We now come to the first Select Committee statement. In a moment I shall call the Chair of the Public Administration and Constitutional Affairs Committee, Mr Bernard Jenkin, who will speak on his subject for up to 10 minutes, during which no interventions may be taken. At the conclusion of his statement, I will call Members to put questions on the subject of it, and to do so briefly. I will then call Mr Jenkin to respond briefly to those questions in turn. Members can expect to be called only once. I reiterate that interventions should be questions, and should be brief. Front Benchers may take part in questioning.
I am grateful to the Backbench Business Committee for giving me time to present PACAC’s 12th report of this Session, which is entitled “Lessons learned from the EU Referendum”, and which is still topical in so many ways. We shall be producing further reports at the fag end of the current Parliament—including, on Monday, a report on the Advisory Committee on Business Appointments—but I fear that there will be no opportunities for them to be presented on the Floor of the House before the Dissolution.
The referendum on our membership of the EU and the vote to leave by a margin of 52% to 48% represents one of the most momentous political events in our politics for decades. It has already had, and will continue to have, far-reaching consequences, and it will shape the destiny of our country. PACAC’s report, published last week, seeks to draw out some important lessons to be learned from the EU referendum in relation to the purpose of referendums and how they should be conducted. It builds on the work of our predecessor, the Public Administration Select Committee, whose report on the Scottish independence referendum was published in the last Parliament. We hope that both reports will be regarded as required reading for anyone who plans a major referendum in the future.
PACAC argues that referendums are appropriate for resolving questions of key constitutional importance that cannot be resolved through the usual medium of party politics. However, it also argues that the system is less satisfactory in the case of what might be called a “bluff call” referendum which—as happened last June—is used by the Government to try to close down an unwelcome debate. Future Parliaments and Governments must consider the potential consequences of promising referendums, especially when, as a result, they may be expected to implement an outcome that they opposed.
PACAC argues that referendums should be limited to matters that are in some way of fundamental or constitutional importance and lend themselves to binary questions, and to instances in which the consequences of both possible outcomes are clear. That is not least because referendums still create a tension in our parliamentary system of government. Although we are becoming used to direct democracy, it contrasts with our constitutional traditions and culture of representative democracy. Direct democracy, as we have just learned, can be a shock to the system, particularly when most of the elected representatives disagree with the result. The forthcoming general election is all the more necessary because it will heal that rift, and will translate the direct mandate from the EU referendum into a representative mandate for a new Government and Parliament.
That, however, is the point of a referendum. It is a new way of challenging entrenched opinion, just as the Anti-Corn Law League overturned agricultural protection and the campaigns to widen suffrage and to open our economy to free trade challenged the establishment in previous centuries. Today, people are educated and have direct access to the information. Voters are therefore more capable of deciding individual questions for themselves, and less willing to accept wisdom handed down from on high.
PACAC also considered the conduct and delivery of referendums in the future. It found that Government fears that the purdah restrictions provided for in section 125 of the Political Parties, Elections and Referendums Act 2000 would impair the conduct of government were groundless. The Committee recommends that those restrictions, which are vital to the fair conduct of referendums, should be extended to cover the full 10 weeks of a referendum period, as recommended by the Electoral Commission. It is something of a testament to Select Committees that we succeeded in persuading the House of Commons to prevent the Government from altering the purdah rules in advance of the referendum. We also recommend that the rules be updated to reflect the digital age. We support the Law Commission’s proposals to consolidate the law regulating the conduct of referendums.
As for the administration of the referendum, the evidence gathered during PACAC’s inquiry suggests that while it was not without some faults, the referendum was on the whole run well. PACAC commends the Electoral Commission for the successful delivery of the referendum, which was of enormous scale and complexity.
One of the most significant problems was the collapse of the voter registration website just hours before the registration deadline on 7 June. The Government attributed that to “unprecedented demand”. More than 500,000 online applications to register to vote were recorded on 7 June alone. According to the Electoral Commission, the problems that led to the website’s crash were aggravated by a large number of duplicate applications: 38% of applications made during the campaign were duplicates, and there was no way of checking online whether an application was a duplicate. PACAC supports the Electoral Commission’s recommendation that the Government develop an online service to enable people to check whether they are already correctly registered to vote, which would be invaluable in preventing the website from collapsing again in the future. Such websites should be better tested for resilience.
The media devoted a great deal of attention to our having raised the possibility that the website collapse had been caused by a cyber-attack. Whether or not that can be proved is not the point; it is important to be aware of the potential for foreign interference in referendums or elections, responsibility for which is actually being claimed by some countries, and attacks experienced by others. Permanent machinery for monitoring cyber-security in respect of elections and referendums should be established.
