By Democratic Audit UK 28/11/2016 Read More →

Referendums, informed voting and the trouble with a ‘truth commission’

Following the EU referendum, there have been demands for a ‘truth commission’ to be set up to oversee future referendum campaigns. Paul Kildea argues that there are significant practical difficulties to the establishment of such a body. These include the possibility of a ‘chilling effect’ on speech, the fact that the accuracy of many controversial campaign statements would be impossible to assess and the probability that the interventions of a ‘truth commission’ would become political flashpoints in themselves. It would therefore be better to focus on other changes that can be made to better prepare voters for their choice at the ballot box – such as improving the design of official pamphlets and the increased use of deliberative mechanisms such as citizens’ assemblies.

A Vote Leave poster in Omagh, Northern Ireland. Photo: Kenneth Allen via a CC-BY-SA 2.0 licence

One of the many talking points to have emerged from the EU referendum in June is whether a ‘truth commission’ should be established to oversee future referendum campaigns. Numerous commentators have expressed frustration at the misleading claims made by both Leave and Remain campaigners, and feel that something must be done to protect voters against the wilful spread of misinformation. In a high profile report, the Electoral Reform Society recommended that ‘[a]n official body – either the Electoral Commission or an appropriate alternative – should be empowered to intervene when overtly misleading information is disseminated by the official campaigns’. At around the same time, a change.org petition called for the establishment of ‘an independent Office of Electoral Integrity (OEI) to factually verify the truthfulness of claims made during political campaigns…with powers to issue fines and factual clarifications’. That petition, which attracted over 165,000 supporters, has received 49 signatures since being published as an Early Day Motion in the House of Commons.

The objectives of improving the quality of referendum debates, and assisting voters to make informed choices, are worthy ones. However, the establishment of a body to monitor the content of campaign statements would be misguided. Efforts to foster informed voting should be directed elsewhere.

Concerns about false and misleading campaign statements

It is understandable why the idea of a truth commission emerged in the aftermath of the EU referendum. In a hard fought campaign, both sides were accused of misleading voters through exaggerations, distortions or outright lies. The Leave campaign was widely criticised for claiming that the UK sent £350 million a week to Brussels, and intimating that it could instead be spent on the NHS. Remain, meanwhile, was singled out for exaggerating the economic impacts of leaving the EU, including a claim that households would be on average £4,300 worse off. Other flashpoints included the release of UKIP’s anti-immigration poster, featuring a huge queue of migrants and refugees and the tagline ‘Breaking Point: The EU has failed us all.’ A survey conducted near the end of the campaign found that nearly one-half of voters (46 per cent) thought that politicians from both sides were ‘mostly telling lies’, while only 19 per cent thought that they were ‘mostly telling the truth’.

While the Brexit debate was particularly fractious, concerns about the spread of misinformation are not isolated to referendums in the UK. On the contrary, referendums everywhere have long been criticised for being prone to unscrupulous campaigning that pays scant regard for the truth. To cite just a few international examples, in the lead up the 1999 referendum on whether Australia should become a republic, No campaigners argued that change would ‘cost us our stability, our certainty and our security’ and would render Australia ‘comparable to Nazi Germany’. And prior to last year’s Irish vote on same-sex marriage, the No campaign circulated material that drew a baseless connection between the referendum and surrogacy laws.

Exaggerations, half-truths and lies are also a feature of election campaigns. However, they are arguably more problematic in relation to referendums. Whereas a candidate elected on the back of misleading statements will have to answer for them after the poll, referendum campaigners face no such accountability. The argument to restrain referendum campaigners from making such statements in the first place is therefore stronger.

Regulating campaign content: some precedents

There are few precedents available to guide the design of a referendum truth commission. Much existing regulation of campaign content concerns elections and tends to be limited. In the UK it is an offence to make false statements of fact in relation to a candidate’s personal character or conduct, without believing them to be true nor having reasonable grounds for doing so. In Australia, it is an offence to mislead or deceive an elector as to the manner in which they cast a vote at an electionor referendum – for example, by providing false instructions on how to fill out a ballot paper. General laws on defamation will often apply to campaign material. Such regulation falls well short of entrusting an independent body to monitor the truthfulness of campaign statements.

The state of South Australia is a rare jurisdiction that has taken a more interventionist approach to regulating the accuracy of campaign content. Its regime provides only a partial guide for proponents of a referendum truth commission, in that it applies only to elections and to the publication of advertisements. Also, decisions about ‘truthfulness’ are ultimately left to the courts, rather than the Electoral Commission or some other independent authority.

Under section 113 of the Electoral Act 1985 (SA), it is an offence to authorise, cause or permit the publication of an election advertisement by any means (including on radio or television) that ‘contains a statement purporting to be a statement of fact that is inaccurate and misleading to a material extent’ (that is, to a substantial or significant extent). The state’s Electoral Commissioner may request that such an advertisement be withdrawn and/or that a retraction be issued, and can seek a court order to support that request. Individuals (including opposition campaigners) can also request a withdrawal and/or retraction, provided that they can demonstrate standing. In the event that a prosecution is launched against an advertiser, the maximum penalties that apply are $5,000 for individuals and $25,000 for corporations. It is a defence for an advertiser to establish that they played no part in determining the content of the advertisement in question and ‘could not reasonably be expected to have known that the statement…was inaccurate and misleading’.

