When I wrote my last article for the Constitution Society in December, it focused on the impact on local authorities of the forthcoming introduction of ID to vote. Following the elections on May 4th there is now a wider discussion taking place: the ‘how was it for you?’ of the electoral community. From legislators to the Electoral Commission, from returning officers to presiding officers, this question is starting to be answered.
As an election observer I, and the others that observe with me, have a privileged access to the electoral process. We can stand in polling stations and, subject to the rules around secrecy on these matters, we can observe the process throughout, reporting on our findings. This access gave our team an unparalleled opportunity to gain insight into the election process on May 4th and, importantly, first sight of those that were being affected by the introduction of the requirement to show ID to vote.
Democracy Volunteers, the organisation which I lead, observed in over half of the councils with elections – 118 of the 230. We observed almost 900 polling stations on polling day and saw 1.2% of voters turned away because they did not possess the correct ID or did not have ID at all. As we spend between 30-45 minutes in each polling station, we cannot necessarily record those that return later in the day. Of those 1.2% excluded, our observations identified that over 53% were ‘non-white passing’. The new policy would appear, based on our first observation of ID to vote to affect this group most, and as others have postulated since the election, would also seem to affect older voters disproportionately as well. This despite older voters having more forms of ID available to them to prove their identity to elections staff, such as elderly travel passes.
The discussion since the election has been moderated by the apparently limited impact it appears to have had, but those of us who saw the process at first hand can see further problems ahead in the 2024 round of elections and the likely general election within months of that.
There are three fundamental problems with regarding May 4th as being something the Government can easily pass off as ‘nothing to see here’. These are: was this election a true test of ID to vote? Can the systems put in place for voter ID cope with the potentially increased challenges of a larger election? And, perhaps most challenging: how do we know it worked – were there really only 1.2% of people ‘turned away’?
I will start by answering my final question first, as the data we collected can shed some light on the question. The simple answer is – probably not. Whilst our observers saw 1.2% of those presenting themselves at the presiding officers’ desks being turned away, some of these should not have been. Some had valid passports from Commonwealth countries (which were valid for this election), and we also saw others turned away at the door by the ‘greeters’ who, as a consequence, never made it to the desk to be recorded. Nor were those that saw the large signs inside and outside polling stations and were turned away long before they encountered the staff in the polling stations recorded. Some would also have been put off by party tellers who could have reminded them unofficially and others who saw the news before leaving home were also excluded – these people were impossible to record. 1.2% underestimates those who did not vote because they lacked valid ID – some will argue the opposite, but this is the lowest number that was recordable – as it only reflects those that were turned away by the presiding officer – not other hurdles in the process.
Secondly, are the systems in place sufficient to cope with the increased workload in future elections? The Voter Authority Certificate, which voters could apply for online to allow them to prove their identity, would seem to be a temporary solution to a problem that will persist. In Northern Ireland, the Electoral Office for Northern Ireland issues photo ID cards for those that need them. In England this May voters could apply for a paper equivalent via a government website, with the application then sent on to the local council for approval and then issued subject to it being deemed valid. There were approximately 90,000 applications for an estimated 900,000 who might have needed them in England this May. To analyse this group, we need to know how many were duplicate or rejected applications. To understand how this process went, the Government will need to answer three questions – what percentage of those applications were approved, what was the makeup of those who were rejected, and what number were able to remedy any problems to ensure they could vote before the final deadline for applications? There are also two other systems in place that could have alleviated the challenges with voter ID. First, the use of emergency proxies, where up until 5pm on polling day a potential voter can nominate a proxy if they can show that there are specific reasons they cannot themselves vote – such as they have lost their ID. Our observers saw no one being offered this route. They could, in theory, if the legislation allowed, also be offered a ‘tendered ballot’ where a voter is issued with a substitute ballot and it is only counted in very specific circumstances, we saw none being offered. So-called ‘off ramp’ options that we see in other countries, where a voter can cure their ballot or simply has longer to resolve their problem, are simply not available in an electoral culture where ballots are counted as soon as possible after voting ends.
Finally, was this a true test of the introduction of ID? This in some ways is the most difficult question to answer. 230 de facto voter ID pilots in the more rural parts of the country are perhaps not a destruction test of this new policy. Our observers did see more problems in urban areas, especially those with a more diverse population. With elections next year in London and other major urban centres, this policy could well come under greater strain, but such centres will also have those turning out who are the most committed to voting. May 2024 will also see the first major elections conducted following the introduction of the rest of the Elections Act 2022, covering issues such as postal voting. In turn, this could see further strain placed on ID to vote, as the parties’ capacity to encourage postal voting may be more limited.
The next Westminster general election will see the largest hurdle for voter ID but, by then, elections staff may have had two or more trial runs in England. In Wales they will have had one. In Scotland, however, the next election will be a testing ground for voter ID north of the border in a context where large numbers will use it for the first time. Not until 2024/25 will we know the full impacts of ID to vote – but we might reasonably assume that 1.2% of those who arrive at polling stations will be turned away and those turned away will be disproportionately from ethnic minorities. For now, that is the data that has been collected. Over the coming months we should see further evidence as to those most affected, and of any government plans to ameliorate these issues going forward.
John Ault.
Dr John Ault is the director of the UK-based election observation group, Democracy Volunteers. He has observed elections across the UK, much of Europe, and in North America. He has previously been an academic, elected councillor, and election campaigner. He is also a former chair of the UK’s Electoral Reform Society.
The Constitution Society is committed to the promotion of informed debate and is politically impartial. Any views expressed in this article are the personal views of the author and not those of The Constitution Society.