However, there is another organisation that could spend your money. Would you be willing to send the 100k to some social media ninjas who could usefully spend it on behalf of this organisation? I am very confident it would be well spent in the final crucial 5 days. Obviously it would be entirely legal.1
DOMINIC CUMMINGS, email to
Vote Leave donor, 11 June 2016
In late August 2019, Boris Johnson wrote a memo to the Westminster cabinet committee tasked with preparing for a no-deal Brexit. The freshly minted prime minister had pledged to leave the European Union “do or die”. Now he told ministers to “act immediately” to share all user data from their departmental websites. The government wanted to create a platform for gathering “targeted and personalised information”. Dominic Cummings, the prime minister’s chief advisor, emailed senior officials telling them that the data collection was a “TOP PRIORITY”.2
A few years earlier, a government plan to transfer masses of data would likely have gone largely unnoticed. A privacy campaigner might have offered a comment; a lowly opposition politician would have raised a question without expecting much of an answer. But the story of Johnson’s diktat hit the headlines. Labour deputy leader Tom Watson described the proposal as “very suspicious”. Others complained that the government was secretly planning to hoover up detailed data about its citizens.
There were reasons for this heightened sensitivity. We are all (slightly) more wary of how our personal information is used, especially by political campaigners. Cambridge Analytica shut down in 2018, following a scandal about the massive misuse of Facebook data from tens of millions of users. But there was another reason the story of the prime minister’s memo made waves. It had ‘Dominic Cummings’ and ‘data’ in the same headline.
Cummings has been the closest contemporary British politics has to a Machiavelli. Like the author of The Prince, he has a piercing stare and a prematurely receding hairline. The mastermind behind the unexpected vote for Brexit in 2016, Cummings presented himself as a British political strategist with an uncanny knack for tapping into voters’ deepest desires. He read military historians, was inspired by Silicon Valley technocrats and wrote voluminous blog posts on everything from the Apollo space programme to Otto von Bismarck. Where others obsessed about appealing to the news media, Cummings – despite his well-heeled Oxbridge background – talked about opposing established elites. When then prime minister David Cameron described him as a “career psychopath”, it only served to feed the Cummings mythology.
A former colleague said of Cummings: “He doesn’t really believe in government at all.”3 If he had faith in anything, it was data. During the EU referendum, Cummings ran what he proudly called “the first campaign in the UK to put almost all our money into digital communication”.4 Vote Leave bought an estimated 1.5 billion Facebook advertisements directed at seven million people. Many of these targeted ads spread misinformation, particularly about immigration and the financial benefits of leaving the European Union. The campaign was also helped by massive illegal overspending, which paid for millions of digital ads. Cummings credited this almost invisible social media blitz, mostly delivered in the final days before the vote, with securing victory in a tight referendum.
Cummings’s installation in Downing Street in July 2019 was followed by a sudden spike in digital activity by the new administration. Almost immediately, hundreds of targeted adverts promoting the new prime minister started appearing online. One Facebook ad paid for by the Conservatives trumpeted a BBC news story reporting a “£14 billion pound cash boost for schools”. But it was not true. The BBC headline had been doctored. The actual figure quoted in the BBC story was just £7.1 billion. The ad was eventually taken down amid a chorus of criticism but it had run for two weeks before anyone flagged it.
This dissembling was a taste of things to come. A few months later, the Conservatives ran the most dishonest general campaign in British political history. Independent fact-checkers accused the Tories of misleading the public, after the party’s official Twitter account was rebranded as a fact-checking website during a television debate. Dark money-funded pro-Johnson ads flooded social media. Conservative Party sources routinely lied to journalists. Many saw the spirit of Dominic Cummings behind the onslaught of disinformation.
Having vowed to “deliver Brexit” in government, Cummings was credited with instigating Boris Johnson’s controversial decision to suspend Parliament in August 2019 in an unsuccessful attempt to force through an accelerated departure from the EU. (The Supreme Court later ruled the prorogation unlawful.) Twenty-one Tory MPs who voted to block a no-deal Brexit had the whip removed. “Dominic Cummings has been hired by Boris to lay waste,” complained Winston Churchill’s grandson, Nicholas Soames. “He is doing the job he was asked to do.”5
Cummings was not the only Vote Leave alumnus that Johnson brought into the heart of his administration. Press officer Lee Cain became the prime minister’s head of communications. The Leave campaign’s youthful social media maven Chloe Westley took up a similar role at Number 10. So many former campaign staff were given jobs in the new administration that Guardian columnist Jonathan Freeland declared it “a Vote Leave government”. It didn’t seem to matter much that the campaign had in fact broken the law in 2016. Pumping hundreds of thousands of pounds more than the legal limit into millions of Facebook adverts targeted at undecided voters was treated as an historical footnote, if it was mentioned at all.
But precisely how Vote Leave broke the law and pushed the boundaries of digital campaigning matters a great deal. Not because it shows that the Brexit vote was fraudulent, but because it’s one of the clearest examples of why our electoral system is broken. It demonstrates how even relatively small sums of money can influence our politics – and how our system is still wide open to abuse, especially through digital means.
Vote Leave’s story is a parable about how modern campaigns, of all kinds, can bend and break laws drafted for a very different era, and how regulators have failed to get to grips with the rapidly changing realities of democratic consultation. To understand all this, we first need to understand how leaving the European Union went from a dream shared by a small band of largely libertarian Eurosceptics to a campaign that changed the face of Britain.
*
The 2016 vote to leave the European Union was a culmination that many Brexiters had spent their lives working towards. A political project long dismissed as the preserve of what then Conservative prime minister David Cameron called “fruitcakes” and “loonies” had won the day. More than anything else, the result was a vindication of the much derided official ‘out’ campaign, Vote Leave, and particularly its bosses Matthew Elliott and Dominic Cummings. Few had given them a chance.
