Written evidence from Professor Justin Fisher[1] (TEB 16)

Public Administration and Constitutional Affairs Committee

The Elections Bill inquiry

Executive Summary

• The proposed introduction of voter identification is disproportionate. However, if it is to be introduced, consideration should be given to a form of identification that could be used for remote online voting.

• The removal of the 15-year limit on overseas’ voters right to vote in UK Parliamentary elections will expand the number of eligible voters. In order to promote participation amongst overseas voters, consideration should be given to providing a means for these voters to cast their ballot online.

• The proposed introduction of a Strategy and Policy Document is both a disproportionate and unnecessary measure, which threatens the independence and effectiveness of the Electoral Commission.

• The wording of the Bill in respect of notional expenditure requires examination in order to protect electoral agents, candidates and party staff.

• Proposed measures for third-parties in respect of dual registration and coordination require further examination and should be dropped from the Bill at this stage.

1. Voter identification

1.1              There is a strong argument to be made that this proposal represents a disproportionate response to a problem that appears not to be widespread. Where it exists, it is largely localised.

1.2              Public perceptions in respect of electoral integrity are frequently exaggerated. Where problems exist, they are rarely universal. As a consequence, recourse to public opinion alone is rarely a good guide to political reform.[2]

1.3              Repeated evidence from electoral agents (who have a far greater understanding of the electoral process) indicates that perceptions of electoral fraud are isolated. At every general election since 2005, I have conducted surveys of electoral agents from all the major parties contesting those elections.[3] These surveys repeatedly show high levels of satisfaction with general electoral administration, and low levels of perception of electoral fraud.

1.4              For example, in 2019, of those electoral agents who expressed an opinion, some 82% were satisfied and 6% were dissatisfied with the administration of the election. While satisfaction rates were high across the board, agents from Labour, Liberal Democrats and the Brexit Party had lower levels of net satisfaction. Agents with more experience were more satisfied in net terms, as were agents in Scotland. This geographical variation reflects finding in the previous three elections.  Agents in Wales were least satisfied.

1.5              Similarly, there was a low level of perception of electoral fraud. Only 11% (10% in 2017) perceived there to be A Little or A Lot. Fully 59% (61% in 2017) of those who expressed an opinion were of the view that there was None at All. Conservative agents were more likely to perceive some fraud than those from Labour or the Liberal Democrats. Indeed, this finding is common over the previous three elections, even when controlling for electoral performance and geography. 

1.6              Not only is concern about electoral integrity relatively isolated, the evidence from agents is also that perceptions of fraud are generally driven by electoral success or failure, party difference and geography, with very urban areas being those most likely to be ones where there are most concerns.[4]

1.7              Notwithstanding, should voter identification be introduced, it would be worthwhile considering a digital form of identification should remote online voting be introduced in the future.

1.8              Remote online voting is becoming more common internationally. Trials began in the UK in 2000, but there have been no further trails since 2007, the Electoral Commission arguing in 2014, that there should be no more trials of online voting until the UK government had put in place ‘…a comprehensive electoral modernisation strategy, setting out how the wider use of technology in elections will ensure the achievement of transparency, public trust and cost effectiveness,’ as well as the implementation of individual voter registration.[5]

1.9              However, the experience of COVID-19 has brought the question into sharp relief, following the postponement of elections in 2020 on grounds of safety. An online option would arguably have enabled these elections to take place as scheduled.

1.10          It appears that the Voter Card required for voter identification could be issued by the Returning Officer for each constituency. If that is the case, there is a risk of variation in practice. The quality of electoral administration in the UK is uneven.[6] That being so, my advice would be for the format of the Voter Card to be standardised. It could still be distributed locally, but it would be better to have a single Voter Card format.

2. Overseas Electors

2.1              The scrapping of the 15-year year limit on overseas electors’ rights to vote in UK Parliamentary elections is not without controversy, principally because those based overseas are not subject to the domestic measures introduced by any government. It is arguably perverse that an individual who will not be subject to any of the measures for whom they vote may nevertheless influence the composition of a UK government.

2.2              Notwithstanding, if the 15-year limit is scrapped, this will expand the pool of eligible voters. In such circumstances, consideration should be given to introducing online voting for those citizens to eliminate any reliance on postal services outside the United Kingdom, over which there is no domestic control nor automatic comparability of service provision. The Committee may wish to look at Switzerland as an example. Here, in three cantons, remote online was introduced to encourage participation amongst expatriates. The impact on increased turnout is not yet conclusive – that is still driven more by levels of party competition. However, remote online voting is very popular amongst Swiss expatriate voters, and they have adopted the approach to casting their ballot with more regularity than resident voters. [7]

3. The Electoral Commission

3.1.1        The Strategy and Policy Statement. Political oversight of the Electoral Commission already exists both in respect of the Speaker’s Committee, and the appointment of party nominated Commissioners.

