Elections Bill (Fourth sitting)

Fleur Anderson Excerpts

Committee stage
Thursday 16th September 2021 Public Bill Committees

[Patrick Grady]

Thank you. That is really helpful.

[Fleur Anderson (Putney) (Lab)]

Q Thank you, Mr Millar. It is a pleasure to see you this afternoon, and thank you for your evidence. Can I just ask about the relationship between this Bill and the European convention on human rights, particular the right to vote, obviously? Concerns have been expressed, for example by the Gypsy, Roma and Traveller community, among others, that the Bill will reduce the ability to vote and will therefore contravene article 3 of the first protocol, which states that conditions imposed by the state must not curtail the rights in question in any way. Do you think that the voter ID scheme set out in the Bill is compatible with the European convention on human rights? Does it risk not being compatible? You have said that the law is unclear in some ways—is it unclear? If someone went to a polling station on election day without voter ID and was unable to get it, because the council was swamped by other such requests, and so they could not vote, could they then return to the European convention on human rights and say that their right to vote had been curtailed? Is that possible under the Bill? If there should be changes, what changes would you make?

Gavin Millar: This strand of convention law—by which I mean, whether a piece of domestic legislation is incompatible with the provisions of the convention—does not work on an individual case-by-case basis. It works on the basis that if you have to look at compatibility in a court case, it is at the impact of the domestic rule of law—here, the voter ID provision—across the piece and the whole of the electoral system in the contracting party.

Is the impact of that legislative provision one that can be justified as being compatible with the convention? The convention—Strasbourg—has its own internal set of rules for saying what is and is not compatible. Very few rights are absolute, which is why you can have laws that prevent certain people—criminals and so on—from voting for a period, but to be compatible with the convention they have to be justifiable, in the sense of achieving a legitimate aim, one that is legitimate in that country for that political system and that voting system. It has to be a proportionate means of achieving that aim.

The question here—I accept that it would be assessed by the impact on individual groups of people, such as the Roma, whom you mentioned, but it would be much broader than that—is, if you try to justify what the Government are proposing to do across the electoral system as a whole, can it be justified as meeting a legitimate aim? Is there a problem that is so bad that it needs addressing in this system in this way? Is this a crude or a proportionate way of addressing it? The problem I have with clause 1 is that I cannot see the problem and, even if there is a problem, I cannot see that this is a targeted and proportionate way of addressing it, because it would just sweep out of the franchise somebody who did not happen to have a card or voter ID but was properly on the electoral register and entitled to vote when they turned up.

Why do I say that there is not a problem? You are all politicians, you have been elected and you know how this works, but you may not have looked at this from the point of view of an election lawyer, a criminal lawyer or someone looking at election fraud, which for my sins I have spent a lot of time doing for the past 20 years. The sort of fraud we are talking about here is called “personation” under the RPA. It is an electoral offence—it is impersonation, but misses off the “im” in the statutory historical categorisation. Personation is A turning up at the polling station pretending to be B, who is validly on the register.

It is not a problem of any great consequence in our system, and I speak from experience. Personation cases are almost non-existent. There are reasons why it is not a problem. First of all, it is extremely risky for anyone to try that. You are liable to be caught because somebody spots you and knows you are not that person. It is also ineffective because there is the alternative possibility that that person turns up and votes later, or indeed has already voted and is marked off the register when you try to impersonate them. If you are going to do it, you have to be absolutely certain that the person is dead or is not going to come and vote, and that you will not be found out that way. It is also hugely inefficient compared with other forms of fraud that have been perpetrated, particularly since postal voting on demand. You have to get a range of people, or yourself, to go around different polling stations at different times in the day, and all you get out of each criminal offence you commit is one vote. It is just not efficient or effective as a fraud, so it does not happen.

As I understand it, this came from the 2014 Tower Hamlets mayoral election. There were a whole range of election offences pleaded in that case and looked at by the court. One of them involved some personation at polling stations, but it was not the core problem. If that were the reason we had got to this point, this would be an example of a hard case making very bad law, and I would counsel against that. The fraud that exists in our system, or has existed since 2000, that everybody has read about and knows about, is a very different type of election fraud. One possibility is what is called roll-stuffing in Australia, where you put additional voters on the register who are not entitled to vote in a concerted fraud before the election, and then vote in their name. You normally apply for a postal vote for those non-existent voters at a particular address, and you pick up the postal vote papers and you vote.

There are various other postal vote frauds that were recounted in the cases that have been cited. That form of fraud has been made much more difficult by Parliament and by the administrators because of the cases over the past 20 years, and there are less cases even of that form of fraud, but it is not a form of fraud that would be addressed by this piece of legislation, so what is the problem? What is it achieving? Why is this a proportionate way of addressing it? I have no answers to any of those questions, and of course in a situation where, by common estimates, we have something like 17% of eligible voters not on the register, one wonders why our efforts are not being concentrated on voter registration measures—getting more people on to the register and facilitating them in voting—rather than making it more difficult for them to do it by imposing this requirement, which we have never had.

I appreciate that advocates of the Bill will say, “It is not a lot to do, to get a piece of photo ID or have a piece of photo ID and bring it along to the polling station,” but we need only look at the Windrush scandal to see how many poor people and ordinary people in our society have difficulties with that sort of thing, not to mention disabled people and other discriminated-against groups who do not want to engage with obtaining this sort of identification, for fear that it will open them up to other scrutiny and investigation of an unjustifiable kind. It is wrong on every count, really.

To answer the question, yes, there will inevitably be challenges to this as incompatible with the European convention on human rights if it is introduced, and it seems to me that there is a strong case for doing that. The impact would be considerable, by all accounts—although somewhat unquantifiable—but I just have not seen the evidence that you would be required to produce at a judicial review or at a case in Strasbourg to justify this as an appropriate state interference with the right to vote.

[Fleur Anderson]

That is very concerning. Thank you.

[The Chair]

If there are no further questions from Members, I thank our witness for his evidence.

Examination of Witness

Fazilet Hadi gave evidence.