Prosecutions under this provision have been rare. In one successful prosecution, a Labor Party advertisement was ruled misleading for stating, without basis, that the Liberal Party supported the closure of schools with fewer than 300 students. In another case, an independent candidate established that a Liberal Party advertisement had misled voters about the workings of the preferential voting system by claiming that a vote for him would ‘give you’ a Labor Premier.

New Zealand also regulates the making of misleading statements during election campaigns, but only for the two days preceding polling day. It is an offence within that period to make ‘a statement of fact that the person knows is false in a material particular’. Notably, both the South Australian and New Zealand laws apply only to statements of fact.

Designing a referendum truth commission: challenges

The dearth of precedents shows that the establishment of a referendum truth commission in the UK would be a novel initiative. It would also be one that is likely to raise concerns about free speech and encounter significant practical difficulties.

One of the main objections to empowering a body to monitor the truthfulness of campaign statements (in advertisements or otherwise) is that it would have a ‘chilling’ effect on campaign speech. The prospect of having the accuracy of one’s campaign utterances second-guessed, especially under threat of a fine, could cause campaigners to hold back from making legitimate claims. This would have the effect of reducing the amount of information available to voters.

A second issue is that many of the problematic claims made during referendum campaigns are promises and predictions, rather than statements of fact. The accuracy of such statements may be impossible to assess. For example, the claim that British families would be £4300 worse off if the UK left the EU is tendentious, but as a prediction of future wellbeing it is difficult to conclude that it is inaccurate and misleading. Statements that amount to exaggerations or emotional appeals (‘reform will put our security/stability/wellbeing at risk’; ‘Breaking Point: The EU has failed us all’) are also difficult to evaluate. Even if a truth commission could only monitor statements of fact (as in South Australia) the task of identifying such statements, and measuring their accuracy, would not always be straightforward.

Another important question concerns who would be responsible for assessing the veracity of campaign statements. A standalone ‘truth commission’ is one option. The effectiveness of such a body would depend, to a large degree, on its reputation for competence and independence. Some steps could be taken to cultivate this – for example, the appointment of commissioners with cross-party appeal and community support. But even this would not guarantee success. In the heat of a referendum campaign, the interventions of a standalone commission would likely become political flashpoints in themselves. Campaigners who have their statements publicly corrected may see political advantage in accusing the commission of bias and otherwise challenging its standing and authority.

As the Electoral Reform Society notes, an alternative would be to empower the Electoral Commission to intervene when misleading statements are made. But, as the Commission itself has acknowledged the impression that it was taking sides in political debates could undermine its independence. It would also divert resources away from its core business of running the referendum.

The courts could also do the job, as in South Australia. In some ways courts are well-placed to assess whether statements made in campaign material are true or false. But judicial processes move slowly and it would usually take months for a court to issue a final determination on a complaint. By that point, the statement that was alleged to be misleading would have had its effect on voters and the referendum itself would be over. Injunctions can help to address this concern with timeliness, but are open to abuse: they can be used as a means of unsettling an opponent’s advertising campaign, and to divert their resources into legal challenges.

A final question concerns the types of interventions that a truth commission (or similar body) should make. A minimalist approach would be to limit the commission’s involvement to the issuing of factual corrections. This would have the advantage of minimising any impact on free speech. In the lead up to the Irish same-sex marriage referendum, the Chairperson of the Referendum Commission made an intervention of this kind when he clarified that the outcome of the vote would not affect surrogacy and adoption. The force and effect of such statements will depend on the standing of the commission and its members.

The weakness of this minimalist model is that the threat of public correction, particularly in intense campaigns, may fail to act as a deterrent. The issuing of a factual clarification may even provide free publicity for inaccurate statements, ensuring that they receive far more airtime than they otherwise would have. This suggests that a more robust approach to remedies is warranted and should include the power to issue fines. But getting the balance right is tricky: small fines may fail to dissuade campaigners from stretching the truth, while large fines will imperil free speech and strengthen the argument for involving the courts.

The experience in South Australia also suggests caution. In the view of a former state Electoral Commissioner, its ‘truth’ laws have not changed behaviour. Campaigners have not been deterred from making misleading or inaccurate advertisements. And the law has been subject to misuse, with one party member ‘fir[ing] off three or four [complaints] a day’ to disrupt his opponents’ campaign.

What is to be done?

Ultimately, the establishment of a truth commission is not the way to address concerns about false and misleading statements made during referendum campaigns. But, as the Electoral Reform Society itself acknowledged in its post-Brexit report, there are other steps that can be taken to better prepare voters to make an informed choice at the ballot box. These include improving the design of official pamphlets (including a stronger focus on impartial information) and ensuring that voters are clear about the precise consequences of their vote (as they were for the AV referendum, when substantive legislation was passed in advance and made contingent on a majority Yes vote). More can also be done in advance of the campaign – for example, by involving citizens in the design of referendum proposals through citizens’ assemblies and other deliberative mechanisms. These initiatives would not prevent unscrupulous campaigning but they would better equip citizens to critically evaluate the claims of campaigners and to cast informed votes on polling day.

This post represents the views of the author and not those of Democratic Audit. It first appeared at the Constitution Unit blog.

Paul Kildea spoke at a Constitution Unit seminar on the regulation of the EU referendum on 25 October. You can read a summary of what was said here.