For the third time in just over a decade, a new map of parliamentary constituencies is being designed. This one will likely be implemented. Charles Pattie and David Rossiter argue that, despite the misconceptions of both Labour and the Conservatives, the review is neither a ‘gerrymander’ against one, nor redressing an imbalance that harmed the other. But these entrenched views could yet threaten the future of First Past the Post as the system for Westminster elections.
Here we go again. For the third time since 2010, a new map of Westminster parliamentary constituencies is being designed. The Boundary Commission for England released its preliminary proposals on 8 June (the Commissions for Scotland, Wales and Northern Ireland will follow suit in the coming months). Final recommendations will appear in the summer of 2023. This time (the previous two attempts at redistricting faltered before being implemented) the new map is very likely to be adopted. And if past reviews are any guide, the process will be carried out amidst claims and counterclaims regarding potential winners and losers, and whether there is deliberate bias in the process.
Of course, redrawing the constituency map inevitably involves winners and losers, even when (as in the UK) done by politically impartial Commissioners. Previous reviews have tended to result in relative losses of seats for Labour and gains for the Conservatives (smaller parties tend to suffer greater disadvantages from the disproportional nature of First Past the Post (FPTP) than from the effects of boundary reviews). Some Labour figures are likely to argue (as they have done in the past) that the review is a gerrymander against their party, and so drives a nail into the coffin of its electoral chances. On the other side some Conservatives will argue the review simply redresses substantial anti-Conservative bias in the old seats – a nail in the coffin in which that bias is to be buried.
Both views are wrong, but for different reasons.
Boundary reviews are not anti-Labour gerrymanders. It is true that, for much of the post-war period, Labour tended to lose, and the Conservatives to gain, seats as a result of redistricting. But that was because of the ways in which the UK’s changing population geography interacted with the geography of party support. Labour-voting areas tended to lose population relative to Conservative-voting ones. As constituencies aged, therefore, electorates in Labour seats tended to become smaller on average, while those in Conservative seats became larger. But this helped Labour, as it meant fewer votes were needed to elect Labour MPs than to elect Conservatives. Far from representing an anti-Labour gerrymander, therefore, the tendency for Labour to lose seats as a result of redistricting (which tends to reduce disparities in electorates between seats) was actually redressing an unfair pro-Labour advantage which grew as seats aged.
The Conservative argument that revising the constituency map to take into account population change redresses biases against them in the electoral system – of which electoral boundaries is just one of several – was true in the past, partly because of the shifts in population over time noted above, but also because Scotland and Wales – both traditional Labour strongholds – were overrepresented in Westminster. But it has become much less true in recent years. The long-established tendency for inner-city areas to lose population relative to suburbs and exurban areas (which depleted electorates in safe Labour seats) has not been as intense in recent years. And since 2005 (when, to take account of devolution, the country’s number of Westminster MPs was reduced to remove its overrepresentation), Labour no longer receives a bonus from its popularity in Scotland. Since 2015, Labour has been eclipsed there by the SNP, removing any residual advantage. And the Conservatives have made inroads into former Labour seats in the Midlands and the North of England. The most dramatic manifestation of this was the collapse of the so-called ‘Red Wall’ in the 2019 election. Many of the smaller constituencies which traditionally returned Labour MPs are now Conservative.
So equalising electorates is now less of a disadvantage for Labour than it used to be, and less of an advantage for the Conservatives. Boundary reviews are not an exercise in anti-Labour gerrymandering. The electoral system is no longer biased against the Conservatives, even in aging seats (quite the opposite, in fact). The main source of bias in elections over the last 30 years comes not from constituency size effects but from vote efficiency (and abstention). And it is no longer inevitable that reviews will tend to favour the Conservatives.
That said, the current review will undoubtedly result in a very substantial redrawing of the UK’s electoral map. And not before time. The current constituencies are based on the electorate from around 20 years ago, and there are now some substantial disparities in constituency electorates. The boundary review currently under way uses the March 2020 electorate as the basis for redrawing constituencies. By then, the average UK constituency contained 73,167 electors. But this ranged from just 42,657 in Arfon to 99,859 in Bristol West (the electorates in the Na h-Eileanan an lar, Orkney and Shetland, and Isle of Wight constituencies were even more extreme, but these seats have protected status under the current boundary review legislation). Under the legislation governing the current review (updated most recently in the Parliamentary Constituencies Act 2020), a boundary review is required to create 650 (no more, no less) constituencies, all (bar 5 exceptions) with electorates within ±5% of the average on the enumeration date for the review. In other words – the exempted seats aside – every seat produced by the current review must contain no fewer than 69,724 and no more than 77,062 registered electors. At present 415 (or nearly 64%) of existing seats have March 2020 electorates outside that range.
