If you dig into the UK election in-person voting system, you’ll find it’s designed around a brilliantly balanced ideal, making voting as easy as possible whilst ensuring that fraud can not be committed at a scale that will normally make any kind of difference.
I’m a hacker by trade, I may work for the good guys, but my job is to look at systems and processes and work out how they could be abused (ideally before the bad guys do) and a few years ago I was a polling clerk at a very quiet polling station, with no internet access and only the Electoral Commission Polling Staff Handbook as reading material, so I spent most of the day studying the guide and working out ways I could subvert the in-person voting system.
One thing is obvious reading the guide, everything possible is done to keep the barrier to voting as low a possible, especially for that that impromptu last minute decision to vote. No polling card, no problem, kids or dog with you, no problem, need physical assistance no problem. It’s all about ways to facilitate voting, not prevent it.
So what about fraud? You may be surprised to know I came up with many ways I could commit voter fraud, but none fitted an important part of my original statement “committed at a scale that will normally make any kind of difference“. Sure I could see numerous ways I could fraudulently cast a single vote, but none that scaled. You’d either need multiple visit to the polling station times (increasing the chances of being noticed or the person you were impersonating having already voted) or have a conspiracy of many people and given each one could only cast a small handful of votes, you’d need a big conspiracy (and even small conspiracies are a big risk as humans are generally not great at keeping secrets).
So, how does mandatory voter-ID change things? Firstly is raises the barrier to entry, unlike polling cards, if you didn’t receive it (in time or at all), no vote. Lost it, no vote. Left it at home, no vote. Unplanned chance to vote as you were near, no vote. But more importantly, abusive spouse or parent who feels that if afforded a secret ballot you may not vote the way you want, they simply hide your voter-ID and boom, no vote.
What do we get on the other side of the scale in return, technically some improved security, but as it’s almost impossible to pull off in-person voter fraud at scale, the number of actual votes changed would be zero and the number of results changed would almost certainly zero. In 2017 44.6m votes were cast, there were 28 allegations (not proof) of fraud, that’s 0.000063%.
It’s spending money and ensuring people who could and would have legitimately voted, don’t, for no real benefit.
I’ll spare you the political analysis on who this would impact the most and why, as I hope even supporters of the parties who may gain from disenfranchising voters will think it’s an abhorrent idea, and just say …..
If you care about electoral integrity, focus on the postal voting system. It’s far easier to abuse (especially at scale) and doesn’t offer many of the same protections (i.e. from coerced or intimidated voting or ensuring the secrecy of the vote). Anyone who thinks fixing in-person voter should take priority over mail-in voting either doesn’t understand the UK election voting system, or has something to gain from a higher barrier to entry when it comes to in-person voting.
2 responses to “Why Voter-ID in UK elections is mostly a terrible idea”
How hard would it be to get a copy of the electoral register and bus people round early in the morning to vote on other people’s behalf?
Obviously that would risk discovery if the legitimate voter arrived later. So, you would need to find non-votes, those recent hospital admissions, bereavements, thosevarrested and detained, surely not that hard.
Combine that with a marginal seat and voila.
I agree that the postal vote is far easier to rig, and should be stopped, but I also believe that if you don’t look then you don’t find, so walkin voter fraud is likely to be much higher than reported.
But that comes back to conspiracies not scaling. A bus load of people and not one would break the confidence, be it to a friend or a national newspaper in return for a large payout? Not one of those involved would have an attack of conscience and turn whistle-blower on their co-conspirators?
Now keep in mind the number of suitable marks (i.e. peoples whose vote you could steal) for any single ward is quite low and an issue would be raised if :-
– The mark was wrongly identified
– The mark managed to attend anyway (released after arrested, discharged from hospital etc)
– The local election office had managed to update the voting record (which they do right up to a day or two before, in cases like bereavements)
– The mark had already opted for a postal or proxy vote
– The polling officer / clarks knew the mark (whilst clerks and officers aren’t necessarily from the ward they are working, it is very common and often their first choice of location)
– In smaller wards when one police officer is assigned multiple polling stations and rotates between them, they spotted the same people attending more than once.
The crux of the article wasn’t that it was impossible (far from it), only that it’s a very high risk maneuver with little chance of making a difference. Whilst 1% or 2% margin sounds tiny, keep in mind that in the 2019 election, the smallest majority was 57 , The next 4 closest would have needed between 100 and 200 votes (and suitable marks), so a team of 10-20 hitting 10 polling stations upwards each. It’s further complicated by the fact that these margins are obviously not known ahead of time, so predicting how big your conspiracy needs to be is impossible.
Frankly it’s much easier hit the voting role to find residents at care homes who can be persuaded to vote in return for a nice trip out. That’s far easier to pull off and doesn’t risk jail time (and yes, this does happen).
BTW Walk in voter fraud is looked for, not just the casual checks and balances above, but the whole reason for the controversial recording of ballot counterfoil IDs against voter roll IDs (which can only be correlated by a court order) allows suspected in-person voter from to be investigated after the fact. If you look at the 2019 data, of the 595 cases of alleged electoral fraud only 142 were alleged voting fraud, of those there were only 3 convictions or cautions (all people who voting as a close family member).