Russia’s remote voting proposals will lessen transparency and trust

In an already controlled environment, the latest moves to change electoral systems in Russia have the potential to further tighten the grip of the Kremlin. A Bill to enable candidate registration signatures to be collected via the state services app has been amended at the last minute to allow remote voting via a number of means. It passed the Duma (the lower house of parliament) after lawmakers were given just 36 minutes to see the proposed amendments. Covid-19 restrictions limited the amount of media and public scrutiny that was possible of the procedings.

What appears to have emerged from the process is a Bill that will allow for the development of internet voting, for postal voting and to expand the range of people who are qualified to vote at home on election day. In addition, for health reasons, voting will now be allowed in the precincts of the polling station as well as the voting room itself.

There is an axiom that any time you take the ballot paper out of the control of election officials, that vote becomes less secure and more susceptible to fraud. These new measures all remove the oversight that election administrators – and observers – will have over the process. It is perhaps no wonder that Russia’s leading independent election observation group – Golos – have said of the changes: 

“Their implementation without simultaneously ensuring guarantees of effective control will increase the level of distrust of citizens in elections.”

To take the changes one by one:

Internet Voting

Internet voting is often seen as the solution to many election problems. In the UK it was trialed as a response to declining turnout in the early 2000s. But just because someone tells a pollster that they are more likely to vote if they can do so from home via the internet, doesn’t mean they will actually do so. TV programmes which use internet voting have many hours of positive broadcast coverage and still only get a small proportion of their audience to vote.

As I have written before, internet voting is doubly problematic. First in that it takes the vote out of the polling station. Second, that it is reliant on ‘black box technology’ so the voter cannot see directly how their vote contributes to the result and there is no paper trail. If some malign actor, either within the election commission or hacking in from outside, wanted to fix the result then it is far more possible with internet voting and almost impossible to prove.

The only country which successfully uses internet voting for national elections is Estonia which has spent many millions (in a very small country) on security. This includes a reader for every household so that voters can insert their national identity card to be validated. Even then, I would argue, it is not completely secure as other members of the household could vote using a person’s card – particularly if they are vulnerable or disinclined to vote. And the chances of pressurized voting are obvious.

Postal Voting

Postal votes have been the subject of many election fraud cases around the world. It is not quite true to assert, as President Trump does, that all postal voting is riddled with fraud. But postal votes are subject to many of the risks of other forms of remote voting.

Where a person cannot make it to vote on election day, postal votes can be a good thing. In the UK we used to have a ‘for cause’ system which meant you needed a valid reason for asking for one. Now we operate an ‘on demand’ system. This avoids the need to tie up doctors and employers and for election administrators to deal with lots of paperwork.

Where the UK – and others – have largely failed is when they seek to adopt universal postal voting – ie every voter is sent a postal vote to their registered address. I have dealt with such issues here. In short, if a person is not aware that their vote is being sent by post then it is easy to abstract and cast illegally. Switzerland is a country where all-postal ballots do work well, but is a very different electoral culture.

In order to have an effective postal vote system, a country needs to have a means of verifying that the application and the resultant vote come from the registered voter. You don’t want to allow others to apply and then vote on your behalf. This means having lots of staff, lots of time and specialist signature matching software. My experience of the Russian system is that the elections staff are generally pretty well trained and motivated, but they are short staffed and would need a significant increase in their budgets and allocation of high quality hardware from local administrations which are often reluctant to let them have anything other than the oldest computers.

Traditionally, Russia has sought to address the problem of people being away from home on election day by allowing ‘place of stay’ voting. This system, managed by the state services app, allows a voter to move their polling station up to a couple of days before the election. If you are away from home on business or an economic migrant, you can simply change where you vote to a local station. And there are special polling stations created in hospitals and railway stations, and even on ice-breakers and at the Antarctic Research Station. So with all these options, are postal votes really needed?

Early Voting

Early voting has been used for some time in a number of countries. It is not the most susceptible to fraud as it still requires the voter to attend a polling station (their own or a central hub) where they are dealt with by election administrators in the same manner as on election day. However, it can stretch the resources of party and other observers who are there to ensure that nothing untoward happens. And it can make it easier for the same voter to cast multiple ballots by going from polling station to polling station.

Home Voting

Home voting has been the traditional means by which Russia allows those who cannot come to a polling station on election day to vote. It has always been restricted to the old and people with disabilities and requires an application by the elector which is then adjudicated by the polling station committee. If approved, then on election day a subset of the committee, plus observers, takes a small version of the ballot box to the home of the voter. Although in most cases this is a workable solution, it requires the intrusion into the voter’s home of up to eight people and it is often difficult to ensure the secrecy of the vote.

