Submission to the Procedure Committee inquiry ‘Elections within the House of Commons’
Professor Meg Russell FBA, Professor Alan Renwick, and Dr Tom Fleming
Constitution Unit, University College London
May 2025
Summary
1. Introduction
The UCL Constitution Unit conducts timely, rigorous independent research into constitutional change and the reform of political institutions. Since its foundation in 1995, its researchers have carried out extensive work on parliament, House of Commons reform, and elections. The Unit’s Director, Professor Meg Russell, was a specialist adviser to the ‘Wright Committee’ that was crucial to the establishment of elections considered by this inquiry.
This submission addresses the first two questions in the inquiry’s call for evidence:
2. Outline of voting systems used in House of Commons elections
2.1. Background
Elections are used in the House of Commons to fill the following posts:
Speaker
Deputy Speakers
chairs of most select committees
members of most select committees.
The current procedures for electing the Speaker and Deputy Speaker were established in 2001 and 2010, respectively. In each case, they were based on proposals by the Procedure Committee. Elections for the chairs and members of (most) select committees were introduced in 2010, implementing recommendations from the House of Commons Reform Committee (known more often as the ‘Wright Committee’ after its chair, the then Labour MP Tony Wright).
These elections take place in three stages: first, most posts (except the Speakership) are allocated to one or other side of the House or to particular political parties; second, eligible candidates are nominated; third, voting takes place to choose among the eligible candidates.
The procedures used at each of these three stages might be debated. For example, two of us wrote last year about the non-transparent nature of the procedure for allocating committee chairs and memberships across political parties.[1] But the inquiry terms of reference refer specifically to voting systems, which we take to mean only the third of the three stages. We therefore focus our remarks primarily on that stage.
2.2. Voting systems
Elections of the Speaker, Deputy Speakers, and committee chairs take place using voting systems that all follow the same basic logic, but that differ somewhat in their mechanics.
The Speaker is elected through a multi-round voting process. At each round, voters can cast a ballot for one candidate, and a candidate securing over 50% of the valid votes is elected. If no candidate meets that threshold, the candidate with fewest votes and any other candidates with fewer than 5% of the votes are eliminated, and other candidates are given the chance to withdraw from the contest; then another round is held.[2]
The Deputy Speakers and committee chairs are elected in a single round using transferable ballots. Where only one office is available in each contest – as for the committee chairs – this system is known as the Alternative Vote (AV), and the logic is identical to that in the multi-round system. First preferences are counted; if no candidate passes 50% support, the lowest candidate is eliminated and their supporters’ second preferences are added to the count; this process continues until a candidate surpasses 50% of the votes that are still in the count.
Where multiple offices are elected concurrently – as for Deputy Speaker elections – the system is known as the Single Transferable Vote (STV). There are three Deputy Speaker positions, so candidates must exceed one quarter of the vote to be elected (on the basis that it is impossible for more than three candidates to achieve that). Last-placed candidates are treated as under AV (i.e. eliminated, with their voters’ subsequent preferences reallocated). In addition, when a candidate is elected, their surplus votes (votes over the threshold required for election) are also transferred, in order to ensure that all voters are treated equitably.[3]
The Deputy Speaker elections have several added complications. First, although the three positions are elected concurrently at the start of each parliament, two must be ‘from the opposite side of the House [i.e. government or opposition] to that from which the Speaker was drawn’ (Standing Order No. 2A(5)(e)(i)) and one from the Speaker’s side. Second, the three positions are ranked. The first-placed candidate from the opposite side to the Speaker becomes Chairman of Ways and Means; the first-placed candidate from the Speaker’s side becomes First Deputy Chairman of Ways and Means; the second-placed candidate from the opposite side to the Speaker becomes Second Deputy Chairman of Ways and Means. Finally, across the four positions of Speaker and Deputy Speakers, at least one man and at least one woman must be elected.
The voting system for electing ordinary members of select committees is not prescribed in standing orders. Indeed, as we discuss below, the standing orders do not acknowledge the existence of these elections, as they are organised by political parties rather than the House. Formally, select committee members are appointed by the House approving a motion to that effect. All that is mentioned in standing orders is that such motions – with some specific exceptions – must be made on behalf of the Committee of Selection.
3. Competitiveness of elections within the House of Commons
The inquiry’s terms of reference ask whether elections within the House of Commons have become more competitive since 2010. This comes against a background of complaints from MPs about the extent of canvassing by candidates for select committee chairs after the 2024 general election. That is addressed by the committee’s third question, which we do not focus upon substantively, but increased competitiveness could be one possible driver of such a trend. Our discussion of competitiveness therefore focuses on select committee chair elections.
