Select committee report shows way forward on voter registration

The Political and Constitutional Reform Select Committee’s 10^threport of session, Individual Electoral Registration and Electoral Administration, has generated widespread media coverage. It is safe to say that a select committee report on electoral registration has never been awaited, and greeted, with such interest. Stuart Wilks-Heeg looks at the reasons why.

The reason for the interest, of course, has been the widely-repeated claim that the government’s proposals for the introduction of individual electoral registration (IER) couldlead to 10 million additional voters disappearing from the electoral registers. The basis for this bleak projection of the impact of IER, the Electoral Commission’s written and oral evidence to the select committee, had been presented as a ‘worst case scenario’. But, perhaps inevitably given the source of the estimate, and the political passions which electoral registration evokes, it quickly acquired the status of ‘fact’ in some circles.

The Commission’s point, which was entirely legitimate, was that the government’s proposal to make registering to vote voluntary and a matter of ‘personal choice’ could feasibly see registration levels fall from the current 90 per cent or so of those eligible to vote to the 65 per cent or so of those who actually do vote. From here, all that was required was a quick few taps on a calculator and a tiny semantic twist, and suddenly others, notably Labour frontbenchers, were proclaiming that the Commission had told MPs that 10 million voters will disappear from the registers. From there, Labour’s claims about mass disenfranchisement for partisan gain and outrageous gerrymandering gathered such pace that some senior Labour figures even seemed to forget that it was they who had legislated for IER in the first place.

The truth is that nobody knows what the impact of IER, voluntary or otherwise, might be on registration levels. To start with, we don’t even know the current number of eligible electors missing from the registers. The figure of 3.5 million unregistered voters iswidely cited, but this is the estimate for England and Wales only, and it relates to comparisons against the 2001 Census. The Cabinet Office has provided funding for the Electoral Commission to undertake a national survey of the completeness and accuracy of the registers, so we should have an updated figure for Great Britain very shortly.

What is certain, however, is that the number of entries on the electoral registers will fall with the full introduction of IER from December 2015. This is partly because we have every reason to expect that the experience of Northern Ireland, which saw a 10 per cent drop in register entries after the introduction in IER in 2002, will be replicated. More importantly, we know that the number of register entries will drop simply because the current registers are known to contain at least some redundant and bogus entries as well as some ineligible duplicate entries. While the new system won’t deal with all of the latter, it should clean up most of the former. Unless eligible voters currently absent from the registers are added to them as a result of IER, then the total number of register entries will fall even if all the valid entries currently on the registers are retained. This much we know for sure.

Two further outcomes, which are virtually certain, are that the decline in the number of register entries will go beyond that resulting from a mere ‘cleaning up’ of the registers and that members of some social groups are far more likely to disappear from the registers than others. Again, the experience in Northern Ireland tells us that registration levels are likely to fall yet further among young people and students, residents in large communal establishments, private sector tenants, those on lower incomes and members of ethnic minority groups. Given the concentrations of these groups in urban and metropolitan areas, there will be substantial geographical variations in the extent to which registration levels decline under IER. The big issue here, as the select committee report crucially notes, is that the December 2015 electoral register, the first to be produced under ‘full’ IER, will be used as the basis for the redrawing of constituency boundaries prior to the 2020 General Election. It is with regard to the use of the registers for boundary reviews that Labour has been absolutely right to raise concerns about what may happen after the next General Election.

Given these observations, the really important issue is whether the government’s proposals for IER have derived enough from the Northern Ireland experience to ensure that registration levels do not plummet, and that they do not decline as unevenly, as they did in the province. As the select committee report notes, there is plenty of evidence to show that the government has considered the lessons of the Northern Ireland experience, but it is by no means clear that they have drawn all of the right conclusions.

Specifically, the government maintains that it is confident that its proposals to enable electoral registration officers (EROs) to access other public databases will result in registers which are both more accurate and more complete. The thinking here is that virtually all eligible electors will be identifiable from one public database or another and by using ‘data matching’ techniques to identify eligible electors who have not returned a registration form, EROs will be able to improve the coverage of the registers.

The trouble with this argument, as the select committee report correctly notes, is twofold. First, since the government intends that electors will be able to choose if they register or not, EROs may well find that the great bulk of the unregistered voters they approach simply ask to ‘opt-out’ of being registered, as the draft Bill would allow them to do. Second, evidence provided by EROs to the select committee raises serious doubts about whether data-matching will prove to be the panacea it is frequently claimed to be. While current pilots are ongoing, the committee took a clear view that the evidence it had received ‘suggests that data matching will be of limited effectiveness, especially in identifying potential electors’.

There can be little doubt that the select committee’s recommendations would improve the draft Bill, and that three of its proposed amendments, in particular, will be widely welcomed by those who have expressed concerns about the proposals. First, there should be a legal duty to register under IER, as is currently the case in Northern Ireland. Second, a full household canvass should take place in 2014. Third, a commitment should be made to use the register in place for the 2015 General Election, rather than the December 2015 register, as the basis for the post-2015 boundary review.

At the heart of these, and the committee’s other, recommendations is a concern to ensure that IER is introduced in a way which maximises the benefits of the change and minimises the risks. The select committee’s report is a reminder of the importance of pre-legislative scrutiny – something which the government was a little too eager to avoid during its ‘first phase’ of constitutional reform (although it has now resolved to conform to its own standards of ‘best practice’ on this matter).

Democratic Audit therefore welcomes and endorses the select committee’s report, and a range of organisations with an interest in electoral registration will surely do the same. Quite appropriately, given the purpose of pre-legislative scrutiny, the ball is now back in the government’s court.

SWH Stewart Wilks-Heeg is the former Executive Director of the Democratic Audit until 2013. He was an undergraduate and M.Sc student in the LSE Department of Government and gained his PhD at the University of Liverpool. He has written extensively on local government and urban politics in the UK. He is the author (with S. Clayton) of Whose Town is it Anyway? The State of Local Democracy in Two Northern Towns (York: Joseph Rowntree Charitable Trust, 2006).