impact on the security of election procedures. In particular, it is not possible to evaluate whether the requirements actually prevented attempts to commit electoral fraud at elections.

Overall findings from the pilot schemes

In the areas where the pilot schemes took place, the local police forces have told us that there were no complaints about electoral fraud at polling stations for the May 2018 elections. However, it is not possible to assess whether the identification requirement prevented any actual attempts to commit impersonation fraud.

In other areas where elections took place in May 2018 there were three complaints about electoral fraud taking place at polling stations. The police forces for those areas told us that they have investigated two of these cases but they did not need to take any further action. They are still investigating the other case.

There is no direct evidence to suggest that this difference was because of the identification requirements in areas where pilot schemes took place.

However, the pilot scheme requirements did introduce new identification checks where there have previously been no checks. Because of this, we can conclude that the pilot scheme requirements are likely to have had some positive impact on reducing the potential for electoral fraud by impersonation at polling stations.

We cannot make any assessment from the available data about the nature or extent of this impact, however. Nor can we make any assessment from the data from these pilot schemes about whether there was any difference in the impact on security between the different requirements in each of the pilot scheme areas.

Beyond the pilots: the impact on security at future elections

Further pilot schemes may not help to provide more robust evidence about the impact of an identification requirement on the security of future elections. This is because it is not possible to evaluate whether the requirements actually prevented attempts to commit electoral fraud at elections.

However, more work should be carried out before any further pilot schemes to further assess the relative security of different approaches to checking the identity of voters. It would be particularly useful to gather more evidence on the levels of assurance about someone’s identity provided by different types of documents.

For example, the security of different types of identification documents needs to be balanced against the availability of those forms of identification. This balance should be considered further before future pilot schemes and any decision to implement a voter identification requirement in Great Britain.