There is some evidence that these elections have become more competitive. In particular, Dr Marc Geddes and Dr Stephen Holden Bates have examined how far elections are contested, rather than a single candidate winning unopposed. Their work highlights that the 2024 elections saw the smallest percentage of unopposed elections since the first such elections in 2010.[4]
However, Constitution Unit research has shown that other ways of measuring the competitiveness of elections lead to rather different conclusions.[5]
First, we can look at the number of candidates taking part in the elections (see Table 1). A larger number of candidates can make an election more competitive by creating greater uncertainty about the outcome and putting additional pressure on candidates to out-campaign their rivals. On this metric, there is no evidence of increasing competitiveness over time. The median number of candidates competing for each individual post has remained broadly consistent, while the mean number of candidates was higher in 2010 and 2015 than at any subsequent set of elections, including 2024.
Table 1. Mean/median candidates in select committee chair elections | |||||
Candidates | 2010 | 2015 | 2017 | 2019 | 2024 |
Mean | 2.38 | 2.35 | 1.93 | 1.93 | 1.96 |
Median | 2 | 2 | 1 | 2 | 2 |
Second, we can examine the margin by which winning candidates were elected. Smaller winning margins can be seen as an indicator of greater competitiveness, as they show elections were genuinely close contests rather than a foregone conclusion. Table 2 therefore shows the mean and median winning margins in contested elections since 2010. Winning margins were calculated as the gap in votes between the winning and second-placed candidates in each election.[6] This second metric again shows no evidence of increasing competitiveness since 2010. If anything, it shows some signs of the opposite, as the typical winning margin of select committee chairs was larger in 2024 – whether measured as the mean or median – than it was in 2010 and 2015.
Table 2. Winning margins in contested select committee chair elections | |||||
Margins | 2010 | 2015 | 2017 | 2019 | 2024 |
Mean | 73.8 | 113.3 | 96.3 | 125.7 | 124.5 |
Median | 72.5 | 71.5 | 113.5 | 121.0 | 98.0 |
Overall, there is no clear trend towards select committee chair elections becoming more competitive over time, and 2024 does not seem distinctive in this respect. There may have been a larger number of contested chair elections than at the start of most other parliaments post-2010. But those elections were not contested by an unusually large number of candidates, and they were not won by unusually small margins.
Any recent increase in candidates’ canvassing efforts therefore seems unlikely to have been driven by an equivalent increase in the underlying competitiveness of the elections, and may instead result from other factors. One plausible contributing factor could be the unusually high number of new MPs elected in 2024, if such MPs were (anticipated to be) more reliant on campaign literature for learning about the chair candidates.
4. Clarity of voting systems used within the House of Commons
The inquiry’s terms of reference also ask whether there is sufficient clarity regarding the voting systems used in House of Commons elections. This question may refer both to whether clear rules exist and to whether those rules are widely understood and recognised as legitimate.
4.1. The Speaker, Deputy Speakers, and committee chairs
As the paragraphs above indicate, the procedures for electing the Speaker, Deputy Speakers, and many committee chairs are precisely defined. At least in the first sense, those voting systems are therefore clear. It is more difficult for us to comment on whether the voting systems fulfil the second aspect of clarity: whether they are widely understood and recognised as legitimate among MPs. But while the committee reflects on this matter, it may be useful for it to consider that the existing voting systems do in fact have a very strong underlying rationale.
As explained above, the voting systems for the Speaker and the committee chairs, though different in their mechanics, follow an identical underlying logic. In each case, the goal is to find a candidate who can command the support of a majority in the House. Through successive rounds of voting/counting, the last-placed candidate is eliminated from the contest until a candidate meeting this requirement (not counting abstentions) is found. This ensures that a candidate with narrow support cannot be elected – as could be the case under First Past the Post. It also encourages candidates to seek support from across the House – as would seem desirable for posts that are intended to act in a non-partisan manner. Elections of this kind are widely used in other contexts as well: for example, Labour, the Conservatives, and the Liberal Democrats all use such systems to elect their leaders.
The STV election for the Deputy Speakers follows a slightly different logic. Here, notwithstanding the ranking of the positions, a collective pool of office-holders is being elected. STV is designed to elect a group of people who together represent the diversity of opinion across the House – at least, to the extent that this is possible with only three positions (or four including the Speaker). If that is the goal, STV is the system best suited to delivering it.