Lessons relating to the protection and resilience of IT systems against possible foreign interference must also extend beyond the technical. Our understanding of cyber is predominantly technical and computer-network-based, but Russia and China use a cognitive approach based on an understanding of mass psychology and how to exploit individuals. In my capacity as Chair of PACAC, I shall be writing to the Intelligence and Security Committee to raise the issue of cyber-security in the EU referendum, and to ask whether it will be following up on PACAC’s concerns. Today, however, I am encouraged by reports that the National Cyber Security Centre will be advising political parties on the matter during the forthcoming election campaign.
PACAC also again looked at the role played by the civil service during the referendum. We expressed concern that the manner of the presentation of some Government reports, particularly those from the Treasury, and the decision to spend £9.3 million on sending a leaflet to all UK households advocating a remain vote, were inappropriate and undermined public confidence in civil service impartiality. PACAC reiterates the recommendation made by its predecessor, PASC, that there should be a new paragraph in “The Civil Service code” to clarify the role and conduct of civil servants during referendums. It currently contains no reference to referendums.
Finally, we looked at the degree of contingency planning that was made in respect of a possible leave vote. In the run-up to the 1975 referendum, according to the contemporary accounts, Whitehall prepared for a possible UK exit from the common market with a “fairly intensive” programme of Cabinet Office-led contingency planning. PACAC was alarmed to learn that in the run-up to the EU referendum last June, the Government’s official position was that there would be no contingency planning. The only exception was planning in the Treasury to anticipate the impact of a leave vote on the UK’s financial stability.
PACAC was relieved to learn that work was undertaken within the civil service on the potential implications of a leave vote, albeit without the knowledge of Ministers and despite their explicit instructions. There was a secret away day. Civil servants should never have been asked to operate in a climate in which contingency planning was officially banned, and the Government should not have shirked their constitutional and public obligation to prepare for both possible outcomes. PACAC recommends that, in the event of future referendums, civil servants should be tasked with preparing for both eventualities, as they do in the case of general elections.
It is essential for referendums to be well run, to be conducted fairly, and to command public trust and confidence. PACAC therefore hopes that the Government will heed our recommendations so that the country will be ready for any further referendums in the future.
Let me take this opportunity to thank the House, but more particularly my Committee and its dedicated staff, for the privilege of serving as Chair of PACAC in the current Parliament.
Thank you, Mr Speaker, for the fairness you have always shown me in this and previous Parliaments.
The report is clear: the referendum was called to call the bluff of the Brexiteers, civil service neutrality was clearly jeopardised and, as the Chair of the Committee said, there was no preparation for the vote to leave. Is not it obvious that the referendum was held not in the national interest but in the governing party’s interest? Now, with 30 of its MPs under investigation, we are having an election, instead of focusing on the outcome of the referendum. Paragraphs 102, 103 and 104 of the Committee’s report should concern the House—and, in fact, the whole country. We have not done enough to secure our systems for either referendums or elections. In the Chair’s view and the view of his Committee, are our systems strong enough, at the time of a snap general election, in the event of a concerted cyber-attack, to which the report refers, by a foreign power or from some other source? Even at this late stage, does he think that there is anything that we can do to strengthen our systems’ resilience?
I am grateful for the hon. Gentleman’s question. I will not tangle with all the things that he raised, but we have a pretty resilient system. The fact that the vast majority of votes cast are pencil or pen marks on bits of paper that are physically counted means that it is basically an impossible system to hack. What we need to be aware of is the vulnerability of electoral registers and systems. The dispersal of our electoral register among different electoral authorities is another source of resilience: there is not one system to hack. However, we need to be aware of what certain countries might want to be seen to be doing—what they might want to be seen to be attempting to influence the result of, or want to be thought to influence the result of. I do not think that any country influenced the result of the leave vote in the EU referendum. I do not think that the result in any election in any major country would have been altered, but we need to understand why certain countries are doing this and what psychological effect they are trying to create by attempting these things, and we need to be alert to the vulnerability of our systems.
I congratulate my hon. Friend and his Committee on a comprehensive report. I agree with the remarks from the Opposition on cyber-activity. Does he agree that we need much better monitoring of cyber-activity as a matter of urgency, not just for referendums but for elections generally?
I do. We made a specific recommendation that a new body be established to monitor cyber-activity in relation to referendums and elections. However, I emphasise that we are in a much stronger position than countries that have electronic voting or single population registers. I have confidence in our system, although we need to be more alert in order to maintain public confidence; that is the main point.