The proposal now is to allow carers as well as those being cared for to vote in this way. That may seem a logical step, but simply extends the problem, I would suggest.

Precinct Voting

The proposal is to allow voters to cast their ballots not just in the confines of the polling booths, but also within designated areas within the precincts of the polling station building – in courtyards, for example. This is being done, it is claimed, for health reasons.

Fairly obviously, loss of secrecy is a big problem with this proposal. If people are wandering around with their ballot then it can be seen by others. In my experience in Russia at least one third of voters do not bother to fold their ballot after completion. 

I don’t know whether there will be polling booths set up in the courtyards where voting will be allowed, but the chances are that these will be as unpopular as those in the officisl voting room if there is a fear of Covid-19.

Perhaps the other major problem is that election officials and observers will find it difficult to track what is happening. This makes frauds such as carousel voting, illicit pressure, family voting and proxy voting all more easy to achieve.

On the other hand…

I certainly would not suggest that the Russian voting system is in the dark ages. The place of stay voting system is very good indeed and deserves to be studied by many supposedly advanced democracies. And the state services portal makes it possible to accomplish a lot of tasks related to the elections process in a simple and speedy manner. That’s a boon to voters as well as to the state. If there were to be a form of internet voting then this might well be the basis for such a system.

That said, however, it is clear that the changes being developed as a result of this Bill are not going to make the Russian system more secure and will actually do only a little to enhance access to voting. Fundamentally, they open the way for those who wish to rig the vote to do so. Citizen confidence in elections stems from knowing that votes are cast freely and that the result is an accurate counting of only legitimate votes. Sadly I think that this Bill takes Russia away from those principles.

Get ready for easy to manipulate fake audio to join the election disruptor’s arsenal

Are we ready for (extremely realistic) fake audio clips to join ‘Photoshopped’ images and DeepFake videos in the arsenal of electoral manipulators?

unnamedThe Campaign Tech newsletter (part of Campaigns and Elections) has the following story:

“Remember Lyrebird, the fake audio startup? They’re running a private beta for a new tool called Overdub that lets users upload an audio clip and then amend what’s being said by the speaker by simply editing the transcript of the original audio clip.

Descript (Overdub/Lyrebird’s parent) stresses that they’re committed to only letting users synthesize their own voice. 

But we all know that beating bad actors is a game of whack-a-mole. And even if Descript is successful on this front, it’s absurd to think every fake audio vendor will be (or will want to be).

The takeaway: sophisticated off-the-shelf audio manipulation tools are going to become as accessible as photo-editing tools.”

And while the American fear is that this might be used by political opponents, much of the rest of the world will be concerned about overseas disruptors, some of whom want to tilt the balance in favour of a particular candidate and others who simply want to cause chaos.

Such fakes, if uncovered, would fall under Twitter and Facebook’s policies of labelling fakes as such and demoting them in their algorithms. But they have said they will not do so if such tweets or posts come from newsworthy figures such as world leaders.

Voter Suppression and poll failures – the latest government election efforts

Two significant bits of news as regards elections in the UK have emerged in the last few days:

Voter ID requirement would be suppression effort

First, the government has said that it wants to press ahead with plans to require voters to take photo identity documents to the polling station. This measure, although a common requirement in countries with ID cards, has been dubbed an attempt at voter suppression in the UK where the most likely to be hit would be young people, the elderly and people from minority communities.

Two rounds of pilots have taken place in local council elections where voters were required to bring either photo ID or poll cards in order to be allowed to vote. In the most recent trial in five local authorities, more than 750 people were turned away and did not return with the correct documents although many others did so after initially being refused. In contrast, during the 2015 elections (which included a general election) there 665 complaints of electoral fraud but most related to nominations or postal voting. There were 26 cases reported of ‘personation’ at a polling station, but none resulted in convictions. A total of around 51.4 million ballots cast.

But if most other countries require ID, why shouldn’t the UK? 

First, because we don’t have a national identity card system. (That’s a whole separate civil liberties argument). So we have to rely on forms of identity that we do have – passports and driving licences. But only about 78% of the adult population have passports and only about 80% have driving licenses. These raw statistics would suggest that 96% of the population would have one or the other, but actually if you don’t have a passport then you are much less likely to have a driving licence. So the real figure is about 11% – one in nine of the population. And these people are statistically more likely to be much younger, not in work or from minority populations – in other words, statiustically less likely to be Conservative voters.

In order to get around the ‘No-ID’ problem, the government would have to issue a special form of voting ID card. With a potential 5.6 million to issue, this would be an expensive and complex business (it would have to be free to the applicant) and very difficult to promote to the audiences who need to see and understand the message.