Vote Leave was both the product of decades of Eurosceptic agitation and rhetoric and, not unusually in British politics, a conversation between a small group of confidants. On a warm summer’s day in 2012, Matthew Elliott met Daniel Hannan in a summer house belonging to a septuagenarian Eurosceptic merchant banker named Rodney Leach.6 For two decades, Hannan had been the nearest thing to an intellectual motor behind Conservative opposition to the EU. In that time he had progressed from the debating halls of Oxford to the chambers of the European Parliament, but his core beliefs about ‘restoring’ national sovereignty had remained unaltered. “Daniel wanted to destroy Brussels,” one former colleague told me.
Now the push for a referendum on Europe was gathering pace. Cameron would soon commit to holding a vote. But the nascent campaign for withdrawal needed a front man. Hannan wanted Elliott to take on the role. “I knew it had to be Matt,” he would later tell Sunday Times journalist Tim Shipman.7
In many ways, Elliott was an obvious choice to lead what became Vote Leave. He was young, had Westminster experience and was steeped in Euroscepticism. He was also a devoted free marketer. Growing up in a solidly middle-class family headed by a trade unionist father in 1980s northern England, Elliott picked up his politics outside the home.
“I remember at Leeds Grammar School there was an economics teacher called Terry Ellsworth, and he basically taught us A-Level economics using Milton Friedman’s Free to Choose books and video series. I think a lot of my free market economics came from that,” Elliott later recalled.8 He went on to the London School of Economics, where he became president of the Hayek Society, named after the Nobel Prize-winning Austrian economist and polemicist against government intervention.
From the beginning of his career, Elliott worked to apply North American libertarian methods to British politics. Shortly after graduating from university, he paid his first visit to Americans for Tax Reform, a conservative Washington DC “taxpayer advocacy group” set up by Grover Norquist in the 1980s at the behest of president Ronald Reagan. Its corporate funders have included the Koch brothers’ various foundations.
Elliott brought Norquist’s model to Britain.9 In 2004, aged just 25, he co-founded the TaxPayers’ Alliance to campaign in favour of tax cuts and privatisation. The TPA called for the television licence fee – which pays for the BBC – to be abolished and the National Health Service to be replaced by private provision.
Like its American counterparts, Elliott’s pressure group was adept at capturing media attention. There was almost nothing on which it would not offer a comment. In 2008 alone, the Daily Mail quoted the TaxPayers’ Alliance in a staggering 517 articles. The Sun did so 307 times.10 When the TPA published a denunciation of green taxes, a topless blonde model named Keeley on the Sun’s now-defunct Page 3 asked readers: “Why should Britain pay over the odds when our energy usage is lower than other countries?”11 Even the revelation that one of the TPA’s founders, Anthony Heath, did not actually pay any tax in the UK did little to dent the group’s media reach.12
Elliott grew the TPA into a £1 million-a-year operation employing over a dozen staff. The impressive growth was undergirded by dark money. Although the TPA campaigned for greater transparency in government, Elliott always refused to say who funded it. “I think people have a right to donate to charities and campaigns anonymously,” he said when pressed.13
Money did come from leading Conservative donors, including JCB tycoon Anthony Bamford, former Tory co-treasurer Peter Cruddas and the secretive Midlands Industrial Council, initially set up in 1946 to oppose Clement Attlee’s nationalisation programme. The TPA was part of a lattice of libertarian and climate-sceptic outlets that often operated in tandem, pushing similar policies and causes. Many are based at the same address, 55 Tufton Street, a four-storey Georgian townhouse on an elegant Westminster side street. The house is owned by Conservative businessman Richard Smith.14
Washington lobbyists frequently pass through Tufton Street and its environs. In September 2010, Elliott hosted a huge “free market roadshow”15 in central London featuring a raft of influential American libertarians: the Cato Institute, the Heritage Foundation, the Krieble Foundation and many more. Pride of place was reserved for the conservative movement sweeping across the United States, the aggressively anti-government Tea Party. A Tea Party-aligned consultant said that she had wanted to “identify groups in Europe” that “could start activist wings”16. At the time, Elliott said: “It will be fascinating to see whether it will transfer to the UK. Will there be the same sort of uprising?”17*
It was not just Elliott’s familiarity with transatlantic libertarianism that would have appealed to the ardent Atlanticist Daniel Hannan as he set about building a campaign to leave the EU. Elliott’s CV also contained a particularly rare qualification: he already had experience winning a British referendum.
In 2011, Britain held its first national referendum since the vote on joining the European project more than thirty-five years earlier. This time around, the proposal was more narrowly procedural. Should Westminster switch from the first-past-the-post electoral system to the slightly more proportional alternative vote?
The proposed alteration was a fudge. The Liberal Democrats, who wanted full proportional representation, had agreed to a referendum on a less democratic electoral measure as part of their coalition agreement with David Cameron’s Conservatives.
Matthew Elliott led the campaign against the measure. In a foreshadowing of Vote Leave five years later, NO to AV quickly framed the terms of the debate with dubious figures and kept to a small number of simple messages. Elliott claimed that the alternative vote would cost £250 million to introduce. The claim was refuted by the Electoral Commission and the non-aligned Political Studies Association. It didn’t matter. Billboards ran photographs of a sick child in an incubator alongside the slogan “She needs a new cardiac facility, not an alternative voting system.” ‘Yes’ lost by a crushing two to one on polling day.18
NO to AV also provided a trial run for the Brexit referendum in other ways. Elliott focused heavily on digital campaigning. He pushed messages on Facebook and built online applications that encouraged voters to attend real-world campaign events. It was cutting-edge for the time.
“Matthew was one of those people who saw early that we needed to put digital at the heart of the campaign. He had seen what was happening in the US,” says Jag Singh, an American digital expert who worked on NO to AV and later set up a digital political consultancy with Elliott and Paul Staines from the right-wing website Guido Fawkes.
Elliott did not immediately establish Vote Leave after his meeting with Hannan. Instead, in 2013 he set up a prototype campaign called Business for Britain, again based at 55 Tufton Street. By then Cameron had pledged to renegotiate Britain’s relationship with Brussels. The aim of Business for Britain was to subtly force Cameron’s hand. “I realised that business was the way into it,” Elliott later said. “We did not do it as a hard Brexit campaign but went along the lines of the renegotiation, albeit pushing further what the PM would be thinking.”19
In May 2015, the EU referendum went from a remote possibility to imminent reality. After the Conservatives won a surprising majority, Cameron found himself having to call a referendum he had never expected or wanted. During the coalition years, Eurosceptics were often left on the backbenches as Liberal Democrats took up payroll posts in government. The attitude of Conservatives hostile to Brussels had hardened from a wariness about federalism into a full-throated demand to leave the EU itself. Business for Britain morphed into Vote Leave.