3.1.2        The 2009 Political Parties & Elections Act introduced party-nominated Commissioners. At the time, I had reservations about the inclusion of party nominees as Commissioners, since the appointment of party nominees ran the risk of diminishing the independence of the Commission. However, the appointment of party nominees does not appear to have damaged the independence of the Commission to date.

3.1.3        For the Commission to function effectively and independently, it should – as far as is possible - be free from interference from those whom it seeks to regulate. The provision to require a Strategy and Policy Statement, approved by Parliament, together with the extension of the function of the Speaker’s Committee to examine the Commissions’ compliance with such a document will significantly diminish the necessary independence of the Commission, both literally and in the eyes of the electorate. Such a risk is amplified by the fact that a majority of the members of the Speaker’s Committee on the Electoral Commission are from the governing party.[8]

3.1.4        Such a Strategy and Policy Statement is also unnecessary. Far from requiring additional oversight, the Commission already delivers excellent work in ensuring the high levels of satisfaction in the integrity of the electoral process amongst those who are most knowledgeable and closely involved. For example, a survey of electoral agents at the 2019 general elections showed that amongst those who expressed an opinion, 78% agreed that the rules in respect of election spending and donations, were clear; 72% viewed the Electoral Commission as a useful source of advice; 75% thought Electoral Commission guidance for candidates and agents was clear and easy to use; and 75% thought Electoral Commission written information on the verification and count was clear and easy to use.[9]

3.1.5        The strength of the Commission is the active work it undertakes with those it regulates to build understanding of the law and ensure compliance. An indication of that strength is best demonstrated not just by levels of satisfaction amongst professional compliance officers, politicians or commentators, but by those working on the ground in elections, the vast majority of whom are volunteers.[10]

3.1.6        The proposed introduction of a Strategy and Policy Document is both a disproportionate and unnecessary measure. It threatens the independence and therefore the effectiveness of the Electoral Commission. There is little evidence to support the need for further political and parliamentary accountability and therefore I would recommend that this measure is dropped from the Bill.

3.2.1        Criminal Prosecutions by the Electoral Commission – There is a logic to this proposal both in terms of the avoidance of duplication and past experience.

3.2.2        When the Commission was first established, the only sanctions available were those of criminal prosecution. There was a reluctance to invoke these, resulting in non-compliance going unpunished.[11]

3.2.3        However, if the Commission is to be prevented from bringing criminal prosecutions (and indeed, even if it is not) it is essential that a wider range of civil sanctions is available to the Commission. The current maximum fine available is £20,000. Such a figure risks making the fine part of the ‘costs of business’ for a party. However, caution must be paid not to make the fine so large as to potentially threaten the entire operation of a party. Such a case occurred in Georgia, when in 2012, the regulator imposed the largest party finance sanction ever in a Council of Europe member state.[12]

3.2.4        Notwithstanding, the logic of the argument in respect of duplication may be undermined if other regulators retain the right to bring criminal prosecutions. If this is the case, there is a far weaker argument is respect of removing this power from the Commission.

4. Notional Expenditure

4.1              The measures outlined are – in principle – an appropriate response to the question of notional expenditure. As I have outlined in a previous submission to the Committee, [13] there is no perfect solution to the question of notional expenditure, and the current arrangements represent the least bad option available – a view also reflected by the Committee on Standards in Public Life.[14]

4.2              However, while I endorse the Bill’s approach to his question, I have some misgivings in respect of some technical aspects.

4.3              The wording “…encouraged by the candidate or candidate’s electoral agent” as an indication of positive intent is problematic, as there may be no proof that could be offered that such encouragement took place. The only workable solution to rely solely on written declarations as in the case, for example, of reviewing donations, such that there is a clear and unambiguous paper trail.

4.4              There needs to be a clear statement in respect of the identity of the responsible person in a political party in terms of the authorisation of spending that may be ascribed to a party or to a candidate. In the ruling relating to R v Mackinlay and others the judge noted that ‘There appears to have been a belief that Central Headquarters staff salaries and accommodation of staff employed by Central Headquarters were a central party expense, even if those staff were living temporarily in a constituency for the duration of the election campaign.’ He added that the official ‘...was not alone in that she worked in a culture which tolerated some of what she did’.[15]

4.5              While there is no suggestion from the judge that there was anyone other than the accused who ‘was aware of the dishonest calculations and concealment of invoices’, the prosecution of a party official rather than the agent or candidate raises the possibility that party staff may in future be liable. The naming of a responsible person in the party would protect party staff from the possibility that blame could be passed either to a member of staff or to the election agent. Such a move would further incentivise parties to prevent such actions and would be consistent with the appointment of a responsible person for accepting party donations.

