• silicon.com
• Management
• CIO Insights

By Will Sturgeon, 9 May 2005 15:35

COMMENT As the post-mortem begins on the General Election, one of the biggest questions being asked is whether a change in voting methods could have changed the outcome.

On the one hand are questions about the benefit of e-voting in combating apathy and persistent low turnout, which this year averaged 62 per cent nationally. On the other hand are continuing questions over the ease with which election fraud could be committed. Some say e-voting will increase that risk while others say fraud is so endemic currently there is nothing to lose.

On the question of apathy it is clear that even minor changes in the turnout could have made for a very different outcome.

Take for example four closely fought constituencies.

In Sittingbourne and Sheppey where the turnout was a paltry 53.7 per cent, Labour candidate Derek Wyatt won by just 79 votes - or 0.2 per cent of the votes cast.

In Croydon Central where turnout was 60.6 per cent, Conservative candidate Andrew Pelling won by just 75 votes - or 0.2 per cent of votes cast.

In Harlow where the turnout was 62.6 per cent, Labour candidate Bill Rammell won by just 97 votes - or 0.2 per cent of the votes cast.

In Battersea where turnout was 59 per cent, Labour candidate Martin Linton won by just 163 votes - or 0.4 per cent of votes cast.

And these aren't the only instances where minor changes in turnout could have, potentially, changed the result.

Such fine margins between winning and losing really are the difference between a few floating voters or a few who just couldn't be bothered to walk to the polling station.

It's believed the use of e-voting technologies, whether via secure internet sites or even SMS text messages, will provide a significant hike in turnout.

In many constituencies this time around it could well have been the difference between winning and losing. In Cardiff West, the most connected constituency in the UK in terms of voters with home broadband internet access, turnout was just 57 per cent. This may have been due in part to it being a very safe Labour seat but with so many apathetic voters already online there is real scope to increase the turnout nationwide.

Tory and Liberal Democrat supporters may not have thought it worth walking to the polling station or even to the post box but they may have been prepared to log on from home.

Recent research from Detica suggests the use of new technologies would certainly increase turnout at elections. A quarter of voters (25 per cent) said they would be more likely to vote if they could do so by SMS. And almost a third said they would be more likely to vote if they could do so by email (30 per cent) or via a secure voting website (32 per cent).

And even where the influence of e-voting is not enough to sway the result - remember there will be areas where supporters of all parties would vote online in roughly equal numbers - as an electorate we owe it to ourselves to ensure the government is genuinely representative of the nation's beliefs and opinions. For this to be the case we need turnout to be as close to 100 per cent as possible.

But looking at the worst case scenario, what if turnout is ever more than 100 per cent? Or to put it another way: how big is the threat of fraud and how much confidence can we have in electronic systems?

The current system is certainly flawed and open to clear abuses. That's beyond doubt. Members of the silicon.com team who showed up at polling stations without polling cards were not asked to show ID - they were simply presented with a list of names at their given address, and allowed to pick one.

Armed with nothing more than the ability to read house numbers, it is possible to vote as anybody, and importantly to vote more than once.

A little bin-raiding, in order to get the names of occupants, may also help.

However, such incidents of fraud are still conducted one-by-one, the concern as ever with technology is that it provides the opportunity to commit such frauds on a grander scale.

David Porter, head of security and risk at Detica, told silicon.com: "There are certainly problems with the current ballot box system but I don't think it could ever be enough to swing the result of an election. Similarly with postal voting, where fraudsters can either apply for your vote or intercept and manipulate it, while fraud poses more of a threat it would still have to be a very large co-ordinated effort to actually sway the result.

"What worries me more is the next step up - to electronic voting. Like in other realms of life, such as shopping or banking, when things go electronic the fraudsters jump on board and can do things quicker and in greater numbers."

While Porter said a secure system of e-voting with controls in place and monitoring and detection techniques applied to unusual voting patterns could one day be worth considering as a trade-off in favour of higher turnouts, he warned against the trivialising of a General Election.

"One benefit with the current system is that people have to actually get off their arses and vote," he said, adding that such a process of walking to the polling station buys thinking time. We should be wary of technological advances that might enable people to make rash decisions or vote on a whim, he said.

Richard Allan, former Liberal Democrat MP for Sheffield Hallam and a former member of the All Party Internet Group, told silicon.com any unsupervised voting throws up concerns and said he'd be against moves in that direction, even though the Lib Dems perhaps stand to gain more than any other party, in his opinion, due to the demographic and typical profile of floating Lib Dem voters.

Allan told silicon.com: "There is no doubt that many people would find it more convenient to vote from home using their computers but the price we would pay for that convenience in terms of loss of confidence in the results is too great - unless and until the fraud issue can be comprehensively resolved."

"With the recent prosecutions for fraud in the postal voting system, concerns about this issue are growing not diminishing, making any widespread introduction of remote e-voting very unlikely," he said.