America is at war with itself over elections. The ongoing voting wars have an uncomfortable resemblance to the position in the Deep South before the passage of the federal Voting Rights Act in 1965. In those days, electoral officials — themselves predominantly white — administered biased rules designed to keep blacks off the voter rolls.
Ever since the US Supreme Court’s controversial decision in Bush v. Gore gave George W. Bush the presidency following a wafer-thin, dubious result in Florida in 2000, there has been a rash of new laws and a surge in electoral litigation. Republican state governments have passed laws to prevent voting fraud. But many Democrats claim these laws are designed to keep blacks and other disadvantaged or pro-Democrat groups from participating in elections. One common anti-voting fraud rule is the requirement by state governments that voters must supply a form of photo identification such as a driving licence when they vote.
These new regulations have been challenged in a rash of court cases in which the judges have split down party lines. When Indiana’s law requiring voters to show photographic identification at the polls came up for judicial review by the Supreme Court in 2008 in the case of Crawford v. Marion County Election Board, only Justice David Souter dissented from his fellow Republican appointees who declared themselves satisfied with the state’s photo ID requirements. Joined by two Democrat appointees, Justices Ruth Bader Ginsburg and Stephen Breyer, Souter claimed that the law worked to the disadvantage of “poor, old, and disabled voters” among whom a significant percentage were “likely to be deterred from voting”.
This conflict between a party of the Left, which has favoured lenient voting regulations, and a party of the Right, which has stressed the need for safeguards against voting fraud, has spread to other English-speaking democracies. In Canada, Stephen Harper’s Conservative administration has come under sustained criticism for enacting safeguards against voting fraud. Until 2014, Canada allowed both election day registration of electors and a system of “vouching”. This allowed an elector to declare that another elector (typically a spouse) who lacked the required identification was entitled to cast a ballot.
As shown by an official report, the system was hard to administer and led to frequent errors. However, opponents of the Fair Elections Act 2014 claimed that administrative problems at polling stations were to be distinguished from voting fraud. They accused the Conservatives of introducing regulations to make it harder for members of disadvantaged communities to vote. In particular, old people, inhabitants of aboriginal “First Nations” reservations, students and the homeless would be disadvantaged. The constitutionality of the 2014 law is currently under legal challenge and it almost certainly will be repealed by the newly elected Liberal Government.
The competing claims of rules designed to promote voter engagement and to deter voting fraud have also become a divisive issue in Australia. A notable campaigner, Dr Amy McGrath, has campaigned for many years to draw attention to systematic fraud, especially in trade union ballots. She has gained considerable support, especially since the discovery of no fewer than 7,743 cases of suspected voting fraud in Australia’s federal election of 2013. In fact, these were just the cases referred to the Australian Federal Police out of “18,770 voters who had multiple marks against their name on the certified list” (that is, people who had voted more than once). In state elections in 2011 in New South Wales, 3,843 multiple votes were cast.
It is typical of the political divisions that have been driving the “voting wars” in the US, Canada and Australia that it has been argued from the Left that the thousands of examples of apparent multiple voting are not a real problem. They reflect administrative error rather than deliberate electoral crime. Moreover, it cannot be shown that they affected the election results. In fact, multiple voting was no more common in marginal seats where there was the greatest motive for organised cheating. Therefore, as a notable report by the Sydney University academic Rodney Smith concluded, the side-effects of the cure — voter identification rules — would be far more damaging than the probably minuscule harm of the alleged problem.
This brings us to the United Kingdom and the current inquiry by Sir Eric Pickles MP into electoral fraud in British elections. So far, debates about electoral rules in the UK have lacked the toxic character of those in other English-speaking nations. All the main parties agreed on the introduction of postal voting on demand a decade ago and the recent change from household to individual voter registration. There is a danger that the situation is changing. For the sake of British democracy, all of the main parties as well as professional electoral administrators and political scientists need to be restrained and evidence-led in their recommendations.
There have been several indications of political bad temper and some ill-founded accusations. Chris Game, a commentator based at Birmingham University’s Institute of Local Government, gave a paper at a meeting of the influential Electoral Integrity Project held in San Francisco in September in which he described individual electoral administration as a “Conservative plot” despite the fact that it had been proposed by the Electoral Commission under the previous Labour government. Its real motive was to limit “the registration of various disadvantaged, marginalised and hard-to-reach groups, who, if they did register and vote, would be likely to support Labour”. Moreover, these registers “will be used to determine the sizes and boundaries of the parliamentary constituencies on which the 2020 General Election will be fought”.
This reflects the empirically weak argument of the Liberal Democrat peers Lord Tyler and Lord Rennard. While they do not oppose the introduction of individual electoral registration per se, they (along with the Electoral Commission) criticise the speed with which it is being implemented. They charge that it will remove two million voters from the register — predominantly from Labour-held urban areas — and, in the process, will skew the long overdue redrawing of parliamentary boundaries.