Then Elliott made his boldest move: he hired Dominic Cummings, once a pugilistic special advisor to education secretary Michael Gove, to run the Leave campaign. It was a decision that would come to define Vote Leave and even Brexit itself.
In person, Cummings and Elliott are very different. Elliott is often circumspect and serious. He prefers to operate behind the scenes. Cummings is truculent, controversial and wildly divisive. As Tim Shipman writes, Cummings is “viewed by his allies as one of the most talented public policy professionals of his generation”. To his enemies, he is “a raging menace, a Tory bastard love-child of Damian McBride and Alastair Campbell”.20 One-time Vote Leave board member John Mills remembers Elliott as “a very nice person to work with”; Cummings, by contrast, was “difficult to manage” and “opinionated”.
A young member of the Brexit campaign team, who spoke on condition of anonymity, had a rather different experience. “Matthew Elliott barely spoke to junior people, just the board and the funders. Dominic Cummings was hands-on. He was the one who was there for you.”
Politically, however, the two men had much in common. Like Elliott, Cummings’s background was solidly Eurosceptic. At Oxford, he studied under right-wing Scottish historian Norman Stone,21 a leading member of the Thatcherite Bruges Group. After a failed attempt to set up an airline in post-Soviet Russia, Cummings had taken over Business for Sterling, an anti-euro lobbying campaign also set up by Rodney Leach. (The same small groups of people recur in a bewildering number of different combinations in recent British politics.)
Like Elliott, Cummings also had experience with British referendums. In November 2004, he had led the campaign against setting up a North-East Regional Assembly, intended as the first step towards English devolution. Cummings had other ideas. He came up with a catchy slogan to capture the key local objection: “Politicians talk, we pay.”22 There were negative billboards and television adverts, as well as photogenic stunts. To illustrate how much the assembly would cost taxpayers, campaign chairman John Elliott – no relation to Matthew – delivered a short talk on local government finance while standing next to a bonfire onto which blonde girls from Durham University wearing tight t-shirts threw bundles of fake £50 notes. Cummings brought an inflatable white elephant on a tour of the north-east. As he would in 2016, he said that the money saved on another layer of government could be used to hire additional doctors, teachers or police.23
In another echo of the Brexit referendum, the campaign was also dogged by arguments with the Electoral Commission and fellow campaigners against the assembly over designations and spending. In his insider account of the North-East referendum, Cummings’s “ground war commander” William Norton writes that the regulator was “utterly useless”. “It had no power to act as a referee, only as an auditor,” he notes. Norton later became Vote Leave’s legal director.
In the end, the disputes mattered little. Those hostile to a new North East Assembly won a resounding 78 per cent of the vote. English devolution was abandoned. The No campaign’s chief spokesperson, Graham Robb, described the operation as “Britain’s first populist campaign”. Cummings would later say that it had been “a training exercise for the EU referendum”.
Along with campaign director James Frayne, Cummings set up a free market think tank called the New Frontiers Foundation. The name was taken from a celebrated 1960 speech by John F. Kennedy. The intention was to create a UK equivalent of the Heritage Foundation, the behemoth of the American conservative think tank world that had set the policy agenda for numerous presidents.24 In late 2004, the Financial Times reported that “the New Frontiers Foundation, one of the most active Eurosceptic groups in Britain, has been energetically cultivating transatlantic ties in the US”.25
But Cummings’s new venture quickly ran into the sand. By the following year, the New Frontiers Foundation was out of money. Frayne went to work for the TaxPayers’ Alliance. Cummings joined Michael Gove as an advisor, where he stayed for seven years before quitting in 2014, only coming out of self-imposed political exile to run Vote Leave.
There is a temptation to retrospectively view a successful political campaign as flawless. With hindsight, every move becomes the right one. This was certainly not the case with Vote Leave. There were internal difficulties, including a failed attempt by the board to sack Cummings. Elliott’s fledgling campaign spent months fighting an often vicious internecine conflict with Arron Banks’s Leave.EU operation over which should be classified as the lead campaign. The designation mattered – the lead campaign would have privileged access to media and could spend up to £7 million. These spending limits would become crucial during the referendum, and after.
The Electoral Commission eventually chose Vote Leave as the official pro-Brexit campaign. Vote Leave also had another major boost: Boris Johnson. On Friday, 19 February, the former London mayor and Conservative MP wrote two columns in the Daily Telegraph about the forthcoming referendum. (Until he became prime minister, the Telegraph paid him £275,000 a year to write a weekly column.) He sent both pieces to his then wife Marina Wheeler. The first was a jeremiad against EU law that opened with a curious section about buses and ended with a rallying cry for a Leave vote. The second was a rather half-hearted endorsement of Cameron’s insubstantial renegotiations with Brussels.
The draft column is most memorable for outlining Johnson’s policy on cake (“pro-having it and pro-eating it”) and Britain’s relationship with Europe (“It is surely a boon for the world and for Europe that she should be so intimately engaged in the EU”26). In the end, Johnson filed the original column, backing Brexit.
Suddenly, the referendum seemed winnable. By then, Vote Leave had moved out of 55 Tufton Street to larger offices in the Westminster Tower beside Lambeth Bridge, right across the river from the Houses of Parliament. It was the same site used by Matthew Elliott’s campaign against electoral reform in 2011.
Donations, initially slow, started to flow in. Much of the money came from major Tory donors who had also contributed to the anti-AV campaign. Jeremy Hosking, co-founder of the Marathon hedge fund, gave almost £1.7 million to pro-Brexit campaigns. Former Conservative Party treasurers Peter Cruddas and Michael Farmer contributed, as did Anthony Bamford. After Cameron’s government sent a pro-Remain leaflet to every house in the country, the Midlands Industrial Council pledged to give up to £5 million to the Leave campaign.27 As referendum day approached, Vote Leave had more offers of money than they could legally spend.
