Ensuring equal representation in Parliament: who counts?

The UK government announced on July 16 that it had decided to end the transition period to Individual Electoral Registration early. The Electoral Commission had recommended that it did not, and thereby allow some 1.9 million people who would otherwise be removed to remain on the electoral roll for a further year. The Commission’s reasons for this concern the potential disenfranchisement of a large number of potential voters at next year’s important elections to devolved bodies and local governments. A further important reason not given is that ending the transition period early could bias the allocation of Parliamentary constituencies for the 2020 general election against the country’s urban areas. Ron Johnston, Charles Pattie and David Rossiter explain why.

Credit: Leo Reynolds, CC BY NC SA 2.0

In 2011 the coalition government passed legislation to reduce the number of MPs and alter the procedure for determining the boundaries of Parliamentary constituencies. To remove substantial differences in levels of representation both between and within the UK’s four countries, with four exceptions all of the 600 new constituencies (replacing the current 650) were to have electorates within 5% of the national average (quota).

The Boundary Commissions implemented the new rules, and had produced revised sets of constituencies by early 2013 when Parliament voted to abort the ongoing redistribution and delay the procedure’s implementation until after the 2015 general election. The Conservatives remain committed to the change – believing that the current variations in levels of representation (i.e. the number of electors per MP) advantages the Labour party. Their 2015 general election manifesto included a statement re-committing them to that policy – We will also continue to reform our political system: make votes of more equal value through long overdue boundary reforms, reducing the number of MPs… – and David Cameron clarified that position on 1 July 2015 at Prime Minister’s Questions, saying that we should get on and create more equal constituencies and reduce the number of MPs to reduce the cost of politics.

The Boundary Commissions will start work again implementing the new rules in March 2016. The quota will be determined by dividing the UK’s registered electorate by 596 (the 600 new constituencies less the four special cases – Orkney & Shetland; the Western Isles; and the Isle of Wight – where the constituency boundaries are pre-determined). This will then determine the number of seats to be allocated to England, Northern Ireland, Scotland and Wales, and to each of England’s nine regions, and individual constituency boundaries will then be recommended.

So the number of registered electors is crucial to this process. For decades the electoral roll has been compiled each autumn following a canvass of every household, but we are now – save in Northern Ireland where the change was made some years ago – in the transition period to a new system of Individual Electoral Registration. This has been designed to make the electoral roll more complete and accurate, but it is probable that many who could be registered won’t be. How many are, and where they live, is crucial to this new constituency-definition exercise.

We can illustrate this issue by looking at what the allocation of constituencies across the four UK countries would have been if recent electoral rolls had been used, compared to the current distribution of 650 MPs.

	Current	2010	2014	2015
England	533	500(+2)	498(+2)	499(+2)
Northern Ireland	18	16	16	16
Scotland	59	50(+2)	53(+2)	51(+2)
Wales	40	30	29	30

The second column shows the number of MPs allocated in the aborted review that started in 2011 using electoral data for 2010. (Only 596 of the 600 seats are allocated because the four ‘protected constituencies’ are excluded from the calculations.) The next column indicates the allocations if the electoral data for the end of 2014 were used: Scotland would gain three seats from its 2010 allocation – presumably because of the extra enrolment for the Scottish independence referendum – whereas England would lose two and the Welsh contingent of MPs would fall to 29. After the electoral roll was compiled in late 2014, however, people not on the roll at their current home could register to vote at the 2015 general election until 10 April of that year, and over one million did. Using the electorate figures for that election, the final column shows a small change from the 2010 data – one extra seat for Scotland and one less for England.

Such a change might seem small, but we calculated that if England were to lose a seat, and (to take just one region as an example) and the reduction was in the Southwest region, then all of the constituencies in Cornwall, Devon and Somerset would have to be redrawn in order to fit the rules: a small change can have a major knock-on effect. A small difference in the allocation of seats can result in substantial changes locally – and it is very likely that will occur.

So far so straightforward: the electoral roll will be compiled and the Commissions will start their work, probably allocating seats across the four countries very close to that shown in the table’s final column. But a major unresolved issue – until the government’s announcement on 16 July – regarding the compilation of the electoral rolls will have a substantial impact on the number of registered electors in December 2015, and hence the determination of the electoral quota and the allocation of seats when the Boundary Commissions start work in March 2016. The December 2015 electoral roll numbers could be very different – especially in some places – from those in place at the time of the May 2015 general election, especially given the government’s decision.

The transition period in the switchover to Individual Electoral Registration was due to end by December 2016 but could – if the government so determined by the end of August 2015 – be ended a year earlier. Under that new system, a household canvass is conducted every autumn to identify the eligible voters there. Those identified and whose presence at that address cannot be verified from other sources are sent an application form on which they must apply to be registered individually – and if they do not do so then at the end of the transition period to IER they will be removed.

