Q Colleagues, as I mentioned earlier, the third witness is no longer able to appear, so we now move to our final witnesses for the morning session: Professor Iain McLean who, again wonderfully, is here physically; and Professor Sir John Curtice, who is appearing virtually. Thank you both for being with us earlier than anticipated. Will you please introduce yourselves?
Thank you, Chair. I am Iain McLean, professor of politics at Oxford University. John—as he will say in a moment—and I are academics who have been working in this area for decades. I have been a witness at various boundary inquiries, at the fourth and fifth English reviews, never on behalf of political parties, but always on behalf of local authorities. I have published academic papers pointing out that the former rules were mutually contradictory. That was fixed in the Fixed-term Parliaments Act 2011, and it is important that the Bill should not unfix it. I will leave it there.
Professor Sir John Curtice:
I am John Curtice, professor of politics at the University of Strathclyde. I have written, as Ian was implying, for about 40 years on the way in which the single member plurality electoral system works in the UK and the way in which the geography affects and has changed its operation in the post-war period. That therefore meets my interest in this area, which has been rather more to do with political ramifications of the commissions’ work and the boundary redrawing rather than some of the more technical side, on which you will find Iain much more expert than I, but I am more than happy to share my observations from the stats in which I am interested. I have written about how the electoral systems operate in virtually every election since 1979.
Q Thank you very much indeed, Sir David. Could we have anything better than more time with Professor Maclean and Professor Sir John? This really is a treat—thank you both very much for joining us. Given that we have a little more time, I would like to start with a question to each of you, although I am sure your paths may cross over as the session goes on.
Professor McLean, you began in your introduction by referring to the rules having been put right in the earlier Bill and said that you would not change them again. Could you go into a little more detail on that? I am taking you to mean the rules that we find in schedule 2 to the Parliamentary Constituencies Act 1986, which, as you will know, the Bill predominantly leaves unchanged. We—perhaps like you—think that they flexible enough to allow the commissions to do their work, but perhaps you could elaborate on that. If I may, I would then like to ask Sir John a question once Iain has had a chance to speak.
The rules, as originally drafted in 1986, were mutually contradictory. Rule 1 said that you should not expand the size of the House of Commons, and there was an equality rule, the unintended effect of which, as it was then written, was to tend to increase the size of the House of Commons after each review, for mathematical reasons that I hope I do not have to go into now, although I can.
They are now expired because two things in the 2011 Act fixed that problem. It gave total priority to a fixed number of seats in the House of Commons, and because that overrules everything else in schedule 2 to the 1986 Act, the creeping enlargement of the House of Commons, which some people thought a problem, is no longer a problem. Secondly, within the other rules, the 2011 Act amends the 1986 Act by giving equality of constituency size priority over the other criteria, including local ties and respect for local government boundaries. Once that priority has been set—I am speaking mathematically, not politically—the contradictions in schedule 2 as it originally operated have disappeared.
I have looked at—with some difficulty during lockdown—the text of the Bill and I have it and the explanatory notes in front of me, via a rather dodgy connection to my iPhone. I have looked rather nerdily at the proposed amendments to the vital schedule 2 to the 1986 Act. From my reading—though I am not a lawyer—I would say that they do not upset the changes that were made in 2011 and, therefore, they should be left as they are. I think that will do at the technical level, although the Committee may have further questions.
Q Thank you very much indeed, Professor McLean. I do not mean to take the role of the Clerk, but I think that I can say that, if it helps, you can take a copy of the Bill and the explanatory notes from the table just behind Mr Efford.
Thank you for that helpful explanation. To clarify it further, do you think that rule 5(1), the list of factors, does a good enough job of providing flexibility to the boundary commissions, given its place in the hierarchy of rules that you have just gone through?
Q Thank you very much indeed.
Sir John, thank you very much for joining us. I wonder whether you might be able to help us with our understanding of the data used for boundary reviews. They are based on electoral registration data; could you give us your views on the adequacy of that?
Professor Sir John Curtice:
The short answer is that over the long run, from the various exercises—most recently by the Electoral Commission, and before the commission was created, by the Office for National Statistics—that have looked at the accuracy and completeness of the electoral register, we know that there are inadequacies in the register that have increased over time. Those inaccuracies are also related to certain circumstances such as having recently moved house, living in private accommodation or being unemployed. The Electoral Commission’s most recent report, for the December 2018 registers, said that they were 85% complete, meaning that only 85% of those people who should be on the register are on it, and 89% accurate, meaning that about 11% of entries relate to people who should not be on the register at the place that they are at.
The Bill makes no difference at all for all practical purposes to the rules for redistribution that were passed in the 2011 Act, but that Act places a premium on allocating constituencies with respect to electorates. We know that those electorates are less than perfect; I guess that if we are really now concerned about the mathematical accuracy of boundaries, what we should probably be worrying about is not the rules for redistribution, but ensuring that those rules are implemented more effectively by improving the accuracy of the electoral register. But that is a long-running problem, and I am not trying to argue that it will be easy to resolve.
