LSE - Small Logo
LSE - Small Logo

Blog Admin

July 2nd, 2014

How to balance the competing demands of arithmetic, continuity and geography in designing constituencies


Estimated reading time: 5 minutes

Blog Admin

July 2nd, 2014

How to balance the competing demands of arithmetic, continuity and geography in designing constituencies


Estimated reading time: 5 minutes

The review of Parliamentary constituencies that ended prematurely in 2013 would have resulted in most of the 600 seats contested at the 2015 general election being very different from the current 650. Another review – again reducing the number of MPs to 600 – is scheduled to start in 2016, preparatory for the 2020 general election. But if the rules for defining constituencies are changed slightly, would this significantly reduce the amount of disruption to the map of constituencies? Recently published research by Ron Johnston, David Rossiter and Charles Pattie addresses this question: the extent of the change could be reduced, but a significant amount (more than at previous reviews) was almost bound to occur – however many MPs there are.

Concern regarding variations in constituency electorates, coupled with a drive to cut the cost of Parliament in the wake of the 2009 expenses scandal, stimulated Conservative Party commitments in its 2010 General Election manifesto to legislate to “ensure every vote will have equal value” and reduce the size of the House of Commons.

Legislation passed in 2011 put that intention into practice and the Boundary Commissions commenced their task of producing a new set of 600 constituencies all, with the exception of four special cases, having electorates within +/-5 per cent of the UK average. They consulted on their proposals and revised them accordingly, but their work was halted by Parliament before its completion because of disagreements within the coalition on the programme of constitutional change. By then, however, MPs and party organisations had become aware that the new system, with its emphasis on electoral equality, necessarily disrupted the map of constituencies very significantly. Fully 54 per cent of the current seats would be subject to major change, compared to only 30 per cent at the last review, and many more constituencies crossed local government boundaries than previously.

The Boundary Commissions are currently required to begin their task again in 2016, producing a new set of constituencies for the 2020 general election. But the very disruptive consequences of the previous exercise generated questions regarding the nature of the new procedure. Would it be possible to reduce the disruption substantially, yet maintain the general principle of electoral equality, with a more relaxed tolerance around the average? And would there be less disruption if the number of MPs was retained at the current 650, rather than reducing it to 600?

Our research answering these two questions found that a more relaxed tolerance would reduce the disruption somewhat, but at least one-third of all constituencies would almost certainly have to experience major change – and the size of the House of Commons was largely irrelevant to those conclusions.


The variations in constituency electorates stimulating the Conservatives’ commitment to ‘ensure every vote will have equal value’ are illustrated by the latest (December 2013) data on the number of electors in each constituency, which ranges from 22,084 in Na h-Eileanan an Iar (the Western Isles) to 111,800 in the Isle of Wight. The average was 71,578, but one-quarter of all seats had fewer than 65,000 and one-tenth less than 60,000, whereas a further one-quarter had more than 76,000 and one-tenth more than 80,000. Furthermore, the larger and smaller constituencies were not evenly distributed across the UK. Of the 42 with fewer than 60,000 voters, for example, six were in Scotland, eight in England, and the remaining 28 were all in Wales: of the 60 with more than 80,000 voters, five were in Scotland and the remaining 55 in England.

The Conservatives’ proposed changes to remove these inequalities were realised by the coalition government in the Parliamentary Voting System and Constituencies Act, 2011. This introduced a new set of Rules for Redistributions that: (a) mandated a UK-wide norm for constituency electorates; (b) required the four Boundary Commissions to create seats within +/-5 per cent of that norm (usually termed the quota); and (c) cut the number of constituencies from 650 to 600.

The main factors influencing the Boundary Commissions’s deliberations at previous redistributions were the map of major local governments and continuity with previous sets of constituencies. The new Rules for Redistribution implemented in the 2011 Act meant there was a single UK quota (76,641) and – with four exceptions – all constituency electorates had to be within 5 per cent of that figure (i.e. between 72,810 and 80,473). That rule was virtually unbreakable, and only within that constraint could the Commissions take continuity of representation and not crossing local government boundaries into account. (The Act allowed for special treatment by the Commissions for areas of very low population density – in effect, the Scottish Highlands only – and in certain circumstances in Northern Ireland; neither was applied in the review that started in 2011.) Consequently the recommended constituencies (both the Commissions’ initial and then, after public consultation, revised proposals) differed starkly from those in previous redistributions: fully 54 per cent of the revised recommendations were major changes from the previous set, for example.

These very different maps of constituencies from those currently in place generated a somewhat belated realisation that what had been gained in arithmetic terms had come at the expense of continuity and geography. The variability amongst constituency electorates was halved compared with previous reviews, but most existing seats experienced major change and across urban Britain a match between local government and constituency boundaries had become the exception rather than the norm.

Wider disagreements within the coalition over constitutional reform led to the Liberal Democrats joining with opposition parties to block implementation of the Commissions’ recommendations (indeed, their final versions were never published). The legislation remains on the statute book, however, but its details have been questioned: ‘Is there a better way to balance the competing demands of arithmetic, continuity and geography?’  To answer that question, computer simulations of the constituency-building process have been undertaken to explore how House size, arithmetic tolerance and Commission policy interact to shape the map of Parliamentary constituencies.