*
In January 2016, a slightly built, bespectacled young Brighton University student logged on to Facebook and set up a new account. Darren Grimes wanted to start a campaign to encourage young people to vote to leave the EU. It was called BeLeave – a knowing wink to the ‘Beliebers’ who flock to Canadian pop star Justin Bieber’s concerts.
Grimes was an unlikely BeLeaver. He had been a Liberal Democrat activist. When Charles Kennedy died in June 2015, Grimes paid tribute to the former party leader’s belief that “having Britain in Europe was the only way forward”.28 But six months later he had left the Lib Dems to join the Conservatives. In a blog post for a Tory website associated with Daniel Hannan and Eurosceptic Conservative MP Steve Baker, Grimes said the Lib Dems had lost “that classical liberalism that first attracted me to politics”.29
Although BeLeave struggled to get much support online – the market for young, liberal Leave voters was narrow – its front man was making a name for himself. Grimes had a knack for crafting simple, stylish videos and graphics. He was also often invited onto broadcast media hungry for youthful pro-Brexit voices. Originally from county Durham, Grimes spoke articulately about working-class attitudes to the EU and their wider concerns.
BeLeave was also getting noticed at Vote Leave’s London HQ, and not just for Grimes’s burgeoning media profile. BeLeave’s online campaign was cheap. The group could spend up to £700,000 during the campaign but in four months had spent just £21. BeLeave didn’t even have a bank account. With Vote Leave starting to approach its spending limit in the weeks leading up to polling day, the larger campaign had an idea. It would transfer excess funds to its far smaller cousin. “When it became obvious that we were in surplus funding, we chose to donate to other campaigns,” Vote Leave finance director Antonia Flockton would later tell a parliamentary public administration committee.30 “There is a question as to whether they were independent campaigns acting independently. They were. Therefore, there is no issue in relation to our expenditure. The rules, to our minds, are quite clear.”
But there was a problem with Vote Leave’s plan. In spite of Flockton’s confidence, it was against the law.
*
British election law is fiendishly complex. Around 50 Acts and 170 statutory instruments govern electoral contests. “Almost no one knows how it all works,” Gavin Millar, a QC and expert in electoral law, told me when we met in his London legal chambers. A keen art lover, Millar compared the labyrinthine legislation to a work by Heath Robinson, the English illustrator best known for his whimsical drawings of elaborate machines constructed to achieve simple objectives. “Election law is like that. There are bits hanging off it everywhere.”
One core tenet of this legislative maze is reasonably straightforward: spending should be capped to ensure a level playing field. This principle dates back to the 1880s, when Parliament first legislated along these lines to try and prevent tyrannical landlords from buying up seats. Nowadays, the central plank of British electoral law is something called the Political Parties, Elections and Referendums Act 2000 (PPERA is its inelegant acronym).
The Act was introduced in the early years of the Blair government, spurred by press stories through the 1980s and 1990s about tax exiles funding the Conservatives. For the first time, political parties had to declare their donors, and overseas donations were banned. PPERA also established a regulator, the Electoral Commission, with the courts as a last resort to deal with any disputes. The regulator oversees general elections and referendums; for the former, the UK is split up into 650 constituencies each with its own spending constraints; for referendums the whole country is treated as a single constituency with no limit on the amount of spending until the ten weeks before polling day. In the EU referendum’s ‘regulated period’, each of the two designated campaigns, for and against, could spend up to £7 million. Other registered campaigners could spend up to £700,000. In an effort to prevent campaigners breaking spending limits by giving money to front groups, any coordinated campaigning must be declared as ‘joint working’ and their spending must be counted together.
The idea of using third-party groups to bypass spending rules was not new. Matthew Elliott had once suggested setting up a slew of different organisations that would be “paper tigers” for the main Leave campaign. “He viewed everything as a front campaign,” UKIP campaigner Chris Bruni-Lowe told the journalist Tim Shipman.31 Elliott even proposed some potential names for these confected groups: Bikers for Britain, Women for Britain, Muslims for Britain. (Three days before the referendum, Vote Leave did actually donate £10,000 to a group called Muslims for Britain.)
In February 2016, Vote Leave coordinator and future Brexit Minister Steve Baker suggested in an email that the campaign could “spend as much money as is necessary to win the referendum” by creating “separate legal entities”.32 A former Labour minister reported Baker’s statement to the police. Vote Leave issued a press release saying: “We never have and never will encourage people to break the law.”
In the lead-up to the referendum, Vote Leave ran a raft of pro-Brexit groups to appeal to less traditionally Eurosceptic voters. They all had catchy names. Out & Proud. Green Leaves. Mark Gettleson, a former Cambridge Analytica consultant, designed the concepts for Vote Leave’s faux grassroots groups. Gettleson and Darren Grimes had worked together before, on Norman Lamb’s unsuccessful bid for the Liberal Democrat leadership in 2015 (along with future Cambridge Analytica whistleblower Chris Wylie). Gettleson’s ginger groups would all fade into history, except for one: BeLeave.
In the crucial final weeks before 23 June 2016, BeLeave became increasingly important to Vote Leave’s strategy to circumvent spending limits. In less than two weeks, the largest pro-Brexit campaign made a series of huge donations to Grimes’s small youth-focused initiative. In all, BeLeave received an astonishing £675,000 in the weeks before the Brexit vote, mostly from Vote Leave. By the standards of the US these are tiny sums, but in Britain, where the average cap on constituency spending in general elections is around £15,000, it was a very significant donation. And even more importantly, none of the money actually went to BeLeave – instead it was all paid directly to AggregateIQ, the same obscure Canadian data analytics firm that Vote Leave used. This was never declared to the Electoral Commission as ‘joint working’, which is where unrelated organisations campaign together.