The Electoral Commission’s recently-published report – Assessment of progress with the transition to Individual Electoral Registration – estimates that currently c.1.9million individuals (some 4 per cent of all of those listed) are on the electoral roll but have not registered individually. If the transition is ended in 2015, then they will all be removed from the electoral rolls assembled at the end of the year (unless they re-register in the short remaining period).

The Electoral Commission recommended that the transition period continue into 2016, and those 1.9million individuals remain on the roll. The rationale for introducing IER was to ensure that the electoral roll was both complete and accurate and the Commission recognises that ending the transition early, in 2015 rather than 2016, would bring both benefits and risks from removing those who have not registered individually. The benefits are that the resulting electoral roll may be more accurate because many of those removed no longer live at the addresses indicated; they have moved and are no longer eligible to vote there (but may have been replaced by others, who are also not registered individually!). The risks concern incompleteness: there will also be many among those removed who are eligible to vote at the address indicated, but they have not registered individually and so could – in effect – be disenfranchised because of that failure (though they could apply to be added to the roll at any time after its publication – but their late additions will not change the numbers on the rolls compiled by the end of the year and which the Boundary Commissions will sue in the seat-allocation process).

The Commission’s advice was that – given the important elections to be held in 2016 (to the Scottish Parliament, the Welsh Assembly, the Greater London Assembly, and many local governments) – the risks from ending the transition early outweigh the benefits and those who have not registered should remain on the rolls for a further year. The government rejected that advice.

The Electoral Commission and the government both recognised that the electoral roll compiled at the end of 2015 will be used for the Boundary Commissions’ next exercise, but the Commission did not take this into account when making its recommendation. The government did when making its decision; its written statement states that IER should be ended early so that the ‘registers used to conduct the Parliamentary boundary review and for next year’s elections are as complete and accurate as they can possibly be’ – which they deem will be the case by removing up to 1.9 people from the electoral roll compiled at the end of this year. (How many eventually have to be removed will depend on the success of Electoral Registration Officers in persuading those who might have to go to register during the next few weeks.)

But will the registers then be as accurate and complete as possible? For the elections to be held in May they may be, because individuals not on the electoral rolls can be registered until a few weeks before the contests are held. But the electoral rolls compiled this autumn will not be updated for the Boundary Commissions’ exercise: they will use the number of electors registered on the ‘purged’ roll and will take no account of any who register later. This could have a significant influence on the country’s next electoral geography.

Research undertaken on the completeness of the electoral rolls indicates that those not registered (some 14 per cent of the eligible population according to the last detailed estimate in 2014) are concentrated in certain social groups (the young; flat-dwellers, especially renters; members of ethnic minorities; recent movers etc.) – and those groups are concentrated in particular places: urban areas, especially the big cities. Those groups are likely to comprise a large proportion of those who may be removed from the electoral roll at the end of the transition to IER.

The government’s decision to end the transition period early means that the Boundary Commissions’ allocation of constituencies to the different parts of the UK will be considerably different from what it might have been if the a million or more voters remained on the roll for a further year. The main ‘loser’ is likely to be London. Estimates undertaken when IER was being introduced suggested that the number of MPs allocated to London could vary by 6-8 depending on the completeness of the roll. (London currently has 73, and the aborted redistribution in 2011-3 reduced this to 68; with the IER transition ended in 2015 it could be as low as that again – or even lower – but if the 1.9million had remained on the roll it will probably be larger.) The less complete the roll the smaller the number of MPs for London, and other large cities such as Birmingham, Glasgow and Manchester, and the larger the number for the shire counties – a different geography with clear political implications.

The Conservative government is committed to equality of representation in the House of Commons. To achieve that, the electoral roll used by the Boundary Commissions to allocate constituencies and determine their boundaries across the UK should not lead to an under-representation of some areas and over-representation of others – but this could well be the consequence of their decision (which Liberal Democrat peer Lord Tyler is trying to get Parliament to reverse). Have political calculations been involved in that decision, just as they were to change the rules that the Boundary Commissions operate – and the number of MPs?

—

Note: this post originally appeared on The LSE British Politics and Policy blog. It represents the views of the authors and not those of Democratic Audit UK or the LSE. Please read our comments policy before posting.

—

Ron Johnston is a professor in the School of Geographical Sciences at the University of Bristol

David Rossiter is a retired Research Fellow who formerly worked at the Universities of Bristol, Leeds, Oxford and Sheffield

Charles Pattie is a professor in the Department of Geography at the University of Sheffield. They have written together on the Boundary Commissions and constituency definition in the UK since the early 1980s and co-authored The Boundary Commissions: Redrawing the UK’s Map of Parliamentary Constituencies (University of Manchester Press, 1999).