Q Indeed; there is always discussion to be had about how we can continue to improve the completeness and accuracy of the registers. Not that I would get into an argument with you about trends over time, but my understanding was that those are rising rather than declining—but as you say, that is a different discussion.
Looking at electoral registration data with its ins and outs, as you have just outlined, is it the right kind of data to base boundaries on—as opposed to census data, for example, or other kinds that you could conceive of being collected?
Professor Sir John Curtice:
The problem with census data, obviously, is that it is now nearly 10 years out of date. You might want to argue that the ONS produces a mid-year population estimate over time, but it does not necessarily have the detail required to set up boundaries.
The second problem is that there is a disjuncture between residency and citizenship. If you went in the same direction as the Scottish Government by giving anybody who is permanently resident in the United Kingdom the right to vote, you might want to consider population as a reasonable proxy for that. However, as long as we are going to limit the franchise to British, Irish and Commonwealth citizens, given that this country has a substantial resident non-citizen population, you are probably not going to want to go down the route of using population. That, again, is tied up with the issue of the franchise.
Q Understood. Thank you for those opening remarks; that is helpful. Maybe Iain has had a chance to think about the other question I left hanging with him.
Thank you Minister, and thank you Chair, for your forbearance. It is quite a jigsaw puzzle, but on page seven of the Bill are what you call “Minor and consequential amendments”. That is a mistaken heading; one of them is neither minor nor consequential. I will not comment on the addition of the county of Blackpool in paragraph 4 of the schedule; the only material amendment here is in paragraph 4(2): “for ‘596’ substitute ‘646’.” As Members know, that is one of the consequences of keeping the House of Commons’ size at 650. The number 646 appears in the paragraph because of the four reserved constituencies, which are islands exempted from the equality criterion. That is all good. What is not in here are the changes to the schedule of the 1986 Act introduced by the 2011 Act. I was in a position to check that yesterday.
Order. I am slightly embarrassed, but I have to share with the Committee that the Bill available in the room is the wrong Bill. Quite how that has happened, I do not know. The Clerks will make sure that the right Bill is available for the next sitting. I was completely unaware of that, and unfortunately there is nothing I can do about it, I am afraid. It is a pity. Professor McLean, one of the Committee members will get the right Bill; it is on its way, and everyone will have the copies.
My view is that it does not really affect that materially, but I felt that I should place on record the fact that the Bill that we had was not the right one.
It turns out that I always was; my document is the correct Bill. To reiterate, for those who are looking at the correct one, paragraph 4 of the schedule to the Bill, “Minor and Consequential Amendments”, addresses schedule 2 to the 1986 Act. That is the one that does all the work. The only material change that is introduced is one of the consequences of keeping the size of the House at 650 Members; after subtracting the four protected constituencies, that is 646. This ensures that the House’s size continues to be fixed absolutely. That removes one of the sources of the incoherence of the schedule as originally drafted.
The other source of the incoherence was that the electoral equality criterion, until the 2011 Act, had no priority over the local ties and local government boundaries criteria. Amendments to the 2011 Act, which is not further changed and is therefore not in front of you here, gave the equality criterion priority over the local ties and local government boundaries criteria. That remains unchanged by the Bill. Ministers and parliamentary drafters have not, therefore, by any mistake reintroduced any of the inconsistencies in the original 1986 Bill. I hope that that is sufficiently clear to Members, but I can expand further if people wish.
Q I direct my question to Professor Sir John Curtice. I would like to ask about the mathematical accuracy of the boundaries that we are drawing up. Obviously, I do not think anyone would disagree that we would like constituencies to be as equally represented as possible, but I have quite a lot of concern that the data that we are using is not accurate, because the electoral register, as you said in your previous answer, is about 85% complete. A huge proportion of people are missing from electoral registers. Can you see any opportunity in the Bill, Sir John, to increase the accuracy of the data that we are working from? Do you have an opinion about the best source of data to use in drawing up constituencies, so that they could be sized most accurately?
Professor Sir John Curtice:
The short answer is that the Bill is not concerned with the process of electoral administration. The process of electoral registration deals with electoral administration. As Professor McLean has just pointed out, frankly the Bill does nothing material to change the rules on redistribution, including on the basis on which the electorate is used to do that. I simply pointed out in my response to the Minister that there are limitations to the data. We know that those limitations are somewhat greater in, for example, inner-city constituencies with a highly mobile population, than in constituencies with lots of older voters and a more stable population. That, undoubtedly, is correlated to some degree with the political proclivity of constituencies.
As I indicated earlier, as long as we wish to make a distinction between permanent residence and the right to vote, and as long as we do not wish to have a national identity card system, it is difficult to think of an alternative to the system we have. The question therefore is whether there are ways of improving the accuracy of the register. One thing we can note is that although we moved from household registration to individual registration—a somewhat controversial move—it is not obvious that it has fundamentally changed the character of the problem before us.
Q For the purpose of drawing boundaries, where it is most important that there be as much accuracy as possible, would combining other sources of information be a way of improving the accuracy of the electoral roll? Perhaps, for example, data held by the Department for Work and Pensions could be added to that on a register, to ensure that it was more accurate. Obviously, that would not be applied in elections; it would just be for the purpose of drawing constituency boundaries, so that the original data source could be made more accurate.