At the committee’s oral evidence session on 23 April, former Deputy Speakers Nigel Evans, Baroness (Eleanor) Laing of Elderslie, and Baroness (Rosie) Winterton of Doncaster put forward the alternative suggestion of splitting the ballot paper, such that one vote would be held (by AV) to elect the Deputy Speaker from the Speaker’s side and a separate vote would be held (using STV) to elect the two Deputy Speakers from the opposite side. They emphasised that all MPs would need to be able to vote in each of these contests, given that the Deputy Speakers’ role is to represent the whole House, rather than just their own side. In making this proposal, they highlighted several concerns regarding the current arrangements:
These are valid and important points. But two additional points were not made that the committee might consider. First, the solution of holding separate elections for the candidates on each side of the House would not address the third problem: it would still be the case that the candidate with most votes on the opposite side from the Speaker would be elected Chairman of Ways and Means whether they wanted that role or not. To address this, one option would be to have three contests: one for Chairman of Ways and Means; one for First Deputy; and one for Second Deputy. Alternatively, the candidate coming first on the opposite side to the Speaker could be allowed to choose between the two roles allocated to that side.
Second, however, any separation of contests would undermine the goal of electing a set of three people who reflect the diversity of opinion across the House. That would be clearest if three separate contests were run for the three Deputy Speaker roles. Each of these contests would use AV, with the majority winner being elected in each case. That would give the largest party preponderant power over each election (even though not all participants may vote along party lines). In a situation such as the present one, with a substantial single-party majority, that party’s MPs would determine the results of all three contests. The proposal to have two contests – one for each side of the House – would be intermediate between the current arrangements and the three-contest option, meaning it would still undermine the goal of reflecting the diversity of views across the House relative to the status quo.
As noted above, there is an alternative – and very simple – solution to the problem that not all candidates may want to be Chairman of Ways and Means: allow the leading candidate to choose which post they will take.
There is also an alternative solution to the problem of an uncontested position: require all candidates to pass a threshold in order to be elected, even if they are unopposed. It would not be an ideal solution: it might lead to the need for a rerun. But it would preserve the basic goal that the outcome should reflect the diversity of the House. It would be akin to requiring a candidate in an uncontested election to run against ‘none of the above’ – which is an option that Baroness Winterton in fact appeared to float during the committee’s oral evidence session.
If the committee thinks that reflecting the diversity of the House matters, then the current system – perhaps with the two adjustments just suggested – is the best one. That would leave the final concern around confusion. It would seem better to address this by working on explaining the system as well as possible than by adopting a system that yields outcomes that are less fair.
4.2. Select committee members
A far bigger question of clarity applies to the election of select committee members, where the system is almost entirely opaque to those outside parliament, with important aspects unclear even to those participating in the elections.
For the Wright Committee, the election of select committee chairs and members went hand in hand, with the distinction that the former were to be elected by the whole House, and the latter to be elected from within the political parties (following an initial agreed allocation of numbers of members on each committee).
It is clear from the Wright Committee’s report that the intention was to introduce election of committee members for the same committees whose chairs would be elected. Recommendation 5 (paragraph 80) of the committee’s report opens with the following words:
We recommend an initial system of election by the whole House of Chairs of departmental and similar select committees, and thereafter the election by secret ballot of members of those committees by each political party, according to their level of representation in the House, and using transparent and democratic means. The committees within this system should be those appointed under SO No 152 [the departmental select committees] together with the Environmental Audit Committee, the Public Administration Committee and the Committee of Public Accounts.[7]
The standing orders to implement the committee’s recommendations on the select committees were debated and agreed on 4 March 2010. At this point the Procedure Committee was added to the list of committees to be elected, and this was approved as part of what became Standing Order No. 122B. It is this standing order which determines, and makes transparent, which select committees will have elected chairs and which sets out the rules for those elections.
The standing order does not explicitly state that these same select committees will have elected members, though this was a clear assumption at the time of the Wright Committee, and was also reflected in the debates on its report.
