Written evidence submitted by Professor Iain McLean FBA FRSE (EHC 02)
1 I am a Senior Research Fellow in Politics, Nuffield College, Oxford, and an expert on electoral procedures. My publications in this area may be accessed from the landing page at https://www.nuffield.ox.ac.uk/people/profiles/iain-mclean/
2 In this note I comment on two of the matters in your Terms of Reference: Is there sufficient clarity regarding the voting systems used in House of Commons elections? and Should all select committee chairs be elected? If not, why not?
3 The best-known result in the formal theory of voting is Arrow’s Impossibility Theorem (1951). This proves that in any election with at least three options and at least three voters, no system may jointly satisfy all of five minimally demanding conditions.
4 In particular, simple majority rule over more than two candidates (or propositions) is vulnerable to the Condorcet paradox wherein, for some A, B, C, it is always possible that A beats B, B beats C, and C beats A, all at the same time. As Condorcet paradoxes (labelled ‘cycles’ by the great 19th-century theorist C.L. Dodgson) are inescapable, it is desirable that they should be revealed where possible, and a fallback procedure devised.
5 The best-known system that cannot generate voting cycles is rank-order voting, known from its main originator as the ‘Borda count’. However, it is highly manipulable. Although we described it in our 2009 evidence, I would not recommend its use.
6 Voting on propositions in assemblies, including the House of Commons, is typically binary. Standing Orders mandate that only one proposition is before the house at a time, and the voter has only the options Yea, Nay, or Abstain. Standing Orders also govern the precedence of amendments both substantive and procedural.
7 Binary procedures are not exempt from the paradoxes of voting when more than two options are to be voted on. This was starkly revealed, for instance, in the sequence of votes on House of Lords reform in February 2003. The Commons voted against each of five options for the composition of the Upper House, including a fully appointed house. As all five motions were defeated, the outcome was the status quo, viz., a fully appointed house, which was one of the options the House of Commons had voted down.
8 This call for evidence relates only to the election of people to posts: specifically to the three Speaker’s deputies and the chairs of select committees.
9 Your predecessor committee considered the procedures for the election of the Speaker and the three deputies, see HC 1080 (Fourth Report of Session 2008-09); and HC 341 (First Report of Session 2009-10). With my then colleague Scott Moser, I gave written and oral evidence to that committee. The oral evidence was at an informal session and was not transcribed. The two tranches of written evidence are in HC341 at pp. Ev 5 – Ev 9.
10 I believe that the conclusions we came to in 2009-10 remain valid, and I do not wish to either change or repeat what we said then. However, some of our evidence touched on rather technical matters, which I can go into in more depth in oral evidence if the present Committee wishes me to.
11 The House supported the election procedure proposed by your predecessor committee for the three deputy speakers, and the rules are now in Standing Order 2A.
12 Those rules provide for a single election to the three posts using the Single Transferable Vote, with two constraints:
13 Running an STV election with constraints is not easy, but the Electoral Reform Society has considerable experience of it in other contexts. I assume that it has not given rise to any problems in practice since the current wording of Standing Order 2A was adopted. On this matter I defer to the Electoral Reform Society.
14 Although STV cannot escape the Arrow trap, because no procedure can, it is relatively hard to manipulate it. I therefore stand by our recommendation of 2009 that it is (still) an appropriate voting procedure to use.
15 Your predecessor committee considered a proposal that instead of simply recommending STV to the House, it should invite the House to choose between STV and a system known in the literature as Approval Voting. It did not adopt that proposal (HC 341, Formal Minutes, p.36). If the current committee so wish, I can discuss Approval Voting in oral evidence, but I see no strong reason for reopening that question.
16 Is there sufficient clarity regarding the voting systems used in House of Commons elections? In one word, yes. The technical issues with implementing STV subject to constraints are ‘hidden wiring’ which I think need not trouble the Committee (see para 13 above).
17 To the best of my knowledge, the electoral system has only been questioned once by a Member since its first implementation (by Sir Bernard Jenkin on 7 January 2020~: see https://hansard.parliament.uk/Commons/2020-01-07/debates/7A9EC2BE-C8A9-4875-9076-1C9749727E88/PointsOfOrder#contribution-AADD3CE3-2BB7-4E7F-92F1-49863C27966B. The Speaker expressed sympathy with Sir Bernard’s view but as far as I know there has been no follow-up.
18 Therefore I recommend no change in the electoral system.
19 Your predecessor committee discussed the question of defining ‘government’ and ‘opposition’ for the purpose of the eligibility criteria for the team of Deputy Speakers, as did our evidence to it.
20 It is important that the categories ‘government’ and ‘opposition’ are mutually exclusive and jointly exhaustive Every MP who is not a (possibly shadow) minister should be eligible to serve. A consequence is that there must not be any insistence that the number of ‘opposition’ Deputy Speakers required for balance (currently 2, given that the current Speaker was originally elected as Labour) must all be drawn from the Official Opposition. This point is particularly important in the current Parliament for obvious reasons.
21 A Parliament in which the governing party has a large majority of seats does not pose any conceptual problems. If anything, it is more important than ever that the Speaker’s team should contain an equal number of Members elected from the government and opposition.
22 If the result of a General Election is that no party has an overall majority, then there is some chance that the Deputy Speaker team has to be elected before it becomes clear which MPs are on the government side and which on the opposition side. Your predecessor committee discussed this, as did our evidence.
23 In 2010 no problem arose. The Coalition Programme for Government was announced on 20 May. The election for Deputy Speakers took place on 7 June, and all candidates were successfully labelled ‘Government side’ or Opposition side’. See https://www.parliament.uk/business/news/news-by-year/2010/06/deputy-speaker-candidates-announced/
24 In 2017 the question might have arisen whether any DUP Member who wished to stand for Deputy Speaker should be classed as ‘Government side’ or ‘Opposition side’. None did stand. If any had, my view is that it would have been appropriate to class them as ‘Government side’. Only one candidate from the ‘Government side’ was nominated (Eleanor Laing), and she was therefore nominated to the team without an election. The election of the two members from the ‘Opposition side’ took place on 28 June: https://www.parliament.uk/business/news/news-by-year/2017/june/election-of-deputy-speakers/
25 The present Speaker was elected at the very end of the 2017-2019 Parliament and therefore no question was raised as to whether the composition of the team should be changed to reflect that Speaker Hoyle originated from a different party than Speaker Bercow. The election of the Deputies did not take place until the new Parliament had met. For the avoidance of doubt, I hope that this committee will reiterate that there should be no change in the composition of the team following any election during the life of a Parliament of a Speaker from a different party to that of the previous Speaker.
26 The Committee may wish to discuss regularising the practice of appointing temporary deputies before the permanent deputies can be elected. See the Commons Library paper at chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://researchbriefings.files.parliament.uk/documents/SN05375/SN05375.pdf.
27 Should all select committee chairs be elected? If not, why not? I see no reason why not, but an electoral systems specialist has no unique expertise to bear.
April 2025