Professor Sir John Curtice:
The answer to that question, to be honest, is technically beyond my competence, in the sense that I guess the question that the boundary commissioners would ask is whether it is possible to get DWP data—which refers to the right to work, not necessarily to the right to vote—at the level of local government wards, which are the principal building block used by the boundary commissioners in building parliamentary constituencies. I would not be surprised to be told that the answer is no, but I do not know. Again, DWP data might rely on whether people have a national insurance record, but that is not the same thing as citizenship.
Q I will direct a question to Professor McLean, who I hope now has the right Bill in front of him. Back in 2010, Professor, you wrote an article for The Guardian about the boundary review commencing then. I was interested to see that in it, you progressed the argument that the most accurate way to ensure that every vote counted equally would be to move towards proportional representation. That is outside the scope of the Bill, but it shows up the conflict that we have. Would not the way for every vote in the United Kingdom to count equally be to have just one constituency—the United Kingdom—and a system of proportional representation, even though that comes into conflict with the communities that we represent? Ultimately, if we are to maintain the constituency link, we have to have a percentage variance between seats; we cannot have every single seat with exactly the same number of electors. It is a question of where we draw the line.
How can that balance be struck? Is the 5% tolerance most appropriate, or if we are not moving towards a system of proportional representation, should there be a larger tolerance, so that community ties are considered more important?
For clarity, it is important to separate the question of proportional representation from that of the 5% tolerance, because they are different questions. As I evidently said in 2010—you have better recall of what I said than I do—a single-member district system cannot be proportional. That is a mathematical truth. Legislators must make a choice, and the choice that the UK Parliament has made is reflected in this Bill and many others: the single-member district system.
I do not think that it would be a good use of this Committee’s time to talk about whether the UK should switch to proportional representation; with your permission, Chair, I would rather duck that part of the Member’s question.
On equality, the Member poses an important question: is it correct that the equality criterion should override the other ones—the ones on local ties, and on the constituency boundaries following local government ones where possible? My view, which is an arithmetical view, not a political one, is that it is right for the equality criterion to override the others.
Becoming somewhat more political, my observation of boundary inquiries is that since local ties are not further defined in the Act, I have observed on several occasions that for a number of very shrewd operators, who will be well known to members of this Committee, Conservative local ties go one way, Labour local ties go another, and Liberal Democrat local ties go yet another. Each of them, because they are paid to do so, makes a plausible case before a commissioner, who in England is deliberately chosen not to be from the area. Moving on from the mathematics, my view as a political scientist is that the local ties criterion is eminently manipulable, whereas the plus or minus 5% criterion is not.
Is the criterion wide enough? In the United States the courts have said that as near as possible to 0%— not 5%—is the accepted tolerance for US congressional districts. So, it is possible to have a tolerance lower than 5%, but that is not in this Bill and it is not in the earlier Acts.
Q Gentlemen, thank you for giving your time today. As you have probably picked up from reading previous reports, one of the issues this evidence inquiry is trying to get to the bottom of is how we are going to advise the commissions about the best way to do these boundaries.
Building on what you have just said, Professor McLean, about keeping the right size and in terms of communities, about which one can always argue, can we look at rule 5(1)(c) in the 1986 Act, which is about keeping boundaries in existing constituencies? My question, to both witnesses, is about whether the Bill needs to have some clarifications put in it, especially around what we are struggling with regarding the Boundary Commission for England. The evidence from the Boundary Commission for England was pretty much, “We are always going to try and do it with wards, and we will just get the numbers to work.” That overrides almost all the rules in clause 5, including geographic considerations. I gave the example of a North Yorkshire ward that one can only get to by completely leaving the constituency and spending a considerable amount of time on the road, but it would make the numbers work.
Can I probe your minds on the resistance to building outside of the wards, or, in other words, splitting wards down, as they do in Scotland, in order to try to keep existing communities together? What are your views on the different definitions of county constituencies and borough constituencies? How does that play into the building of constituencies? Does the Bill need further guidance to try to equalise the United Kingdom’s approach to how it builds constituencies, with the gold standard of Scotland being a good example?
If John is willing, I will go first, but John will wish to add something about the practicalities of the Boundary Commission for Scotland, which he has written about in academic articles.
The presumption against disturbing existing constituencies is no longer sustainable because these are based on electorates in 2000. Population movements, in what will be 24 years before the new constituencies are implemented, will make it impossible, in more than the odd coincidental case, to give any priority to the maintaining of existing constituencies preference. I think 5% plus or minus should be enough for the boundary commissions and the county-by-county inquiries to deal with difficult situations, such as the one the Member mentioned of a large, empty area in the middle of a constituency. I take it that that is the geographical problem that the Member mentions.
There are other well-known problems of estuaries, such as the problems in the Wirral area last time. Plus or minus 5% should be enough to cope with that. At the risk of sounding like a stuck record, I think it is right that in the 2011 Act, which this Bill importantly does not modify, the plus or minus 5% is given priority over the other local ties rules.
As to whether local government wards are the essential building blocks, that is non-statutory. It is the practice of the English commission, but it has not been the practice of the Scottish commission. I will now hand the floor to John to answer that part of the question.
Professor Sir John Curtice:
There is a crucial difference these days between local government wards in Scotland and those in England. Scottish local government is run under the single transferable vote in multiple constituencies system. When that system was introduced, it was introduced without changing the number of local government councillors significantly. All the wards elect three or four members. As a result, every ward in Scotland was increased by three or four. That means, therefore, that the building blocks in Scotland are large, making it difficult for the Boundary Commission to respect more badges. There are one or two instances in England, such as Birmingham, where that issue can also arise, but it is relatively limited.
It is also true—this is not the area of my own expertise—that some entrepreneurial past secretaries of the Boundary Commission for Scotland have ensured that the Boundary Commission has a much better geographically-referenced database than the one in England. I was reading some of the evidence given to the Committee last week and that came out. I am tempted to say that that is one of the advantages of living in a small country: it becomes possible to administer things in finer detail. We have referred to county and borough constituencies. That only relates to the rules for expenditure. It does not otherwise make a great deal of difference.
Beyond that, I simply observe that in this conversation and this morning, and in much of what the Committee seemed to be talking about last night, seems to be about what this Bill is not about, as opposed to what it is about. The Bill does not fundamentally change the rules of redistribution that were introduced by the 2011 Act and implemented by the Boundary Commissions in their 2013 and 2018 reviews—sadly, neither of which were implemented. Apart from changing the number of MPs, it does nothing to change that—apart from a minor and perfectly sensible change with the rules about respected local government boundaries. I suggest that at some point the Committee might want to focus on the significant changes the Bill does introduce as opposed to the areas that the Bill does not propose to change at all. I understand, of course, that some Members may wish to unpick the provisions of the 2011 Act.
Q One reason we are probing how constituencies are built is because there is removal of parliamentary oversight. It needs to be done properly the first time. You rightly referenced the size of the wards in Birmingham. I am a West Yorkshire MP. There are two councils in West Yorkshire, Kirklees and Leeds—out of the five councils—where the wards are far too big not to be split.
This comes down to guidance. As you pointed out, the large wards and the way they are managed in Scotland has allowed a more detailed approach. When you get to the arguments of whether it should be plus or minus 10% or 5%, I am seeking your view as to whether the arguments about the variations can be overcome by the guidance, which goes more explicitly to the Boundary Commission for England in splitting wards.
In the past, there has been a habit of them trying to form some strange shapes, like American congressional districts, just to get the numbers right, forming very strange communities. They have almost always then changed the first draft significantly in the second draft. The guidance that will go in this Bill, especially for the Boundary Commission for England, should try to avoid that situation.
The parliamentary oversight is going, which I believe is the correct thing to do. But we must get this right the first time and use this Bill to iron out these issues. Is this Bill strong enough, in terms of the Boundary Commission for England, to construct constituencies, which have an eye to what has gone on in the past, but do not end up with peculiar shapes and communities just to make the numbers work?
Professor Sir John Curtice:
Can I respond to that? It is true that the current arrangements for parliamentary oversight do not make it very easy for the House of Commons to change the detail of the provisions. It basically has to say yes or no, and only after it has said no can the Government attempt to change the provisions of the Commission. That is the first point; otherwise, it is a guess on my part, but I would anticipate that now we are going to a House of 650 seats rather than one of 600, some of the difficulties with supposedly major constituencies may be less sharp.
The final thing to say is that even with us going for 650 seats rather than 600, the next boundary revision is bound to be a major one. Because Parliament has blocked both of the last two redistributions that it ordered, we now have boundaries that are 20 years out of date. We are also finally getting around to dealing with the differences in the allocation of constituencies to England, Scotland and Wales, so this is bound to be a disruptive redistribution. It will be somewhat less disruptive than it would have been with 600 seats, but it is bound to be disruptive, in much the same way as the one that was introduced in 1983, because that got affected by the direction of local government.
You might want to investigate the forces that have resulted in boundaries going out of date—that is, population movements, which historically for most of the post-war period meant people moving out of the inner city into more suburban and rural areas. The last analysis of this I read, which was by the expert Tony Champion, indicates that this has been going on to a lesser extent; it is notable that somewhere like London is now gaining population and is certainly not going to lose out from the current redistribution. Of course, nobody knows what is going to happen in the wake of the pandemic, but it is worth being aware that some of the demographic forces that have given rise to the kinds of inequalities we have been used to may no longer have quite the same force as in the past.
If time permits, Chair, may I come in on part of the Member’s question, which was to do with whether the guidance in the Bill should be more explicit than this current draft? My view is no, for the following reasons.
The legislation is UK-wide, as you all know. As this discussion has revealed, the English and Scottish—and, may I say, Northern Irish—commissions have all taken different approaches to the local government boundary question. Those different approaches are all legitimate within the text of the Act that this Bill amends, and it does not amend that Act in any material way. Therefore, I do not think there is any need to give guidance to the Boundary Commission for England that, if it wishes, it can be more flexible in Birmingham and West Yorkshire than its predecessors have been. It already has that discretion; that discretion is exercised by the Boundary Commission for Scotland, and to pick up a point of John’s, if at the last review the Boundary Commission for England had invested in geographic information systems that were as up to date as the Scottish commission’s, some of the problems that the Member mentioned—which I know concern a lot of Members—could have been avoided. My view is that as the existing statutory framework gives the commission the authority to ignore local government boundaries if it has to, there is no need to change the draft Bill in that respect.
Professor Sir John Curtice:
The research on this goes back quite a way, and the answer is “to a degree”. For the purposes of answering this question, I will go back 20 years psephologically, because the psephology of party support has changed so much over the past 20 years that this is not necessarily true now. If we go back 20 years, to an era when a middle-class person was markedly more likely to vote Conservative than Labour, and the opposite was true of someone who was working class—that, by the way, is not currently the case—historically, it had long been demonstrated that if you were a middle-class person living in an area that was predominantly populated by people in working-class occupations, you were more likely to vote Labour than if you were a middle-class person living in a more middle-class area.
There were two potential forces going on there. One is that, to some degree, middle-class people who choose to live in a more working-class area may actually already be rather more of a Labour disposition, but equally, it has certainly long been argued that to some degree, you are influenced by the social interaction to which you are exposed, so if you are living in a working-class community, you are more likely to be exposed to pro-Labour arguments than if you were living in a Conservative one.
Of course, the world has moved on in terms of the demography of party support, which is much less clearly structured by class, and social interaction is no longer as geographically bound as it once was and can now take place over social media. Iain may know more than me, but it has certainly been a while since I have seen anybody doing anything major on the extent to which community makes a difference. The only thing that I would say is that, undoubtedly, one of the reasons why MPs will always be concerned about any redistribution is that it upsets the connection between them and their existing electorate.
One of the things that we certainly do know—again, this may also be relevant to your question—is that if somebody has been elected for the first time at the last election and defeated the incumbent MP from another party, there is a fairly consistent tendency now whereby, in view of the next election, that new Member, who has probably just won a marginal seat, has a great deal of incentive to be representing their community and to be visible and so on, to get something of a personal bonus. You can see that in the way that the Labour party defended some seats in 2019, with newly incumbent, first-term Labour MPs doing well, and it was similar for the Conservative party in 2017. To that extent at least, yes, you can certainly also argue that a minority of voters—in some instances a crucial minority—will vote for their individual MP rather than for the party, but of course, if you get a boundary redistribution that carves up an individual MP’s constituency, that link is broken.
In truth, in our electoral system, there is a continuous and perpetual tension. We want our electoral system to do two things: on the one hand, we want it to provide local representation, and on the other, we want it to be a system that provides a means by which the electorate can choose between alternative Governments. I am afraid that I have spent the last 40 years pointing out the potential conflict between those two objectives and that, if you wish to ensure that the system is fair in the ability of voters to choose between alternative Governments, at some point you have to let go of the emphasis on local representation.
In a sense, the debate that we are having now about mathematical equality versus respecting community ties is a sub-part of that broader debate. Decide what your elections are about: if they are about the election of individual MPs and less to do with Governments, you can focus on representing communities; if you think that it is a system for enabling us to choose between alternative Governments, which is the traditional defence of the single member plurality system, I am afraid that local representation has to be given a lower priority.
Q If a community has a shared experience—perhaps, for instance, the “red wall” seats that people have talked about a lot since the last election—and wants to express a collective view through the ballot box, is it not important that those communities are connected and represented in a cohesive and clearly identifiable way, where they have common characteristics, so that their votes will count?
Professor Sir John Curtice:
That is what we used to have in the system of parliamentary representation when both boroughs and counties were represented and they were often of considerably unequal size. That comes back to the fundamental question about what we think elections should be about. Are they about providing MPs who represent communities, or are they a mechanism for choosing between alternative Governments? I am afraid that is just an inherent tension within the electoral system that we are looking at.
Q But if those views are diluted because communities are divided up in a mathematical exercise, do people not become frustrated because their collective view, brought about by their collective experiences in a locality, cannot be represented?
Professor Sir John Curtice:
Well, you are assuming that the current decisions of parliamentary constituents in some way already play out in—[Inaudible.] As Professor McLean has pointed out, what we regard as our community is sometimes in the eye of the beholder.
Professor Sir John Curtice:
There is a certain geographical concentration of voters who may or may not feel a sense of community, or who may in fact feel that they are an aggregation of many different communities. For example, I expect that relatively few of the constituencies in the far north of Scotland necessarily think that their constituency represents one agreed community, as opposed to a collection of villages. Indeed, if we go out to the Western Isles, where even the concept of village does not really exist, they will not necessarily think that the constituency is some clear, single, coterminous and homogenous community.
Q That is true. There are communities within boundaries, but it is important that they are not subdivided, just to satisfy a tight, rigid, mathematical exercise, is it not?
Professor Sir John Curtice:
The truth is that whatever set of rules you come up with, you may discover that you have got a choice about exactly how you try to represent community interest. At the end of the day, you may well simply discover that whatever rules you come up with, you end up dividing some places that you think—acknowledging that there is a question mark—might be a community.
Q My final point would be that, in that case, should we not allow the Boundary Commission more flexibility than the 5%, in order to meet those concerns, where there is a genuine expression of concern from a local community?
Professor Sir John Curtice:
I think my answer is that, while you might make it somewhat easier to avoid some of the cries that “This community is being divided”, the fact is that—if you go back to the current constituencies—communities are divided. Do we think that some of the lines that are drawn down the middle of Birmingham or London boroughs necessarily represent a community boundary? I suggest that they do not always do so.
I would urge Members not to go down that road. Of course, it is a political judgment for the Committee and the House of Commons. This is somewhat of a knight’s move answer to Mr Efford, but paragraphs 86 to 89 of the explanatory notes have a section about compatibility with the European convention on human rights. The criteria to be met are in paragraph 88 of the explanatory notes:
“The Bill maintains the principle of equal suffrage”.
The wider the margin, the less equal is the suffrage. That is the trade-off, which Parliament must decide to make. My view is that plus or minus 5% is ample, given that we have the device of protected constituencies. Of course, Members may wish to add to that number. I see that an amendment has been tabled that Ynys Môn should be added to the list, and Members might feel that Wirral should be added. Those are further instances of geographical peculiarities that might make the application of the 5% plus or minus more difficult. That is a political judgment for Members; as political scientists, or electoral mathematicians, we cannot say anything about it, except that those might be plausible cases. I would be against relaxing the plus or minus 5%, in the light of compatibility with the European convention on human rights, among other things.
Q We have had some powerful evidence that parliamentary boundaries are to a greater or lesser extent an artificial construct, although rules are put in place to try to acknowledge issues, which should be taken into account. I want to probe further something that Sir John talked about earlier. Because we are dealing with boundaries that are 20 years out of date, this will be a disruptive redistribution.
What comments can be made about trying to future-proof any proposals, to take into account any proposed developments and house building, while noting that those cannot be taken into any analysis of the quota? Do our experts have any views on whether that should be taken into account with regards to the geographical boundaries, so as to avoid unnecessary disruption in the future?
Professor Sir John Curtice:
There is a difference between the rules of the Local Government Boundary Commission for England and the parliamentary boundary commissions. The local government boundary commissions are permitted to take into account anticipated housing developments. I have had the occasional private conversation with people about this. You may want to quiz the Local Government Boundary Commission for England. The question that arises is how accurate the forecasts of house building and demolition activity are and the extent to which that ensure that the local government ward boundaries do not get out of date.
The answer to you is that it is certainly possible—see the rules of the Local Government Boundary Commission for England—but regarding the extent to which it is effective, you should ask the Local Government Boundary Commission for England, because I am not certain. There is a difference and you could anticipate doing a degree of that.
May I add to that? It is rather unfortunate that there are two sets of boundary commissions with different operating rules. Although it is not in the Bill, I do not understand why there needs to be a separate local government boundary commission, in particular one that operates under different rules, as John has just highlighted, from those used by the parliamentary boundary commission.
If one had to choose between these sets of rules—the Local Government Boundary Commission for England permitting evidence about future housing developments and the rules currently before you not permitting them —I would go with the rules that are in front of you, for the same reason that I gave in an earlier answer. One person’s likely housing development, which may just happen to favour that political party could be countered by another person’s likely future housing development, which may favour another party. I feel for the poor inspector, who is, by construction, not a specialist in the area, and is faced with claims that are very hard to adjudicate. You can adjudicate numbers, but future housing development is much more difficult.
Q Thank you to our witnesses for their evidence thus far. Professor Curtice was probably right to say that we should focus on things in the Bill. The two major things are going from 600 seats to 650, and parliamentary approval. To take the first issue in a question to both witnesses, why do you think that the Government changed their position, going from 600 seats to 650?
Professor Sir John Curtice:
That is not difficult. Turkeys were persuaded to put Christmas in the calendar in 2011 but, when Christmas eve came along, they decided to abandon it. There was always going to be a question mark about the willingness of MPs to vote for their own demise.
The reason why we were to have the cuts in the first place is that in 2010 both parties in the coalition proposed reductions in the size of the House of Commons. That was a populist response to the MPs’ expenses scandal. In the end, the cut to 600 that they introduced was less than those in the two parties’ manifestos. Then, of course, implementing it became a victim in 2013 of the spat within the coalition over the failure to reform the House of Lords, and in 2018 of the anticipated inability of the then Conservative Administration to get the provisions through—because they were asking turkeys to vote for Christmas. I am indicating that that is a classic case of how, at the end of the day, it is difficult to persuade Members of the House of Commons to engage in a radical reform that will make their lives difficult.
By the way, given that you have asked this question, let me expand its scope slightly. This is an aspect of the Bill that matters, and this is the question of the attempt at automaticity. To make it clear, there is an issue about automaticity—that is, the ability of Parliament to intervene. Parliament intervened in 2013 and stopped the boundary commissioners working—that was the work of Labour and the Liberal Democrats together—and in 2018 the Conservative Government failed to push the provisions through. Back in the late 1960s, the then Labour Government got their MPs to vote down the provisions. To that extent, there is clearly an issue. Although we have a process of neutral boundary proposals operating under rules set by the House of Commons, in effect the Commons has on three occasions, under different Administrations, ended up not implementing the rule, so there is an automaticity question.
My concern, however, is that although the Bill might make it more difficult for that to happen again, it will not stop it happening again. Given that in clause 8 the Bill stops implementation of the 2018 review, going on to have provisions that supposedly make it impossible for Parliament to overturn things in future, the truth is that the same is perfectly possible for a future House of Commons—a boundary review comes along, the current Administration does not like it, saying, “Actually, we should delay it”, and all they need to do is to introduce a quick piece of primary legislation to overturn it.
As we saw with the Fixed-term Parliaments Act, it is very difficult to introduce provisions that discipline the House of Commons to keep to a set of constitutional rules, given that we do not have an entrenched constitution. Although all of us would laud the fact that the provisions of the Bill are an improvement, reducing the ability of Parliament to stop things, we should not fool ourselves into thinking that it will necessarily stop Parliament, not least because even within the terms of the Bill an order has to be laid—instead of
“as soon as is reasonably practicable” at the moment—under the new provisions,
“as soon as may be reasonably practicable”.
I am not a lawyer, but the distinction between those two things still strikes me as rather fine on whether or not we could still be left in the situation that we had in the last Parliament, when the provisions were simply were not put before the House of Commons in a timely fashion. That could be repeated.
I have very little to add. The automaticity may look worrying to some, because it removes the rule from Parliament, but parliamentary supremacy is mentioned in the explanatory notes and of course the Bill could be enacted and then repealed by a future Parliament. That is the nature of parliamentary supremacy. It would be very embarrassing—the mother of Parliaments, one of the oldest parliamentary democracies and so on: it is already very embarrassing that it is operating on the basis of 20-year-old boundaries and therefore we did not have equal suffrage in the 2019 general election, to put it at its most blunt. I would concur with John that Parliament could do it again. It would be embarrassing, and I rather hope it does not.
Q Of all the things that are embarrassing about the mother of Parliaments I do not think that is the one that would come top of my list. Can I ask specifically about the distribution of seats, and the idea that based on what is before us there would be a reduction in the number of seats for Scotland and Wales? Professor Curtice mentioned that it was a destructive process. Would you go so far as to say that that would impact on the harmony of the Union?
It was bound to be disruptive once a uniform electoral quota was introduced for the four nations of the UK. John, the Minister, or others can correct me, but I think that that was done by the 2011 Act. The fact that, as has already been mentioned, the two instances of review that should have happened under the 2011 Act have not yet happened, means that that bomb, as it seems to some in Scotland and in Wales, was primed in 2011. It has not yet exploded, but it will with the implementation of this Bill; but that is a necessary consequence, as all Members know, of a uniform electoral quota for the United Kingdom. I cannot say any more than that.
Professor Sir John Curtice:
Can we go back a bit on the history of this? The truth, as Iain will explain much more eloquently than me, is that he original over-representation of Scotland and Wales was entirely the product of accident rather than design. When the Scottish Parliament was introduced in 1999 by the Labour Administration one of the things that was done as a result was indeed to reduce the size of Scotland’s representation in the House of Commons—although it was done in a manner that was arguably technically deficient, and did not necessarily deal with the possibility that there would be future disparities between the growth in population in Scotland and that in England.
The principle of basically saying that Scotland’s representation should be proportionate to England’s representation was already embodied by the Labour party and Labour Administration at the beginning of the century. The same thing was not done for Wales because of course when the then Welsh Assembly was first created it had only secondary legislative powers, and it was therefore felt that the devolution was not on a scale that justified the reduction in the number of Welsh MPs. Given that we now have a Welsh Senedd that has primary legislative powers that are not commensurate with, but not that dissimilar from, those of the Scottish Parliament, as it were, what has already been done for Scotland seems to be relevant for Wales.
As to the actual effect, now we are talking about a 650 Parliament: by my calculation, which is based on the electorates as of the election—but, given we are now going to do the electorate on
This is essentially a redistribution from Wales to England, and then within England it is a redistribution really from a line from East Anglia southwards—as opposed to the northern parts of England. Of course one of the ironies of the situation we are now in is that because the Conservative party gained so many seats—they had the so-called red wall seats in the north of England and so on—actually the disparity in the size of the electorate between constituencies that are represented by the Labour party and those represented by the Conservatives is smaller than it has been at any point during these current set of constituencies. In other words, changes in electoral geography are changing the politics of redistribution. London is one of the places that will benefit; it is now a Labour city. The north-east of England, which now has a non-trivial number of Conservative MPs, will lose out heavily. Therefore, actually the redistributive consequences politically are perhaps not quite as toxic as we might have imagined 10 or 15 years ago.
Professor McLean, you mentioned that you felt it would be difficult for the preference of existing constituencies to be kept to if we keep within the 5% quota, because there would need to be quite a substantial revision, and Professor Curtice made similar remarks. Could you expand a little on your analysis of how that might shake out? In terms of our recent electoral history, where do you think this disruption will rank?Q
A problem is caused when you are going by a regional area. The practice of the English commission has been to go by counties for some of its units, including administrative counties such as the former metropolitan counties that were abolished in 1986. That is a defensible practice, because the larger the unit within which you operate, the easier it is to reconcile conflicting criteria. Therefore, if you are in a unit of, let us say, three constituencies, one of which by happenstance is the right size and the others are not, it might be difficult to maintain the right-sized one and observe the other rules. If you are in a unit of 15 constituencies, one of which is the right size, the commissioners have more freedom to draw a map that retains the constituency that happens to be the right size while altering the others.
I said earlier that it is likely—I do not have the data, but John may—that there are now very few constituencies anywhere in the UK that are the right size, which is to say, one 600th of the House, given that we have had 20 years of migration and the disruption mentioned in Scotland and especially Wales. So I think it will be very hard to preserve existing constituencies.
Professor Sir John Curtice:
All I can add is that I did look quickly at what statisticians call the standard deviation of constituency size—that is simply a measure of the extent to which the number of registered electors in a constituency varies between one seat and another—and that number is constantly increasing. Basically, there is now a greater difference in the size of constituencies than there was in 2017, there was a greater difference in 2017 than in 2015, and there was a greater difference in 2015 than in 2010. Although politically this redistribution may not be as dramatic as people on both sides of the House might imagine, there is no doubt that getting the constituencies to reflect electorate sizes is bound to be disruptive.
That is going to be up to the operating practices of the Boundary Commission for England if it remains non-statutory, and it is not proposed in this Bill that it should be given statutory instructions different from those in the 1986 Bill. Thinking on my feet, I think that with the exception of the Isle of Wight, which is not a true exception because it is one of the preserved areas, county populations in the south of England are sufficiently large that—sorry, we are not here treating Rutland as a county—
If we take out the Isle of Wight and possibly Rutland, it should be reasonably feasible for the English commission to operate at county level, but that is an operating matter for the commission. At present it is not in the Bill. If an amendment to give greater respect to county boundaries were introduced to the Bill during its progress, that might imperil the equality rule, which the current law gives as trumps.
Q One final question, if I may. We have talked a lot about the automaticity of this and how Parliament does not have the ability to vote it down. Bearing that in mind, do you both have a sense of how you feel the boundary commissioners might behave differently now that they would almost have a bit more freedom because of Parliament’s not having its ability to vote it down? Do you think they will behave differently in their judgments? They are obviously very professional people and will do their work as best they can, but we all live in the real world and we know that if we think that Parliament has the ability to vote it down, that might affect how radical the final findings we present are.
The only one of the four commissions that has possibly felt itself at risk under those conditions in the past is the Northern Ireland one, where there are deep issues of community and sectarianism. I am all for protecting commissions from that sort of pressure. Having observed the operations of county inquiries in England—I have never done a Scottish inquiry—I would say that the boundary commissions’ staff and inspectors have always maintained great professionalism. I would not expect that to change under the sort of behavioural issues that you raise.
Q Thank you, Sir David. I will be brief. We have talked about automaticity and the House of Commons using its political interests to reject proposals from the boundary commissions in the past. Is there not a danger, however, that the instructions given to boundary commissions at the outset through the legislation will also have political considerations in them based upon who has the majority in the House of Commons at the time, and therefore a further return to the boundary commission at the end gives a safety valve to perhaps counterbalance the political considerations that might have outweighed the criteria given to the boundary commissions.
To Mr Matheson’s question, I am not too concerned about this Bill, perhaps precisely because this Bill does not go into the level of detail that some people might have wanted. It does not give instructions to the commissions, for instance, to always respect local government boundaries or not. The commissions have that discretion. If this Bill is enacted in the rather spare form in which it is in front of you, I would not be too worried about the sorts of issues that the Member has just raised.
Professor Sir John Curtice:
Yes, of course you are right that the rules for redistribution are always politically contentious. That said, and to give him due praise, the rules that are now being devised, in so far as how you allocate seats to the parts of the United Kingdom and within England, do follow the rules that Professor McLean was crucial in persuading the Electoral Commission were the right rules to use for allocating MEPs to the regions in the European electoral system. That can be shown to be the fairest way of doing it. On the first point, yes, you are right.
On the second point, as I have been keen to point out to you, if at the end of the day the House of Commons thinks a boundary commission has fouled up, it can still stop the boundary commission. Any new Administration, in particular, can stop it by simply passing new legislation, so you still have the nuclear weapon if you want it.
On the subject of political aspects, that is a part of the Bill that should be discussed; I am concerned that there is some political consideration going on here. Nobody has raised the point that the next review under this is supposed to end in July 2023 rather than in October 2023. No justification is given for that in the Cabinet Office memo or in the explanatory notes. The only explanation that I can think of—maybe I am being unfair—is that somebody is wanting to pave the way to make it possible to hold a general election in autumn 2023 rather than in spring 2024. Certainly, somebody needs to explain why the next procedure is going to be foreshortened by three months for a set of boundaries that are then going to be in place for another eight years, and this is not going to happen thereafter. There is no justification so far, and I encourage the Committee to inquire further.