Can tech keep the US elections on track during coronavirus?

Paper ballots have held firm in the face of numerous attempts to bring in electronic voting. Is COVID-19 the final straw?

Elections are expensive, time-consuming and labour-intensive affairs. Paper-based ballots are used in the vast majority of countries and for the vast majority of important elections. Each major election accounts for millions of sheets of paper, thousands of litres of ink, tens of thousands of workers and volunteers, untold millions of human hours spent turning up to cast a vote with a pencil and paper, or some other paper-based system, plus many more associated costs and materials. It all seems incredibly wasteful, so why do we persevere?

In fact, there are good reasons. Paper is convenient, simple, low-tech and usable by anyone. It's hard to misread - the "hanging chads" controversy of George W. Bush's election notwithstanding - and it's also surprisingly secure. True, paper is readily available, but to swing an election of tens of millions of votes would potentially require millions of forged items of paperwork being somehow inserted into what is usually a fairly secure logistical chain. This, then, is paper voting's true strength: it's hard to hack.

By contrast, a competent hacker could change tens of millions of votes with a few taps of the keyboard, or at least a well-crafted script. At least, that's the argument put forward by those opposed to electronic voting, and they do potentially have a point.

However, the same argument could be made for electronic banking, online shopping and any other form of transaction that is supposed to be secure. Making electronic voting secure isn't easy, and it's almost certainly harder than its proponents think, but that doesn't mean it's not possible.

In any event, opponents of electronic voting may sometimes take their zeal too far. Paper voting is far from ideal, since not everyone finds it easy to get to a polling station within the necessary time window. Postal votes, as often used in the UK, also run the risk of being used fraudulently. No voting system is perfect, and electronic voting or app voting is no exception: if implemented well, it could be an improvement on the status quo.

The current US presidential election was seen as a potential proving ground for new electronic voting technology even before COVID-19 became an issue. Now - even if the majority of restrictions will hopefully be over by the time the final vote takes place - there's an added urgency. If people can't easily get out to vote, allowing them to vote electronically or by app makes sense.

However, so far the results of electronic voting haven't been spectacular. The Iowa Democrat caucus in February this year proved to be disastrous when the data-collation layer of an app provided for voting was shown to be dismally incapable of doing the job, to the extent that the final numbers weren't known even two weeks later.

"I'm really disappointed that some of our technology created an issue that made the caucus difficult," said Shadow CEO Gerard Niemira. "We feel really terrible about that."

Although not a conventional election, this was hardly a good advertisement for electronic voting. If relatively small numbers of votes can't be accurately processed, what would happen with tens of millions?

Shadow isn't a large, well-known, international company, unlike Microsoft. The latter had its shot at proving the benefits of electronic voting later in February, when its ElectionGuard software was used in Fulton, Wisconsin, to allow voters to choose candidates for the state Supreme Court. That process went off without any apparent hitches, using a combination of existing voting systems and encryption software. Crucially, ElectionGuard doesn't replace existing voting systems; it adds accountability.

Encryption is going to be a vital part of any effective electronic voting system, probably in the shape of a blockchain of some sort that transparently yet anonymously keeps track of votes and prevents anybody from voting twice, or fraudulently. That's going to be difficult to achieve in practice.

It's not just the US that is facing difficulties with conventional voting structures. Poland has elections scheduled for May, yet it also currently has strict COVID-19 limitations on individual movement. The ruling party wants the elections to go ahead but - unless the science narrative shifts soon - that would appear to be foolhardy. People congregating at polling stations would be likely to spread the virus to others. Electronic voting might offer a way out, but that's already been ruled out by the country's electoral authorities. Votes will be counted by hand and by calculators due to the lack of a reliable electronic alternative.

Similar issues face any country with major elections due this year - and there are plenty of them. Many of these will take place during a time of COVID-19 restrictions and so could potentially benefit from e-voting, but that's unlikely to happen.

To get some idea of the challenge facing e-voting, whether online or by app, one need only look to Switzerland. The country carries out more referenda than any other on a per-capita basis and has dabbled with e-voting since 2003. However, last year there were increasing calls for all electronic voting to be banned for at least five years, following the discovery of a security flaw in the main Swiss Post system. If the Swiss can't make it work, who can?

So, to answer the question posed in the title of this article: no, tech can't keep the US elections on track. If coronavirus lockdowns continue right up to the date of the election - which seems unlikely but not impossible - it won't be e-voting that comes to the rescue. A postponement of the election is the more likely and more logical outcome.