There may be trouble ahead

Wider issues such as party politics aside, some of the trickiest challenges are around how we cast our votes and how those votes are counted.

Each of us will be able to cast up to 38 votes, choosing from a field of candidates that is generally predicted to be in the region of 80 to 100, but could quite possibly be bigger.

SACC has ruled out electronic voting in 2020 (more on this shortly) so there is no getting away from the fact that the ballot paper will be on the large side: less of a ballot slip and more of a ballot broadsheet, perhaps.

We certainly won’t be the first to have an unwieldy ballot paper.

The Nigerian presidential election earlier this year featured 73 hopefuls vying for the post: the authorities there published a video demonstrating how to roll and then flatten the paper in order to fit it into the ballot box.

In Australia, meanwhile, ballot papers for senate elections can be over a metre wide, meaning voters can’t even open them out flat inside the polling booths.

However, it was the recent Indian general election that produced what is surely one of the biggest ballot papers of all time: 185 candidates contested the seat for the Nizamabad constituency – 178 of them disgruntled farmers, standing to make a point about crop prices. Mercifully, members of the electorate could cast just one vote each, or we might still be waiting for the result. (The BJP candidate won, in case you were wondering.)

One of the interesting things about lists generally is that their order matters. Names that appear at the beginning and end of a list will have an advantage over those in the middle, thanks to what cognitive scientists call the primacy and recency effects – basically a subconscious bias in favour of the top and tail, to the detriment of everything in between.

It might sound ridiculous but position on a ballot paper really does influence the result: a 2009 study of English local government elections found that a candidate positioned highest on the ballot order was more than twice as likely as the candidate beneath to finish top on the party slate, and four times more likely to finish ahead of the candidate whose name was listed third.

Randomising ballot papers gets around that problem but creates others – such as the length of time needed to cast and count the votes, especially where each elector has up to 38. It also complicates things if electronic vote counting machines are used, which – with potentially over a million individual votes to count in our next general election – is an option that should of course be investigated.

However superficially sensible, though, electronic vote counting machines are not a panacea. They are essentially a big black box with ballot papers going in one end and a set of numbers popping out the other, but they’re vulnerable to both technical malfunctions and deliberate interference, and the only way of verifying their results is by a manual recount. Cost benefit analyses on UK e-counts suggest that they’re more expensive than a manual count and don’t necessarily deliver time savings either, so I’d encourage SACC to do their homework carefully.

In this digital age we expect to be able to do pretty much anything from a laptop or smartphone, so instinctively it feels as though i-voting – internet-enabled voting, as distinct from electronic voting machines at polling stations – is the next natural embodiment of this democratic process. The committee is looking to introduce i-voting in Guernsey, but has suggested that it won’t be ready for 2020 simply because there isn’t enough time to manage it as an IT project. This, however, misses a far more important point: the question of whether it should be introduced at all.

The technology exists and works already, but i-voting hasn’t been widely adopted because of fundamental concerns around its security and accuracy.

Estonia is the only country in the world where a significant proportion of ballots in national elections are cast online.

An international group of e-voting security experts carried out an independent evaluation of the Estonian i-voting system and were alarmed by what they found. The system was vulnerable to cyber attacks from foreign powers, server-side attacks, for example malware introduced by a dishonest election official, and client-side attacks – a hacker with a bot that can overwrite votes, for instance.

The analysts urgently recommended that Estonia discontinue the use of the system and went on to say that, in their opinion, ‘no country in the world can do internet voting safely, and it’s going to be a decade – if ever – before we’re able to solve some of the central security problems at stake’.

With the countdown clock ticking towards 17 June 2020, that at least is one issue the States’ Assembly & Constitution Committee should put firmly on the backburner – but even without it, the length of their list of challenges might rival the Nizamabad ballot paper... I wish them luck.