In reducing the number of MPs, from 650 to 600, the coalition government was less concerned with the impact on boundaries than in reducing the cost of Parliament. But as MPs and others saw seats disappear from the map amid the general disruption, the two issues became somewhat conflated. Surely this reduction had to be part of the cause?  Our research suggests the impact was slight, however. A few more seats might have escaped change had the number of MPs not been altered, but the causes (and possible solutions) of the major disruption were elsewhere.

The imposition of a single electoral quota plus the reduction in the number of MPs meant that the formerly over-represented parts of the United Kingdom would experience larger decreases in their Parliamentary delegations than others: the number of Welsh MPs would decline from 40 to 30 (a 25 per cent reduction) and Scottish MPs from 59 to 52 (a 12 per cent loss); Northern Ireland’s decline was from 18 to 16 (11 per cent) and England’s from 533 to 502 (6 per cent).

The tight +/-5 per cent tolerance around the UK quota meant that it was impossible for the Boundary Commissions to allocate constituencies to individual local authorities in many parts of the country; nor was it feasible to leave large numbers of the existing constituencies either unchanged or only minimally changed. So our research asked:

If the maximum allowed variation around the electoral quota were relaxed somewhat, would that give the Commissions greater scope both to reduce the number of constituencies crossing local authority boundaries and to ensure greater continuity in the pattern of constituency representation after reviews, while still ensuring much greater equality in constituency electorates than heretofore – with either 600 or 650 MPs?

We divided the United Kingdom into 75 areas, comprising either single local authorities or groups of neighbouring small  authorities, each with electorates averaging 400-800,000. A computer program was developed to identify whether it was possible to construct a substantial number of configurations of constituencies within each of those areas, using wards as the building blocks (in line with the Commissions’ general policy), within a size tolerance ranging from +/-5 to +/-12 per cent, inclusive. We concluded that:

  • With a House of Commons comprising 600 MPs, there would be less disruption to the current map of constituencies and a need for fewer constituencies crossing local government boundaries if the tolerance around the electoral quota for individual constituency electorates was relaxed to +/-8 per cent
  • A change in policy on ward-splitting by the Boundary Commission for England would result in a benefit of similar magnitude in terms of continuity but would be less effective in improving the match between Parliamentary constituency and local government boundaries;
  • The greatest benefits would accrue from a relaxed tolerance combined with a more flexible approach to ward-splitting; and
  • With a House of Commons containing 650 MPs similar conclusions apply, though with a marginally better overall match between local authority and Parliamentary constituency boundaries.


The redistributions undertaken by the Boundary Commissions in 2011-2013 were aborted by Parliament for political reasons before their completion, so the 2015 general election will be fought in the current constituencies. But implementation of the 2011 Act was merely delayed until 2016 and a new set of reviews initiated then will, if conducted under the same Rules, be as disruptive to the current map of constituencies as those aborted in 2013.

That next set of reviews will not be able to build on the revised proposals produced by the Commissions in 2012, largely because of uneven population changes across the UK since 2010; already, for example, Scotland is entitled to one more MP than in those recommendations and the south-west of England to one less. Our estimates suggest that both the 2016- review and its successors (the legislation requires redistributions every five years) will be significantly more disruptive than might have been expected and lead us to conclude that:

  • Around one-third of all constituencies are likely to experience major change at each quinquennial review and only about one-third may escape unchanged; and
  • Ward-splitting and a relaxation of the tolerance constraint could reduce the proportion of constituencies experiencing major change at any quinquennial review to below one-fifth and increase those experiencing no change to as much as one-half.

The next scheduled reviews will begin in 2016 and their recommendations submitted to Parliament by October 2018; those constituencies will then be used for the general election scheduled for May 2020. After the 2015 general election, however, there would be time for Parliament to amend the 2011 legislation – if it wished to maintain its principle of greater electoral equality than is currently the case but have less disruption to the constituency map than occurred with the reviews aborted in 2013. It could relax the tolerance around the quota somewhat, or give clearer guidance regarding the use of wards as the building blocks for constituencies – or both. It may also reconsider the size of the House of Commons, but whatever number of MPs is to be elected, removal of the major difference in average constituency electorates across the four UK countries will ensure considerable disruption from the current map – not only at the next but also at subsequent redistributions.

The full report of this research – Equality, Community and Continuity: Reviewing the UK Rules for Constituency Redistributions – is available on our sponsor’s (The McDougall Trust) website.

Note: This article gives the views of the authors, and not the position of the British Politics and Policy blog, nor of the London School of Economics. Please read our comments policy before posting. Featured image credit: John Proctor

About the Authors

ron_johnstonRon Johnston is a Professor in the School of Geographical Sciences at the University of Bristol


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


Charles Pattie thumbnailCharles 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).


Print Friendly, PDF & Email

About the author

Blog Admin

Posted In: Electoral and constitutional reform | Featured

Leave a Reply

Your email address will not be published. Required fields are marked *

Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported
This work by British Politics and Policy at LSE is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported.