Parliamentary Constituencies Bill

Written evidence submitted by Mr John Bryant (PCB02)

Parliamentary Constituencies Bill

Submission to the Public Bill Committee of the House of Commons

Key proposal

The opportunity should be taken to amend the Parliamentary Constituencies Act 1986 so that proposed seats are required to fall within 7.5% either side of the electoral quota instead of 5% as at present.


This submission relates to the Parliamentary Constituencies Bill currently before Parliament. The submission addresses the tolerance either side of the electoral quota within which proposed constituencies are required to fall.

Although I am employed as a policy advisor by a national trade body, I should make it clear that I am making this submission in an individual capacity as a member of the public with a long-standing interest in this subject. This same interest led me to make submissions to the 2018 review, particularly regarding constituencies in London, and I gave evidence in person at the public hearing at Romford. In its final report for London the Boundary Commission adopted my suggestions across quite wide parts of Greater London and made several favourable references to my submission in the narrative report.

My previous involvement was with the review process in England and this submission reflects that experience. Although most of the points will be equally applicable to other parts of the UK, I acknowledge there will be some differences owing to local factors, for instance the fact that the voting system used for local elections in Scotland means that wards tend to be much larger than in England.

I understand that the Committee may publish submissions.

650 MPs rather than 600: Implications for defining boundaries

The most important change in the current Bill is that the number of constituencies in the UK will remain at 650 rather than be cut to 600 as was provided in the legislation governing the 2018 review. I do not wish to make any representation one way or the other about the overall size number of the House of Commons, but I do wish to point out that any material change in the number of seats – in this case, an increase of 8.3% compared with the 2018 review – will have a significant impact on the mechanics of boundary-drawing.

In the case of the 2018 review, and on the assumption that local authority wards should be treated as the building-blocks of which Parliamentary constituencies are composed, the 5% tolerance imposed a severe restraint on the way boundaries could be drawn. But despite this restriction, it proved possible in all cases, although sometimes only with difficulty, to create a coherent and viable pattern of constituencies with no need to resort to splitting wards. (It is true that the BCE did, in its final reports, recommend a very small number of ward splits; however, in my view, none of these was strictly necessary under the rules).

Obviously a reasonable equality between electorates in a highly desirable feature of any representative system; while on the other hand, a degree of flexibility is needed to allow natural and coherent boundaries to be drawn. It must be a matter of judgment where to strike the balance between equality and flexibility.

For example, in 2015 the Parliamentary and Constitutional Reform Committee of the House of Commons considered the matter and, having heard evidence from, among others, Mr Tony Bellringer of the Boundary Commission for England, recommended an increase to 10%: (see paragraph 40 for the specific recommendation).

With respect to the Committee and to Mr Bellringer, however, I submit that a 10% tolerance would result in serious inequalities. Had such a tolerance applied at the 2018 review, for instance, with an electoral quota of 74769, permitted electorates would have ranged from 67293 to 82245. A difference of nearly fifteen thousand electors is surely too much. In my view, despite the difficulties to which it occasionally gave rise, the tolerance of 5%, giving a range of 71031 to 78507, was set at an appropriate level in the context of the 2018 review.

However, the increase now proposed in the number of seats means of course that their average size will be reduced by the corresponding proportion of 8.3%. Given that it was far from straightforward, in the 2018 review, to produce sensible boundaries without ward splits, it is reasonable to ask whether the effect of significantly reducing the size of seats will be to render impossible a task that was previously merely challenging.

Without having access to the electorate figures that will be used for the new review, it is not possible to answer this question conclusively. What can be done, however, is to ‘war-game’ the process in a highly realistic manner by applying a 650-seat approach to the wards and electorates used in the 2018 review.

Clearly, the most difficult areas are those with relatively large wards. I have therefore used London as a test case and have attempted to map 74 constituencies (its likely allocation with 650 seats, compared with 68 in the 2018 review). My efforts suggest that it is not possible to produce a sensible outcome without significant numbers of ward splits. I anticipate that these difficulties will be mirrored in other areas with large wards, Leeds and Sheffield being the most obvious cases (but there will be others). This leaves a choice between easing the tolerance or accepting substantial numbers of ward splits, especially in built-up areas with large wards.

Reluctantly, therefore (because, as mentioned above, I supported the 5% tolerance at the previous review), I urge the Committee to consider amending the Bill to allow a somewhat greater tolerance in constituency sizes. 10% would be too much, since it would allow constituencies to vary by some fourteen thousand electors, so I suggest that the tolerance should be increased to 7.5%. Even this would allow a variation of over ten thousand, but this is unavoidable if the local authority ward is to remain the basis for drawing Parliamentary boundaries.

I therefore suggest that the Bill should amend paragraph 2(1) of Schedule 2 to the Parliamentary Constituencies Act 1986 so that it reads, "The electorate of any constituency shall be no less than 92.5% of the United Kingdom electoral quota, and no more than 107.5% of that quota." This would not affect paragraph 2(2), which exempts protected constituencies and very extensive constituencies from the usual quota.

Why wards?

Some commentators with an interest in these matters have criticized the use of wards as basic units and have urged the Boundary Commissions to show greater willingness to divide or even disregard wards. I should like to add a note explaining why I feel that it is important to use ward boundaries wherever possible.

The use of a pre-existing unit such as a local authority ward assists the boundary-drawing process by reducing the range of available options. The explanation of this seeming paradox is that if the line may be placed anywhere, the range of options becomes virtually infinite. We see this in the United States, where (at least in some states) lines are drawn with scant regard for other political or administrative boundaries. Only a computer with detailed mapping software can handle all the possible permutations, which effectively shuts out the general public from any meaningful engagement with the process. This approach enables the grotesquely distorted gerrymanders seen in states like Illinois, Maryland and Texas, which would be impossible to achieve if electoral districts had to be composed of pre-existing smaller units. (Governor Elbridge Gerry of Massachusetts, credited with the (relatively innocuous) prototype back in 1812, would surely blush at some of the modern creations perpetuating his name.)

A requirement to use existing lines wherever possible not only makes the process simpler and more accessible; by reducing the number of options available under the rules it lessens the scope for disputes and the pursuit of partisan advantage.

In theory it would be possible to use other pre-existing boundaries such as parishes (where they exist) or even postal districts. Wards, however, already have an electoral function, have precisely-defined boundaries, and cover the entire country, with each ward lying wholly within a particular local authority. Moreover, the electorates are readily available.

It follows that I respectfully differ from the views expressed by Mr Bellringer in his evidence to the Parliamentary Reform Committee, in which he said (see paragraph 45) that although the BCE would seek not to split wards, the presumption against doing so would be less strong than in the past. I should advise the opposite approach: if a ward split is recommended the onus should be on the Commission to show that no reasonable arrangement is otherwise possible.

If the division of a ward cannot be avoided, the split should use polling district boundaries (although this is inherently less satisfactory as polling districts are drawn purely for administrative electoral convenience and (unlike wards) do not serve any representative function). However, it is hoped that the suggested easing of the tolerance to 7.5% will avert the necessity for any ward splits.


While equality in electoral districts is highly desirable, a degree of variation is essential to allow flexibility. How to strike this balance depends on the size of constituencies relative to the smaller units, i.e. local authority wards, of which they are to be composed. The retention of 650 MPs, instead of the reduction to 600 previously proposed, reduces the average size of constituencies by 8.3% and means that the previous tolerance of 5% would be likely to require numerous ward splits. To avoid this, the tolerance should be eased to 7.5%.

John Bryant

15 June 2020


Prepared 18th June 2020