A further source of partisan division was the manner in which the Labour MP Graham Allen chaired the House of Commons Political and Constitutional Reform Committee in the last Parliament, when he strongly directed its investigations and reports into promoting “engagement” to the neglect of the issue of voting fraud.
In the UK, unlike some other developed democracies, there has been ample proof of voting fraud, especially in some urban local government elections. The decision of an electoral court last April to invalidate the results of a mayoral election in the London Borough of Tower Hamlets was the most dramatic but by no means the only proven example. It followed notorious cases in the 2004 municipal elections in the Birmingham wards of Aston (where a court found there had been at least 1,000 fraudulent votes) and Bordesley Green (1,500 to 2,000). Between 2004 and 2013, courts imposed jail sentences for electoral fraud in Ashford, Blackburn, Bradford, Bristol, Burnley, Coleraine, Coventry, Derby, Guildford, Oldham, Peterborough, Slough, Walsall, and Wolverhampton. In several of these places, it was Conservative candidates who were found guilty.
Since voting fraud is extremely hard to document and to prove to the standards required by the courts, and since — as shown by the failure of previous investigations by the Metropolitan Police in Tower Hamlets — few police forces possess the special skills required to identify it, electoral corruption is almost certainly more widespread than the list of convictions indicates. We do know that it mainly affects local elections in urban areas and that candidates of all of the main parties have been involved. Evidently, the perks of local office and the ability to provide jobs, contracts and other benefits to political cronies provide incentives.
Voting fraud, though now associated with immigrant communities, is a long-established British tradition celebrated in Victorian novels. In Northern Ireland, the slogan “vote early, vote often” was an indication of malpractice which led to the Labour government’s introduction in 2003 of a requirement that voters produce photo identification. Such identification, though controversial in the US and Canada, is the international norm. Before the 2015 general election, the UK Electoral Commission announced plans to require voters to produce a form of identification at the polls by 2019. Therefore, the Conservative proposal to implement this proposal seems wholly reasonable.
Nevertheless, the review being conducted by Sir Eric faces two major problems. First, as evidence from Northern Ireland shows, and as has been argued repeatedly in foreign court cases, the introduction of more stringent rules for administering elections does tend to lower voter registration and turnout. If the UK is to avoid the conflicts over voting rules which are plaguing the United States and Canada, it is important as far as possible to minimise conflict between the main parties over these rules. This is a hard task because party officials naturally advocate regulations which they calculate will work to their advantage. Certainly, the current government needs to dismiss bogus pleas such as those of Liberal Democrat peers to postpone yet again the equalisation of constituency electorates by redrawing the boundaries. At the same time, the Cameron administration would do well to take the high ground both by consulting fully with opposition parties and by considering measures to promote registration and political engagement among underprivileged sections of the electorate.
Difficult though this may prove, there is a second even larger problem with the way elections are conducted in our country. Voting fraud is facilitated by a combination of two things: an outdated and wholly inefficient system of voter registration and the system introduced in 2001 of postal voting on demand.
Richard Mawrey QC, the lawyer who has been the judge in most of the major election petitions of recent years, has made clear that the evidence presented in court indicates that voting fraud at the polls is risky and therefore practised on a minuscule scale. This is because it requires pretending that the person casting the ballot is someone else. By contrast, as he declared in his judgment in the Slough case in 2007, “Postal voting on demand . . . put the roll-stuffers in business in a big way.” Moreover, “Postal voting on demand, however many safeguards you build into it, is wide open to fraud. And that it’s open to fraud on a scale that will make election rigging a possibility — and indeed, in some areas, a probability.”
In spite of this, both Labour and the Conservatives are by now so wedded to the system of postal voting on demand that there seems little prospect that it will be abandoned. The most that can be hoped for is the introduction of partial safeguards.
Inaccurate voter rolls also facilitate fraud. By 2014, there were a staggering 15 million errors on the UK registers — some 6.5 million names were incorrectly included and 8.5 million were omitted. It is no coincidence that the other countries which have witnessed voting wars are also countries with highly deficient voting rolls. Why are democracies in Western Europe, in other respects so inferior in their democratic structures and traditions, so much more efficient in registering electors?
The simple answer is that they have consolidated civilian registers which are used for a variety of purposes. Most of these countries also have a system of identity cards. For the English-speaking democracies, this smacks of Big Brother. But in an era where supermarkets record the minutiae of shoppers’ habits, surely the time has arrived to put the case for a civilian population register onto the agenda of political debate in Britain. There are, of course, a number of partial measures to improve the efficiency of our electoral management. In the short run, these may have to suffice.