“Vote Leave didn’t really give us that money,” former BeLeave treasurer Shahmir Sanni later told Guardian journalist Carole Cadwalladr. “They just pretended to. We had no control over it. We were 22-year-old students. You’re not going to just give nearly a million pounds to a pair of students and let them do whatever.”33
As well as spending £625,000 with AggregateIQ on BeLeave’s behalf, Dominic Cummings also brokered other donations for the youth campaign. In early June 2016, Cummings emailed a potential donor, hedge fund millionaire Anthony Clake, saying that Vote Leave had all the money it could use but “there is another organisation that could spend your money”. Cummings suggested that Clake “send the 100k to some social media ninjas” in Canada. “Obviously it would be entirely legal,” he added.34
On 15 June 2016, Clake wrote to Cummings, saying that he would split a donation between Vote Leave and BeLeave. Cummings replied the same day, saying, “We are also giving money to them – you can just send us the full amount and we’ll add yours onto what we are giving them and save you the admin.”35 Clake later said that he wanted to give the money to Vote Leave but was discouraged from doing so by the official Leave campaign, as “they were close to their spending limits”.36 Cummings said, in a tweet subsequently deleted, that the Electoral Commission had given Vote Leave the go-ahead to give excess funds to other campaigners. The regulator strenuously denied that assertion.
After the referendum, the suggestion that Vote Leave and BeLeave had been working together hung in the air. There were press stories and calls for the Electoral Commission to step in. Nevertheless, the authorities were slow to investigate. One of the main sparks for action was a series of emails published by openDemocracy that shone a light on the internal workings of the Commission.37 Staff thought the BeLeave donations “unusual” but that there were no grounds to suspect coordination between the campaigns.
In response, the barrister Jolyon Maugham threatened to pursue a judicial review, writing that BeLeave was in fact a puppet entity being used to breach spending rules. In November 2017, the Electoral Commission opened a formal investigation. There were “reasonable grounds”, the regulator said, “to suspect an offence may have been committed”.38
The investigation was slow, partly owing to Vote Leave’s intransigence. Matthew Elliott later told the BBC that the Electoral Commission had refused to engage with his campaign, but the official report paints a very different picture: of Vote Leave repeatedly failing to cooperate. The regulator suggested dates for interviews; Vote Leave never responded. The campaign did not disclose documents. When they finally offered to give up the paperwork, it was only on the condition that they meet the regulator to discuss closing the investigation. The Electoral Commission sent authorised officers to inspect and take copies of these documents. “In the event, these documents were incomplete and some were not the correct documents,” the regulator reported. Vote Leave sent legal letters to the Electoral Commission, threatening to judicially review the decision to investigate them. Proceedings were never initiated.
In a submission to the investigation, Cummings said that he “had zero to do with BeLeave’s marketing, including what they did with AIQ”. When I emailed Cummings requesting an interview for this book, he sent back a one-line response: “I’ve no interest in Remain campaigns.” He never replied to my follow-up questions.
There were other allegations against Vote Leave. In early 2018, their former treasurer Shahmir Sanni claimed that Vote Leave had “tried to delete” key evidence after the Electoral Commission investigation had begun.39 The whistleblower also said that he had been introduced to BeLeave by Stephen Parkinson, one of Vote Leave’s best-paid employees and another veteran of the NO to AV campaign. Parkinson told the New York Times that he had been offering personal advice in the context of a romantic relationship with Sanni. Sanni said Parkinson’s statement had forced him to come out to his family and put relatives in Pakistan, where homosexuality is illegal and gay people are persecuted, in potential danger. Parkinson was kept on as an aide to prime minister Theresa May. (In 2019, May appointed him to the House of Lords.)
After Sanni spoke to the press, alleging that there had been coordination between BeLeave and Vote Leave, he was dismissed by the TaxPayers’ Alliance, where he worked on social media. Matthew Elliott’s pressure group later accepted that Sanni had been unlawfully dismissed and that there had been an active campaign to discredit his allegations of electoral wrongdoing.40
In July 2018, the Electoral Commission investigation reported that there was “significant evidence” of joint working between Vote Leave and BeLeave. The largest Leave campaign had broken the law. Gavin Millar QC described the offences as being of “a scale and seriousness” with no parallel in modern British politics. But the penalties were minuscule. Vote Leave was fined a total of £61,000, barely a tenth of the amount by which it had overspent during the referendum. David Alan Halsall, a Conservative donor who was legally responsible for Vote Leave, was reported to the Metropolitan Police. That was the extent of the punishment. No one else in Vote Leave was sanctioned.
Running a campaign that broke the law was not an impediment to career progression. Matthew Elliott went on to a succession of prominent roles: at the Legatum think tank, as campaign manager for Sajid Javid’s ill-fated Conservative leadership bid, as advisor to a hedge fund. Cummings joined Johnson’s government. Darren Grimes became a paid-up culture warrior, first at Elliott’s partisan news site Brexit Central then with the libertarians at the Institute of Economic Affairs. Boris Johnson – who had dismissed claims of Vote Leave’s law-breaking as “utterly ludicrous” – became prime minister, with Michael Gove by his side.
One former Labour MP says that senior British government figures were aware of Vote Leave’s overspending long before the Electoral Commission reported. “Johnson, Gove and Cummings all knew what was going on,” Ian Lucas told me when we met a few days before the 2019 election general, in a bar near Westminster fitted out in shiny Christmas decorations. Lucas had been an MP for Wrexham for almost twenty years and was a member of the Department of Culture, Media and Sport select committee’s inquiry into ‘fake news’ which had examined the Brexit referendum.
“I voted against the election because I thought we needed regulation before the election. It’s a charade,” he said. With a short, unkempt beard and a natty brown jumper, Lucas looked like a man pleased to be leaving the glare of frontline politics. Four days later, Labour lost Wrexham for the first time since 1935, one of dozens of so-called ‘red wall’ seats to fall in the Conservative surge.
Sipping a mineral water among a smattering of early afternoon drinkers, Lucas complained about the failure to hold the prime minister to account for Vote Leave’s law-breaking. “There are still huge questions to be asked about what Vote Leave did, but they’re not being asked. No journalist seems to ask Boris Johnson what he knew.”