All this begins to look very similar to efforts in the United States to suppress the likely support for opposition parties. 

I used to work for the Electoral Reform Society. As part of that work I highlighted the much more real danger of electoral fraud connected with postal votes and this work resulted in significant changes to the law to try to stamp this out. Even with those changes I would still regard postal voting as a much greater risk of fraud than in-person voting. In response to the new government proposals, the ERS have said:

“Ministers should focus on combating the real threats to our democracy – dark ads, disinformation and dodgy donations – rather than suppressing voters’ rights.”

Electoral Commission criticises government over Euro-poll failures

Second, the Electoral Commission has blamed the government for the failures of the European Parliamentary elections when thousands of eligible voters were denied the chance to cast their ballot. The Commission has blamed “outdated laws” and “the failure by the government to act on recommendations made four years ago”.

The Commission also said voter confidence in the election was lower than in any other recent polls, denting the democratic contract with the public.

Thousands of EU citizens who were on the electoral roll were turned away as they had not been informed they also had to fill in a form confirming they would be exercising their vote in the UK rather than in their country of origin. There were a number of reports of councils giving misleading information to voters who made enquiries and of authorities who failed to process returned forms properly.

Many UK citizens living overseas complained that they did not receive their ballot papers in time to have their vote count.

Social media platforms must do more to prevent election attacks

It’s pretty clear that social media attacks have a real potential to affect not just elections, but political life in general. That’s why Facebook’s ‘two steps forward, one step back’ strategy is so disappointing. They – with their subsidiary WhatsApp – are the biggest players in the social media market and they have a responsibility to act. Only when platforms are completely transparent will election authorities be able to act and we, the voters, will have confidence that our elections are not being distorted.

How election regulations work

Different countries have different laws regarding elections and this applies to online campaigning and social media too. In most countries, the principle means of regulating election campaigns is via spending limits – although there may be a range of other controls. Parties and candidates are required to submit a spending return after the election (and sometimes interim returns mid-campaign). They may have to open a dedicated bank account and there may be limits as to who can contribute and how much.

Many countries view day to day non-commercial uses of social media as being essentially free and so they do not fall under the scope of election expenses. Even websites are often viewed as being low cost and are an under-regulated form of influencing votes.

Such ‘free’ uses include:

  • Setting up a Facebook page to promote a candidate or party and gather ‘likes’ for them. People who have ‘liked’ the candidate can then be sent messages and other information. Likers and other users can view live streams of campaign events
  • A twitter account to promote the candidate, to encourage retweets and to retweet others (endorsements, party leaders etc)
  • A WhatsApp account to create groups and to share information among those groups and encourage other group members to forward the information to others.
  • An Instagram account to share images and engage in conversation with followers and others.

There are, of course, many other social media platforms, but they broadly fall into one of these basic use profiles.

Increasingly, social media is also being used to host paid-for advertisements of a political or campaign nature during elections. These may come from parties or candidates themselves and can be positive or negative in nature. Or they can come from third party actors within the country or from outside. Different rules apply in each country with some countries permitting third party groups to spend money campaigning during an election either for or against a candidate or on the basis of issues. And whilst some countries permit funding by citizens living overseas, broadly speaking no country permits out of country election spending by non-citizens.

Why parties use social media

The advantage of social media advertising is that it allows an advert to be targeted at a specific audience. To take Facebook, the company knows enough about its users that it can sell advertising so that it reaches a very specific group. It is easy to target, for example, women aged 24-35 in a particular city. And, the company knows much more than simple demographics. They also know about an individual’s likes and dislikes (quite literally because of the ‘like’ buttons clicked). So Facebook can sell advertising enabling very precise targeting. And because the user data is not shared with the advertiser – they only receive personal information if the recipient of the advert chooses to share it with them – this practice is seen as compliant with data laws around the world.

The attractiveness of social media to parties, candidates and other political campaigners is obvious and not a bad thing. Lots of voters complain they don’t know enough about what politicians or parties stand for, so this means of communication should help. But a platform that allows genuine communication is also open to fake news and outside interference.

unnamedA disclaimer here: As a campaign manager in the 2016 EU referendum, I commissioned and paid for Facebook adverts on a number of occasions. I was able to define the audience I wanted to see these and I thought they were good value for money. I didn’t of course, have access to the private data that the platform used to target that audience. Our advert spending was properly declared to the Electoral Commission.

The Cambridge Analytica/AIQ case is something different. In this case data was harvested for one reason and then given or sold to political advertisers for completely different reasons. Facebook has been shown to have known about this illegal transfer to the extent that they have been fined the maximum amount permitted in the UK. But even if the company acted illegally in that case, it does not currently inhibit the legal act of selling advertising by Facebook and other social media companies.