4.6              Separately, the response by parties to R v Mackinlay and others has produced some perverse outcomes. For example, one party at the 2019 election sought to keep the expenses associated with the visit of a leading party figure as party expenditure by ensuring that the candidate was not seen with the visitor, going so far as to producing dedicated doorstep canvas scripts.[16]

4.7              While the Electoral Commission provides some guidance on such matters,[17], it is not always definitive due to the necessary judgement calls in respect of responsibility. Indeed, in the case of the leader’s visit where a candidate attends but is not actively promoted, the law only allows the Commission to recommend that a ‘…portion which is reasonable to reflect your use of the costs must be reported in your candidate spending return’. There is no definition of ‘reasonable’.

4.8              For these reasons, and to avoid disputes about the level of spending allocated to candidates and the responsibility in the case of any violation of the rules, a clear paper trail and identification of responsible persons at party level is necessary.

5. Political Finance

5.1.1        Third party campaigner registration. The new lower tier for registration makes good sense and serves to enhance transparency at elections.

5.2.1        Restriction of all third-party campaigning to UK-based entities and eligible overseas electors. This restriction also makes good sense; particularly in respect of foreign third-party campaigners. In general, I see no problem in imposing further registration responsibilities on third parties, given that during election campaign periods, it is those standing for election who should have prominence given that they are accountable to the electorate (whereas third parties are not). However, it should be recognised that this will be almost impossible to enforce in respect of online activity by overseas third parties. While digital imprints will identify authorised online campaigning, it will be extraordinarily difficult to prevent online campaigning by foreign third parties.

5.3.1        Ban on registering as both a political party and a third-party campaigner. At first sight, this looks like a logical step to close an unexpected loophole. It also prevents spending limits being artificially inflated through the combination of third party and political party spending.

5.3.2        However, there is a danger that this provision may unfairly disadvantage third parties. It is quite conceivable that a third party may run a national campaign (perhaps on saving hospitals), but field candidates in isolated seats, where there is a particular local issue (perhaps a specific local hospital under threat). In effect, the third party and the political party would be fighting the same campaign, but under this proposal, would have to separate themselves artificially, thereby creating barriers to entry (though increased administrative burden), but also creating additional difficulties in respect of the coordination between third parties and political parties that the Bill also seeks to prevent.

5.3.3        It is quite appropriate to seek to prevent the artificial inflation of spending limits by combining political party and third-party limits, but this proposal as is stands represents a disproportionate response. This problem is by no means widespread. That being so, and in order to ensure to that there is not an unexpected adverse consequence for the third parties, I would recommend either limiting the number of seats in which a co-registered third party and political party could stand (perhaps three), or dropping the provision altogether.

5.4.1        Restrictions on coordinated spending between parties and third parties. The logic of this proposal is sound, but the existing comparable legislation on coordination – the so-called Working Together rules as they apply in referendums - is not fit for purpose. This was tested at the 2016 referendum and caused considerable difficulties to both designated campaigns (analogous to political parties) and registered campaigners (analogous to third parties).

5.4.2        A study of perceptions of the administration of the referendum and efficacy of the regulations amongst permitted participants was conducted soon after the 2016 referendum.[18] The study found that the legislation on Working Together was very poorly understood. Some 56% of campaign participants found the rules difficult to understand, compared, for example, with only 12% who found the process of recording and reporting donations difficult.  Difficulties were experienced by both the Leave and Remain sides and particularly amongst small campaign groups.

5.4.3        The rules were so complex that campaigners effectively divorced themselves artificially from like-minded groups to avoid any charge of potential coordination. A significant proportion (42%) decided not to work together after initially considering the option.

5.4.4        Participants claimed that the Working Together rules effectively blocked official coordination as they presented too much of a risk in respect of compliance. For example, one designated group said: “We ended up having to send our Legal Director along to each meeting to make sure of that, and report back to the responsible person that there was no coordination happening.”

5.4.5        Participants also reported difficulties whereby one participant complied with coordination declarations while another did not. In this case, it was unclear, which participant was to blame in such circumstances. Moreover, there was a concern from a designated group that they would not know if a non-designated group had claimed to have worked with them until after the non-designated group had submitted its return.

5.4.6        The oral evidence presented on 20th July 2021 from non-party campaigners to this committee in respect of the Review of the Electoral Commission highlights similar concerns, such that while the guidance from the Electoral Commission on coordination may have improved, the weaknesses of the existing legislation remains,[19] just as in the case of the 2016 referendum.

5.4.7        In short, while this provision is well-intentioned, I would strongly urge further examination in the light of the 2016 experience and of the experience of non-party campaigners to ensure that the problems demonstrated there are not replicated. Given the regularity of elections, there is a real danger that the difficulties observed in 2016 could be amplified.

5.4.8        I would strongly recommend that the principle of coordination is dropped from the Bill until sufficient consideration has been given to making the regulations work effectively. At the same time, the rules on coordination in referendum campaigns should be re-drafted.

August 2021