Our first indication of just how disruptive this is likely to be is contained in the Boundary Commission for England’s initial proposals. One consequence is that the same rules on constituency size now operate in every part of the UK. The biggest loser will be Wales, currently substantially overrepresented with 40 MPs: after the review its complement will drop by eight, to 32. Scotland will lose two MPs, down from 59 to 57. Northern Ireland’s representation, meanwhile, remains unchanged. And, as the size of the Commons is now fixed at 650 MPs, the size of England’s contingent will increase by 10, from 533 to 543.
At present, 324 of England’s 533 constituencies have electorates outside the target range. But that does not mean that the remaining 209 seats (just under 40% of the total) can be left unchanged while the remainder are revised to both add the extra seats needed and to bring all seats inside the target. Because the Boundary Commission for England uses local government wards as its main building blocks, and because wards tend to come in unwieldy sizes, many of the seats currently in target will have to be changed too, in order to make space for changes elsewhere (the English Commission is open to splitting wards, and indeed has proposed doing so in some areas, but it has done this as little as possible). Based on its own figures, only 43 of the 543 seats described in its initial proposals are unchanged from the current constituencies, while a further 80 are changed only very slightly to take into account revisions to local ward boundaries since the last review. In other words, 77% of the seats named in the initial proposals are noticeably different from the constituencies currently in use. And many of these changed constituencies are replacing some of the 209 current constituencies with electorates which are in range, just under half of which are earmarked for some non-trivial change.
This is, of course, just the start of the process. The initial proposals now go out for public consultation and will be revised. Even so, the eventual map will be substantially different from the pre-review map. And, because the new rules on equalising electorates substantially reduce the Commission’s ability to exercise discretion in order to minimise disruption, this is not likely to be a one-off. The level of dislocation produced by future reviews is likely to be as large as we will see this time. Major disruption to the constituency map will be the new normal.
The Boundary Commissions have a fairly thankless task on their hands. They must (and do) operate strictly within the relevant legislation, which makes achieving equality of electorates (within the ±5% margin) their top priority: only after that has been done can they begin to think about the effects of their suggestions on community representation. But they face objections from voters concerned that their community is split across constituencies, from local MPs who may find a seat they have represented assiduously has changed beyond recognition, and from political parties which are concerned with the electoral consequences of the changes.
The political parties’ views of the reasons behind redistricting are not necessarily accurate, but they are entrenched. Many Conservatives believe they benefit from tightly circumscribed and rigidly applied equality of electorates across constituencies. Many on the Labour side feel they are bound to lose out from this. So where next?
In the past, both the Labour and Conservative parties supported the use of FPTP for Westminster elections. The reasons are not noble or hard to find: they supported a system which seemed to guarantee that one or other would be likely to form a majority government after most elections. But those links are potentially loosening. Since losing many of its Scottish seats, Labour faces a larger uphill struggle to win under FPTP. And the emphasis on strict electoral equality across constituencies in the current redistricting legislation is likely to further erode Labour support for FPTP, in the belief (even though no longer as true as it once was) that this benefits their rival. Support for electoral reform is likely to grow within Labour ranks as a result. In opposition in the early 1990s, Neil Kinnock and Tony Blair both explored electoral reform (and a potential ‘progressive alliance’ with other broadly like-minded parties) as a possible solution to a semi-permanent Labour lock-out under FPTP. Having won a huge majority in 1997, Blair did not follow through on the idea. But things may be changing. The chances of a Blair-style Labour landslide in the future look remote, increasing the chances that Labour might commit to a more proportional electoral system. If it did so and managed, either on its own or in coalition, to secure power, then it could be that any discussion of nails and coffins could yet apply to FPTP itself.
About the authors
David Rossiter is a former Research Fellow who worked at the Universities of Bristol, Leeds, Oxford and Sheffield.
Charles Pattie is Professor of Politics at the University of Sheffield.
The authors have written together on the Boundary Commissions and constituency definition in the UK since the early 1980s and with Ron Johnston co-authored The Boundary Commissions: Redrawing the UK’s Map of Parliamentary Constituencies (1999) and Representative Democracy? Geography and the British Electoral System (2021: both Manchester University Press