This lack of interest is revealing. Unlike in the US, breaking British election law is often treated as a bagatelle. The maximum fine is just £20,000. Louise Edwards, the Electoral Commission’s head of regulation, told me that the fines she can levy are “scandalously low”. For British political parties and campaigners, paying a few thousand pounds for breaking the rules has become the cost of doing business. “We are seeing people who don’t want to work within the spirit of the law anymore,” says Gavin Millar QC. “Once that consensus, that you won’t break the rules, goes, you’re stuffed.”
Vote Leave was not the only campaign fined by the elections watchdog following the Brexit referendum. Arron Banks’s Leave.EU was sanctioned, as were the Liberal Democrats and the official Remain campaign, for failing to provide receipts and invoices. A small anti-Brexit campaign was fined £1,800 for not declaring more than £50,000 of joint working. Campaigners complained that election reporting processes are overly complex and difficult to follow. “It is very easy to get it wrong and there is very little help,” says John Mills, whose Labour Leave was fined £9,000 for failing to declare £20,000 worth of in-kind donations. Mills said that his sanction was “totally unreasonable given what Vote Leave was fined for”.
A number of the Electoral Commission’s decisions ended up in the courts. A judge overturned the decision to fine Darren Grimes £20,000 for allegedly making a false declaration. Grimes, whose case was supported by Vote Leave, said that he was “relieved” at the verdict and that the sanction “was based on an incorrectly ticked box on an application form”.41 Vote Leave repeatedly denied that it had broken the law, but in early 2019 the campaign quietly dropped its appeal and paid the fine, as well as the Commission’s £200,000 legal costs.42
The Metropolitan Police investigation into Vote Leave’s breaches of electoral law was still going on as this book was being written. In November 2019, the police passed a file on Vote Leave and BeLeave to the Crown Prosecution Service “for early investigative advice”.43 “The time the police have taken has been absolutely ludicrous,” Ian Lucas told me. “All these years on, we still don’t know what happened.” In an unguarded moment, a Scotland Yard spokesman admitted to an openDemocracy reporter that “political sensitivities” contributed to the glacial pace of the force’s inquiries.44
Electoral legal expert Gavin Millar said “police are notoriously reluctant to get involved in this sort of thing. Understandably so.” One option, he says, is to give Britain’s electoral regulator the power to pursue criminal investigations. “Unless you put people in the dock and threaten them with prison, why will people play by the rules?”
*
In early 2019, Brexit: The Uncivil War, a feature film about the EU referendum, aired on British television. Cummings, played by Benedict Cumberbatch with a prosthetically enhanced receding hairline, is very clearly the protagonist. The drama closes with Cummings giving evidence at a future government inquiry into Brexit. The scene is fictional but most of the action that follows is largely factual (with some poetic licence).
In one of the most memorable moments, Cummings is told that there are three million people that do not traditionally vote who could be persuaded to support Brexit, if Vote Leave can reach them. Instead of mail drops and billboard adverts, he decides to target the “missing three million” with ads on social media. “We are going to be making decisions based on science and data,” Cumberbatch-as-Cummings tells the camera. “No advertisers, no snake oil salesmen, or fucking Saatchis. We’re gonna follow algorithmic, statistical analysis.”
In one of his many lengthy post-referendum blog posts, Cummings wrote that Vote Leave “spent 98% of its marketing budget” on Facebook in the final days of the EU referendum. This claim is almost impossible to verify due to the paucity of regulations governing digital politics in Britain, but Cummings certainly put a lot of money and effort into online campaigning. Vote Leave deluged around seven million swing voters with an unprecedented 1.5 billion pro-Brexit videos and messages on Facebook, mainly in the last three or four days of the campaign. “Adverts are more effective the closer to the decision moment they hit the brain,” Cummings later wrote.45 Internally, Vote Leave staff nicknamed the strategy ‘Waterloo’.
Vote Leave and its sister campaigns spent more than £3.6 million on Facebook adverts. Cummings’s decision to focus on Facebook was adroit. Around six in ten British voters use the social network. The platform is particularly good for reaching those that don’t care much about politics. What was surprising, however, is how Vote Leave spent money on Facebook. All of the campaign’s spending went through AggregateIQ, a company that, in early 2016, had no web presence and was based above an optician’s in a shopping centre in the small city of Victoria, British Columbia. Cummings said he found AIQ “on the internet” before the Brexit referendum. “It was strange. Nobody in the UK had heard of AIQ. They seemed to come out of nowhere,” says Sam Jeffers, an expert on digital campaigning in British politics.
Vote Leave was relatively late to the digital party. Both Arron Banks’s Leave.EU and the Remain campaign already had access to large databases of supporters by the time AIQ began running social media ads remotely from Canada. AIQ staff soon moved into the campaign’s London offices. They were even given their own room. “They were really the ones in charge of the online advertising,” a Vote Leave insider told me.
AggregateIQ set about building an audience for Vote Leave’s message. Co-founder Zack Massingham started by identifying people who expressed an interest in Eurosceptic causes online. This demographic was mainly middle-aged and older, less well-off and non-university-educated.46 Once he had identified a core of potential voters, Massingham could use Facebook’s tools to build much bigger ‘look-alike’ audiences to target adverts. Potential supporters were sent an increasing volume of online adverts. Messages were rigorously tested, and adverts that failed to get a response rate of at least 30 per cent were junked.
The size of this organic audience – well into the hundreds of thousands – allowed Cummings’s campaign to drill down into highly specific sub-samples of voters. Then, in turn, the campaign could advertise directly to those groups of ‘persuadables’. These key cohorts would be sent different messages. “We could say, for example, we will target women between 35 and 45 who live in these particular geographical entities, who don’t have a degree,” Cummings later explained.
He boasted of using physicists and experts in “quantum information” to crunch voter data. Vote Leave recorded spending over £70,000 with a firm called Advanced Skills Initiative.47 The company is better known as ASI Data Science, a tech start-up that marketed itself as a world leader in artificial intelligence, and which employed a number of data scientists that worked for Cambridge Analytica.
Cummings also came up with clever ruses to find data on the “missing three million”. He ran an online competition during the 2016 European Football Championship. Win £50 million, the advertisement proclaimed, by successfully predicting the winner of all 51 games. Entry was free; all you had to do was give your personal details. Nobody won – the odds were calculated at an eye-watering 5,000 trillion to one – but more than 120,000 participated.48
Cummings said the point of the competition was to gather “data from people who usually ignore politics”.49 Harvesting such personal information was not against the law, but Vote Leave was later fined by the data regulator for sending hundreds of thousands of unsolicited text messages to contestants. The texts contained a link to a website operated by Vote Leave.50
Leave’s digital operation was not completely sui generis. Remain spent heavily on social media ads, too. It also had access to huge amounts of voter data from political parties. A former staff member of the Stronger In campaign told me that Vote Leave’s big advantage was that it had much freer rein to craft messages without the involvement of political parties. There were no Conservative, Labour and Lib Dem MPs squabbling over what the campaign should and shouldn’t say. Cummings used this space to run the biggest targeted digital misinformation campaign ever seen in Britain.
Vote Leave was often seen as the ‘nice’ face of an often vicious Brexit referendum campaign. Daniel Hannan preferred to talk about sovereignty and Brussels bureaucracy, rather than immigration. UKIP’s only MP, Douglas Carswell, joined Vote Leave ostensibly because it offered a respectable alternative to Nigel Farage and Arron Banks.
But many voters looking at their Facebook news feeds in the final days before the vote saw a very, very different face of the campaign. In the final 72 hours before polls opened, Vote Leave placed millions of ads that spread wildly fantastic narratives about immigration, particularly from Muslim states. One Facebook advert claimed: “Turkey’s 76 million people are being granted visa-free travel by the EU.” The accompanying graphic showed a map of Turkey with Dad’s Army-style arrows heading towards Britain. Other variations on the same advert featured maps of Syria and Iraq.
Vote Leave’s ads were textbook examples of misinformation. They took a kernel of truth – Turkey had applied to join the EU – and distorted it by removing any actual political and historical context. By 2016, Turkey’s bid for EU membership, first made as long ago as 1987, had completely stalled. Prime Minister Recep Tayyip Erdogan was increasingly seen as an autocrat. He had suppressed protests in Istanbul’s Gezi Park with tear gas and water cannons, killing 22 people. In Brussels, Turkey’s accession was firmly off the agenda.
But in Vote Leave’s offices, “every week was Turkey week”, according to a former employee. Head of media Rob Oxley bought yellow rubber turkey masks for his staff. Politicians and campaign surrogates were encouraged to focus on Turkey as much as possible. A few weeks before the vote, Conservative minister Penny Mordaunt told the BBC that Britain could not veto Turkey’s membership of the EU. This was not true.
Cummings knew the political value of dividing voters by playing on their fears around migrants coming to Britain. “Fear is such a powerful driving emotion, especially online,” says Claire Wardle, an expert in digital disinformation. Unlike in traditional campaigns, Vote Leave’s emotive images were not seen on posters or buses. They appeared privately, on people’s social media feeds, as I discovered in Sunderland. They were part of an invisible conversation conducted away from the media’s prying eyes. “You put all the negativity on the Internet and you put it at arm’s length from where journalists can see,” explained Sam Jeffers.
The only reason Vote Leave’s digital adverts did not disappear completely into the digital ether was Facebook’s decision to release them – with the campaign’s approval – to a Westminster parliamentary inquiry in 2018, two years after the referendum.51 Some of the 1,433 ads published had no imprint, a breach of electoral law if they had appeared in a leaflet or a newspaper. But the requirement that voters be told who paid for political adverts does not extend to the digital sphere, even though this is where campaigns now spend huge chunks of their budgets.52
As well as immigration, Vote Leave made great play with the now notorious claim that Britain sent £350 million to Brussels every week, and that if this money was repatriated it could be spent on the National Health Service. The misleading figure was debunked almost as soon as it had been unveiled on the side of a red Vote Leave-branded bus. The UK paid around £250 million each week into the EU budget, as a consequence of the rebate negotiated by Margaret Thatcher in the 1980s. The chair of the independent UK Statistics Authority described Vote Leave’s claim as “a clear misuse of official statistics” in a letter to Boris Johnson. Nevertheless, Cummings continued to push this message through targeted Facebook ads.
Cummings believed this was a crucial factor in winning a tight referendum. “If Boris, Gove and [Labour MP] Gisela [Stuart] had not supported us and picked up the baseball bat marked ‘Turkey/NHS/£350 million’ with five weeks to go, then 650,000 votes might have been lost,” he wrote.53
That is not to say that a Facebook ad or a shonky figure on the side of a campaign bus is the reason the UK voted to leave the EU. Remain had countless advantages. Indeed, if anything the pro-EU campaign was exactly what it was accused of being: an establishment stitch-up. Cameron spent more than £9 million of taxpayer money sending leaflets to every household in the country, urging a vote to stay in the EU. Stronger In peddled its own line in dubious claims. Brexit would cost families £4,300 a year. Britain would go into recession overnight. Remain had access to the machinery of government and a trove of party data.
Unlike in the case of pro-Brexit campaigners, however, these attempts to load the dice in Remain’s favour were clumsy and inefficient, analogue weapons in a digital war. The government-funded pamphlets motivated Vote Leave donors more than ordinary voters. ‘Project Fear’, as it was dubbed by Eurosceptic hardliners, repelled more than it attracted.
Leave’s illegitimate overspending and misinformation, on the other hand, was targeted and effective. Here, in this new world of political campaigning, even small sums of money can go a long, long way, while multimillion-pound government public relations exercises can prove practically worthless.
Cummings led Vote Leave unilaterally. He ignored members of his own board who wanted to push sunny visions of a buccaneering post-Brexit Britain, dismissing what he called “‘go global’ trade babble”. Instead, he kept buying huge numbers of Facebook adverts that played on voters’ fears with simple, compelling messages. In the whirr of the referendum, this digital offensive hardly registered. By the time recalcitrant regulators started to ask what was happening, Brexit had won. The victorious campaigners had already packed up and gone home.
AggregateIQ largely disappeared too – until, in the spring of 2018, the company became embroiled in one of the most controversial political scandals of recent times: the operations of Cambridge Analytica.
As Carole Cadwalladr revealed, Cambridge Analytica had illegally harvested millions of US Facebook users’ data for political advertising ahead of the 2016 US presidential campaign. The London-based company had also been involved in election ‘black ops’ around the world. Reports noted how closely connected AIQ was to Strategic Communications Laboratories, Cambridge Analytica’s parent company. AIQ had the same phone number as an SCL subsidiary and had worked with Cambridge Analytica on Republican Ted Cruz’s unsuccessful run for the US presidency in 2016.54
Amid a barrage of negative publicity, Facebook decided to suspend AIQ as it may “have improperly received FB user data”. Vote Leave’s biggest digital supplier had been banned by the platform on which it had pushed over a billion ads at British voters.
With Cambridge Analytica accused of everything from hiring prostitutes to bribing opposition politicians to coordinating voter suppression campaigns, the question of who the people behind AIQ were – and how Dominic Cummings found a firm so small that its Twitter followers in early 2016 would have fitted into the back of a London black taxi – suddenly became a matter of major interest. The British information commissioner, Elizabeth Denham, announced she was launching an investigation into AIQ.
Certainly, the exact relationship between AIQ and Cambridge Analytica was opaque. Chris Wylie, the sharp-suited, Brexit-supporting data scientist turned pink-haired Cambridge Analytica whistleblower, claimed that AIQ was essentially SCL’s Canadian office.55 Documents later provided by Wylie’s friend and sometime business partner Mark Gettleson to a parliamentary inquiry suggested that it was Gettleson, a former Cambridge Analytica employee, who sourced AIQ for Vote Leave.56 AIQ co-founder Zack Massingham maintained that his company “has never been and is not a part of Cambridge Analytica or SCL”57.
In November 2019, an investigation by Canada’s privacy commissioner found that AIQ had broken laws in its work for Vote Leave. AIQ, the Canadian watchdog ruled, did not have proper legal consent from British voters to disclose their personal information to Facebook for the Brexit advertising blitz ordered by Cummings. The company was also criticised for security failures. AIQ had failed to properly protect the data it had misused during the EU referendum.58
Immediately after the Brexit vote, AIQ put a quote from Dominic Cummings on its homepage: “Without a doubt, the Vote Leave campaign owes a great deal of its success to the work of AggregateIQ. We couldn’t have done it without them.” This testimonial was later removed.
*
On a warm afternoon in early September 2019, I arrived in Westminster to meet Conservative MP Damian Collins. It was noticeably quiet; Boris Johnson’s prorogation of Parliament had not yet been declared unlawful. A handful of protestors waving starry EU banners stood in the autumn sunlight. Police in high-vis jackets leaned on bright red bollards blocking a road where they had, a few days earlier, clashed with far-right demonstrators chanting “We love you, Boris”. When I asked Collins if he was busy, he smiled politely. “Always.”
Collins had emerged as an unlikely poster child for anti-Brexit campaigners. A Tory MP for Folkestone, a heavily Leave-supporting southern English port town, he had voted for Theresa May’s doomed withdrawal agreement with the European Union in the spring of 2019 and then backed Johnson’s leadership bid. But as chairman of the Department of Culture, Media and Sport select committee’s inquiry into fake news, the burly, cricket-playing former Saatchi advertising executive had overseen Westminster’s only sustained examination of electoral malfeasance during the 2016 referendum and beyond. Everyone from experts in online misinformation to senior staff at Facebook and Cambridge Analytica had been brought before Collins’s committee.
The inquiry initially began as a bipartisan examination of the spread of disinformation, inspired by the efforts in Washington of Democratic senator Mark Warner and his Republican counterpart Richard Burr to understand the role of Russia in the 2016 presidential election. Quickly, Collins’s remit widened. “We realised that fake news was not just a content problem,” he said, “but that the systems of social media could be used to create audiences and direct people to content.”
As the biggest digital campaign in British political history to that point, Vote Leave swam quickly onto the committee’s radar. Vote Leave’s law-breaking was, Collins believed, “just a tiny example of something that is easily achievable at scale”. Inspired by US political campaigns and Cummings and Elliott’s own prior referendum experience, Vote Leave had imported “the cutting edge of American techniques of using data and micro-targeting to drive messages”, he said.
But Collins’s inquiry had very limited powers to delve deeper into Vote Leave. Witnesses were not compelled to give evidence. When Dominic Cummings refused to attend the inquiry, he was found in contempt of Parliament. That sounds very serious, but it was no impediment to becoming the prime minister’s special advisor four months later. “We are still interested in talking to Cummings,” Collins told me, almost wistfully. “There are still questions we want to ask him.”
Despite the serious findings against Vote Leave, there was little political will in Britain for a major investigation into how election law was broken in 2016. Unlike a general election, which can be scrutinised by High Court judges if evidence of serious cheating emerges, only Parliament can investigate a referendum. Many commentators dismissed the criticism of Vote Leave’s law-breaking as a displacement activity for those who did not like the result. That might be fair as a description of political emotions, but it is hardly the point. The law had been broken.
Imagine, as Chris Wylie does, if we were talking about sport rather than politics. “If an athlete in the Olympics is caught doping and using illicit drugs, we don’t ask the question, ‘How much has that drug influenced the result? Maybe they would have come in first, maybe they would have come in second?’ If you cheat, you don’t get your medal. And in something as important as our democracy, we shouldn’t stand for cheating.”
In December 2019, Boris Johnson won a general election landslide. The campaign was a ramped-up reprise of Vote Leave. There were dubious claims and disinformation, anonymous adverts on social media and incessant lying. Journalists who asked awkward questions were sidelined. On the morning after the vote, Dominic Cummings stood smiling outside Number 10 Downing Street. Asked if he should take any credit for the emphatic win, Johnson’s special advisor said “no, not at all”. But it was not hard to see his fingerprints on a campaign that more than any other had pushed the limits, willingly eroding voters’ capacity to separate truth from falsehood. Vote Leave had taken over British politics.
* Elliott’s American wife, Sarah, previously worked as a Republican lobbyist in Washington DC. After the vote to leave the EU, she often appeared on British political talk shows as chairwoman of Republican Overseas UK, talking up the prospects of a post-Brexit trade deal with Donald Trump.