I am a member of the Committee, so there is plenty I could say about the report, but I will respect your wish, Mr Speaker, for us to keep comments and questions short. The most important line in the report is perhaps that there should be
“careful and restrained use of the machinery of government”—
for example, of the civil service and the purdah provisions. Will the hon. Gentleman join me in encouraging the UK Government to trust the devolved Administrations and allow them to organise and run their referendums without external interference from this place?
It is a fact—I make no comment on it, as an impartial Chairman of my Committee—that referendums are constitutional matters and therefore are reserved to the United Kingdom Parliament. I recognise that there is some demand for a new referendum in Scotland, but even the Good Friday agreement says that there should not be a referendum more than once every seven years. There needs to be a respectable interval between referendums, otherwise they just become meaningless. How many times in the European Union have we seen another referendum called when the first gave the wrong result? I do not put the Scottish National party in that category, but calling referendums too often is actually a contempt for democracy.
Was there any discussion in the Committee about the franchise for the referendum? If 16 and 17-year-olds and EU citizens had been allowed to vote, we might have had a very different result. They will be allowed to vote in the Scottish council elections in two weeks, but they will be denied a vote in the UK general election about four weeks after that. Would it not be appropriate to have consistency, and for the franchise to be as wide as possible?
I ask the hon. Gentleman: why stop at 16? Why not 14 or 12? These are subjective judgments made by different bodies in different parts of the constitution. That franchise is a devolved matter; it is a matter for the Scottish Parliament. Personally, I favour maintaining the status quo in the UK.
Does not the hon. Gentleman agree that our system for referendums and elections is more vulnerable to invisible manipulation and corruption than at any time since 1880? The great weakness of the report is that it ignores the evidence provided, principally by the journalist Carole Cadwalladr, on the use of botnets, artificial intelligence and algorithms to influence millions of voters. The evidence is there from the United States and from this country. Under-the-counter systems are being used that we do not understand; they seek to trawl through websites to get information and subsequently influence voters. We are trying to deal with tomorrow’s systems and tomorrow’s high technology with regulations that are long out of date.
Is not it likely that in the coming election there will be more manipulation, that there could well be cyber-attacks, and that we cannot trust these results, because what is happening is under the counter and the Electoral Commission has no tools to deal with it in the way it should? We should not have a general election without finding out the truth about the manipulation that has taken place here, in the US and possibly in other countries that we do not know about. We have not heard about that from GCHQ; we should have heard from it. It has reported on what has happened in the US, where there were cyber-attacks and manipulation. That could well have happened here; we do not know because we have not asked.
With respect, I have asked that question, and I feel I have been rather brushed off by Ministers, perhaps on the advice of officials who are perhaps not as knowledgeable as the Committee is about the technicalities, algorithms or indeed the cognitive approach taken by some of the countries with which the Committee has made itself familiar. I am always grateful for the hon. Gentleman’s contributions to the Committee—I think he is our longest-serving member—but personally I do not agree that this threatens the credibility of our elections. In 1880, one of my predecessors in North Essex conducted his election with his wife walking behind him down the high street handing out gold sovereigns. We have come a long way since that kind of corruption in elections, but we need to be alert to the things that the hon. Gentleman draws attention to, and to be ever more alert to the fake news that appears on the internet and is designed to manipulate people’s expectations.
I pay tribute to the hon. Gentleman and to the work of the Committee. I was proud to be a member of its predecessor Committee, the Public Administration Committee, in the previous Parliament. Who knows, if there are more colleagues on the Liberal Democrat Benches in the next Parliament, as I am sure there will be, perhaps we will qualify for a place on the Committee.
Clearly, the consequences of the referendum, whatever view people take, were not properly considered. Planning was not done and the referendum Act was very shoddy and ill considered. It need not have been. Does the hon. Gentleman think that we need more clarity, and perhaps even legislation, to avoid that kind of thing and such a political referendum being organised in future, without planning?
I think there is always an advantage in what one might call a post-legislative referendum, or a referendum on a proposal on which a White Paper is produced. The devolution referendums in 1997 were both premised on pretty well developed Government policy. One might even pay tribute to the Scottish National party and say that at least it produced a comprehensive document. The leave campaign did produce 600 pages explaining what leave might look like, but the Government had done no preparation, and it is for the Government to prepare for the outcome of a referendum that the Government initiated. I agree with the hon. Gentleman, and perhaps he will join us on the Committee again; I miss him.