Recent problems

There have been a number of scandals to hit election related social media in recent years:

  • During the 2016 UK referendum on membership of the EU the Electoral Commission found that the Vote Leave campaign illegally co-ordinated their campaigning with BeLeave by passing on funding which was spent on social media advertising;
  • During the 2016 US Presidential Election, it is alleged that Russia (and possibly China) sought to interfere with the contest through the promotion of fake news and the use of ‘bots’ to spread false information. (Other claims about Russian interference have been made but they don’t come under the heading of social media;
  • During the 2018 Brazillian Presidential Election, it is claimed that fake news aimed at both candidsates has been spread via WhatsApp groups;
  • During the 2018 Macedonian name referendum, it is alleged that many hundreds of websites, Facebook groups and other means were created from outside the country to promote a boycott and therefore to lessen the credibility of the outcome which was expected to be a Yes vote;
  • Allegations of foreign interference have also been made about French, German and other elections in Europe and elsewhere.

In addition to social media, voters may see election related content on news sites, gossip sites, blogs and so on. Frequently, these sites encourage interaction via comments and these are often un-moderated. Whilst parties can campaigns will endeavour to push messages out via these sites – as they do through mainstream media – the comments sections are often the territory where activists and others will seek to promote points of view and stories which are less factually robust.

So what action have Facebook taken?

They have made two significant changes which are broadly positive. They have required that every political advert carries a form of identification so the viewer can see who produced it. However this ‘imprint’ is often not as clear as one might like, providing little real clue as to who is behind it. A recent example are adverts urging constituents to contact therir MP and ‘stand up for Brexit’. A number or groups have produced these and some are clear whilst others are far from.

Second, Facebook will periodically release the details who who has spent what on political advertising. That’s great, but it won’t be linked to specific content.

They have also announced a ‘war room’ to tackle fake news during there EU elections.

On the downside, Facebook appears to have restricted the ability of plug-ins to monitor advertising content. This has hit the Who Targets Me platform even though the use of plug-ins in that case is entirely consensual. So one of the prime investigators of shady political advertising is no longer able to undertake its investigations.

And, as I’ve previously written, WhatsApp in India has restricted the ability for users to forward messages. However this make the spreading of fake news slightly harder rather than eliminating the possibility entirely.

Has fake news swung elections?

It’s impossible to tell. Governments do not like to admit that they might have come to power or their course of action might have been set via a referendum that was fundamentally flawed. And courts and election commissions have been very reticent in declaring a ballot to be void. That is not to say that it has never happened, but these remedies do not appear to be the most reliable.

Whilst in the past a second country (or people based in a second country) might have sought to influence the conduct of an election by means of radio broadcasts and the like, the advent of the internet, and particularly of social media, has made it much easier to seek to influence an election in another country whether through ‘fake news’ or truthful campaigning.

There is also a question as to how much a vote is actually changed by a piece of fake news. In most cases it appears that a voter is likely to cast their ballot in a certain way and the information they choose to listen to or accept (whether fake or otherwise) simply confirms their choice.

And what constitutes ‘fake’? An outright lie or doctored photo such as the one claiming that the former Brazillian President Dilma was a prodigy of Fidel Castro is simple to categorise. But the ‘£350m for the NHS’ slogan on the side of the Vote Leave bus during the UK’s referendum is not so obviously fake. Had politicians decided to do so, they could have made this come true, regardless of the impact of Brexit on public finances. It is fair to point out that the pretext of the claim – that Brexit would make the UK better off – is probably not the case, but we are then into a political debate – something that should not be policed in a heavy handed fashion, if at all.

However, it does seem probable that there have been significant numbers of votes affected by fake news or international campaigning in various elections and that this is something that should be taken seriously. Respected NGOs in various countries have raised concerns about this issue.

Next steps

Governments across the world have been reluctant even to address this issue. But some have and they have chosen different approaches. In the UK, ministers have said that recognising and ignoring fake news is the responsible of the individual. They don’t propose to take any action to stamp it out. France, however, has indicated that it might try to set up an official body to make rulings. The difficulty here is that such rulings are likely to come after the horse has well and truly bolted.

What seems logical as a first step is for platforms such as Facebook to be much more open about who is funding political advertising and what it says to whom. It is not necessarily for social media executives to do the work of electoral commissions, but they need to enable the official regulators to do their jobs properly. If an individual, organisation or even foreign country is trying to influence elections then this should be clear and, if it is against the law, then action should be taken. But until the social media platforms come clean, this can’t happen.