The Redistribution of Parliamentary Constituencies (and What It Means)

The 2001 and 2005 general election results convinced the Conservatives that they were treated unfairly by the electoral system. Compared to Labour, they got a much smaller share of the seats than of the votes cast. A major reason for that unfairness, the Conservatives reasoned, was that they tend to win seats with larger-than-average electorates. In contrast, Labour tend to win those with smaller-than average electorates. Because of population movements, this difference – and the subsequent anti-Conservative bias – is exacerbated over time (although research shows that other factors contribute much more to that bias than variations in constituency electorates).

To remove this unfairness, revised Rules for Redistribution are to be implemented by the Boundary Commissions, and these are included in the Parliamentary Voting System and Constituencies Act, 2011. The revisions include:

  • The introduction of a UK-wide electoral quota rather than a separate one for each country (Scotland and, especially, Wales – both Labour heartlands – currently have smaller constituencies relative to England);
  • The requirement that all constituencies (with four exceptions) have electorates within +/-5 per cent of that quota (for the current redistribution this is between 72,810 and 80,473), and only within that range could the Commissions take into account factors such as local authority boundaries, communities of interest and previous constituency boundaries (the previous rules prioritised representation of communities over electoral equality – they made no stipulation regarding the permissible range of constituency electorates, which merely had to be ‘as near the electoral quota as is practicable’);
  • Increase the frequency of redistribution to once every five years (it is currently every 8-12 years); and also
  • Fix the number of constituencies at 600 (currently 650).

The Boundary Commissions published their initial proposals under these new rules in late 2011-early 2012. As anticipated, they incorporated much less continuity in the pattern of constituencies than previous reviews – most existing constituencies were dismembered and many new ones incorporated parts of two, if not three, local authorities.

But the fracturing of the country’s electoral map was much greater than many – including MPs – expected.

In a paper  published by Parliamentary Affairs, we analyse these proposals and show that:

  • The greatest fracturing has been in England’s major urban areas (examples are given from Birmingham, Leeds, London, and Sheffield);
  • This is because the building blocks for constituency-formation (wards) there are too large relative to the task; the average Sheffield ward has 13,804 electors, for example, and Birmingham, 18,294, whereas the range between the largest and smallest allowed constituency electorate is just 7,663. (Haringey is entitled to 2 constituencies but it is impossible using the 19 wards to create two constituencies within the specified range.)
  • In Scotland, the Boundary Commission was prepared to split wards where it considered it impracticable to create constituencies otherwise; with 29 split wards it proposed many constituencies that fitted within local authority boundaries (as in Aberdeen, Edinburgh and Glasgow) – without negative reactions from either the political parties or the general public.
  • The Boundary Commission for England decided to split wards only where ‘exceptional and compelling circumstances’ made it impossible to ‘create constituencies that meet the statutory electorate range without dividing them’. It identified no such cases and rather than split wards it created many seats combining large city wards with smaller ones in surrounding areas.
    • Birmingham, for example, currently has ten constituencies each comprising four of the city’s wards: the Commission proposed seven entirely contained within the city’s boundaries and a further six which each contain one or more Birmingham wards plus smaller wards from neighbouring authorities (in two cases, wards from two authorities are combined with a single Birmingham ward); and
    • In London, 37 of the proposed 68 constituencies comprise wards from two adjacent boroughs (this was the case with only 10 of the 73 constituencies at the previous redistribution); nine boroughs lack even a single constituency comprising wards in that borough alone; and only two have no constituency which does not include wards from a neighbour.
    • In an experiment the authors created alternative sets of constituencies: (a) in Scotland with no split wards; and (b) in metropolitan England with 65 split wards. With ward-splitting the constituencies were both much less disruptive from the previous pattern and involved much less crossing of local authority boundaries: in metropolitan England they halved the amount of disruption compared to the Boundary Commission’s proposals.

If the Commissions’ proposals are implemented – or some variant of them with very similar characteristics (which is very likely) – it will start a process whereby – because a numerical criterion is paramount and geographical criteria secondary – the MPs’ representative role will change. The long tradition that UK MPs represent places and communities will be rapidly eroded; many will just represent numerical aggregates.

Some anticipate that when this exercise is repeated after 2015 there will be much less disruption. However:

  • The allocation of seats across the four countries and England’s nine regions may change, requiring extensive remapping (an extra seat for Northern Ireland could generate significant change to most if not all of its constituencies, for example);
  • With population changes, individual constituency electorates may fall outside the required electorate range and necessitate alterations that ripple through neighbouring areas;
  • Many local authorities are being re-warded by local government commissions; and
  • If the introduction of Individual Electoral Registration is successful and the electoral rolls are more complete, the allocation of seats could change considerably (London, for example, could get up to eight more than the current allocation with consequent losses elsewhere).

In conclusion, the nature of the UK’s Parliamentary representation is being changed as an unintended consequence of a decision to tackle a source of unfairness in the operation of the electoral system – which research has shown to be only a minor cause of the anti-Conservative bias at Labour’s three recent election victories.

David Rossiter, Ron Johnston and Charles Pattie. Access the full paper via Parliamentary Affairs here.

4 Responses to “The Redistribution of Parliamentary Constituencies (and What It Means)”

  1. Mike Killingworth
    July 4, 2012 at 8:14 am #

    It’s perhaps worth mentioning first that the single most important cause of anti-Conservative bias in the electoral system is the electorate itself. Conservatives vote solidaristically (as an expression of their citizeship), Labour supporters instrumentally (to achieve an outcome). This results in the poll in safe Tory seats matching that in marginals, whilst in safe Labour seats it is typically far lower.

    On the “cross-border” issue, all I can report is that from 1997 to 2010 I lived in such a seat (Regent’s Park and North Kensington) and am now back in Westminster North. I’ve never heard anyone round here complain about Parliamentary boundaries, and I believe both parties were pretty happy with them both before and after the change. And so they should be: the breaking of local links suits party machines.

  2. Líam (@doktorb)
    July 7, 2012 at 11:16 am #

    I don’t agree with the conclusion that the consequence will be an MP will represent aggregates rather than population centres.

    Inevitably boundaries will wrap around towns much wider than they ever did before, and stretch across the countryside wider than they did before. That’s not an unnatural development. Dividing towns and cities in half is unnatural, something which has been protested against in the review process and should be ‘worked out’ by the next review in five years time.

    The in-built bias in the system will always be towards Labour. This is why Labour chose to campaign against AV, because FPTP and small constituencies tightly bound around city centres always delivered massive Labour majorities.

    It’ll always be shameful that Labour chose to campaign against AV. It’ll be interesting to see if they decide to vote with backbench Conservatives to defeat the boundary change vote, therefore keeping both their inbuilt parliamentary advantage.

  3. RedShift
    July 8, 2012 at 10:24 am #

    In response to Liam:-

    Labour didn’t campaign against AV. Instead they spent the time campaigning to kick Tory and Lib Dem councillors out of their seats – very sensibly in my humble opinion.

    There were both pro-AV and anti-AV campaigns from Labour Party members that were not officially endorsed campaigns

  4. Duncan Stott (@DuncanStott)
    January 14, 2013 at 3:43 pm #

    @RedShift

    Labour No to AV spent £192,084 during the referendum campaign.
    Labour Yes spent £0.

    http://www.electoralcommission.org.uk/party-finance/party-finance-analysis/referendum-expenditure/2011-parliamentary-voting-system-referendum#spending

Leave a Reply