The Wright Committee’s recommendation 6 (paragraphs 87–88) states the following:
We propose that in the new Parliament members of departmental and similar select committees should be elected by secret ballot within party groups, by transparent and democratic processes, with the outcome reported to and endorsed by the House. Party groups would in effect be acting on behalf of the House as electoral colleges. They would therefore expect to act under some constraints as to the methods used to elect committee members. We do not think it necessary that the House should interfere so far as to lay down one particular method of election rather than another. But the method chosen should be one approved by the Speaker, following independent advice, as transparent and democratic: ‘kite-marked’ as legitimate in effect. Officers nominated by the Speaker would be obliged to assure themselves that the processes followed by each party, as notified by its Leader, were indeed in accordance with these norms. And each party would be obliged to publish the method it had adopted.
In the same debate on 4 March 2010, a motion was approved endorsing the key principle of election for committee members, whose relevant words state:
That this House takes note of recommendation 6 of the First Report of the Select Committee on Reform of the House of Commons, Session 2008–09, HC 1117, and endorses the principle that parties should elect members of select committees in a secret ballot by whichever transparent and democratic method they choose…
This fell short of the Wright Committee recommendation that the method of election should be overseen by the Speaker, but emphasised that election should be ‘transparent and democratic’.
Subsequently, there have been three difficulties with this election process compared to the expectations of the Wright Committee:
There is no public record of the committees that have been elected inside the parties since 2010, and how those elections have been conducted. While it might be disproportionate for standing orders to set out these procedures, since they are internal to the parties, it would seem desirable for such a public record to be kept.
In the most recent elections in 2024, for example, it came to light that the Labour Party had not included your own committee on the list of those whose members would be elected. Whether this applied within other parties, or to other committees, in 2024 or at any previous election, is unknown. There may perhaps be various other cases. This seems problematic, and in breach of the motion agreed by the House in 2010.
For the sake of transparency (a core principle emphasised in that motion, echoing repeated words from the Wright Committee) and comprehensibility of the system for the public, and indeed MPs themselves, it seems desirable that there should be an official and public list of committees to be elected, and some public record of the electoral process followed by the parties. It could also, for example (in line with the discussion above about competitiveness), be valuable to record the number of candidates nominated in each party for each committee. As a first step in this direction, your committee could use the report of the current inquiry to restate the select committees whose members are currently expected to be chosen by internal elections.
While it may have been decided in 2010 that asking the Speaker to oversee the elections was burdensome or otherwise undesirable, it would also seem reasonable for the Procedure Committee to routinely report on these elections, perhaps alongside reporting on the elections of select committee chairs, at the start of each parliament.
8
[1] Alan Renwick and Tom Fleming (2024). ‘Select committee elections: how should a ‘proportional’ allocation between parties be calculated?’, Constitution Unit Blog, 12 September 2024.
[2] This process is only followed when a vacancy arises in the Speakership, or when – at the start of a new parliament – the House votes against the motion that the incumbent Speaker continues in that role. The only two Speaker elections since 2001 – in 2009 and 2019 – were due to the first of these circumstances.
[3] The correct procedure here is to count the next preferences on all of the ballot papers cast for the successful candidate, but to give each ballot only a fractional value when it is transferred, so that the total value of the ballots transferred is equal to the size of the candidate’s surplus. For example, if the threshold for election is 200 votes and a candidate secures 400 votes in the first round, their surplus is 200 votes (400 minus 200). The second preferences on all 400 ballot papers are examined and the totals for each remaining candidate are counted up. But these totals are then divided by 2 (400/200) so that the total value of the transfer adds up to 200. In the past, some countries have used a process where ballot papers are added to a candidate’s pile until the candidate reaches the threshold, and then remaining ballot papers are allocated according to second preferences. But that means that the outcome may be affected by the random fact of the order in which the ballot papers are counted. The justification for this approach was just that it was computationally simpler; but there is no longer any case for that in the age of modern computers.
This example also illustrates why equity requires transfers to be made. If 600 MPs voted and 400 of them backed one candidate, then, without transfers, those 400 MPs would determine the winner for one of the three posts available, while the remaining 200 MPs would determine the outcomes for the remaining two. Transferring the surplus of 200 votes ensures that all MPs have the same amount of influence over the outcome.
[4] Marc Geddes and Stephen Holden Bates (2024). ‘Select committee chair elections: what do the results tell us about the new Parliament?’, Hansard Society website, 27 September 2024.
[5] Tom Fleming (2024). ‘Have select committee chair elections got more competitive?’, Constitution Unit Blog, 21 November 2024.
[6] In cases where no candidate secured a majority of first-preference votes, this figure is based on the margin after the final round of redistributing votes.
[7] Reform of the House of Commons Select Committee (2009), Rebuilding the House (First Report of Session 2008–09), HC1117 (London: House of Commons).