Senedd Cymru | Welsh Parliament
Pwyllgor y Bil Atebolrwydd Aelodau | Member Accountability Bill Committee
Bil Senedd Cymru (Atebolrwydd Aelodau ac Etholiadau) | Senedd Cymru (Member Accountability and Elections) Bill
Ymateb gan Keith Bush CyB | Evidence from Keith Bush KC (Honoris Causa)[1]|
Introduction
1.The author would like to submit the following comments in relation to the general principles of the Bill. These comments are limited to Part 1 of the Bill (Recall of Members of the Senedd).
General
2. The general aim of the Bill, which is to 'enhance the accountability of Members of the Senedd to the electorate' (to quote the Explanatory Memorandum), is commendable and enjoys universal public support. It is disappointing that Senedd Cymru does not have an equivalent mechanism for recalling Members who have engaged in serious misconduct to the mechanism that was established for Members of the UK Parliament under the Recall of MPs Act 2015. Members of the Senedd will also be aware that the Scottish Parliament has recently[2] approved the general principles of the Scottish Parliament (Recall and Removal of Members) Bill, particularly given that the Senedd’s Standards of Conduct Committee heard evidence from Graham Simpson MSP, the Member of the Scottish Parliament who introduced that Bill, before that Committee recommended pursuing legislation for Wales with the same objective.
3. The laudability of the Bill's aims, however, cannot justify bad legislation, and in the author's opinion, taking this Bill forward would create a complex system that would involve significant costs but would not achieve those aims. Indeed, there is a risk that it would make Members of the Senedd appear less accountable to the public. The problem (as explained below) is that the Bill (like the Scottish Bill) tries to emulate, as far as possible, the example of the Recall of MPs Act 2015, despite the fact that the system for electing Members of the Senedd, from May onwards, will be fundamentally different to that used in Westminster. In the case of the Scottish Bill, the majority of Members will still be elected via the same constituency system as that which is in place in Westminster. As a result, the Stage 1 Committee that scrutinised the Scottish Bill was content with the way in which the Bill deals with constituency Members. But in the case of regional list Members, who will still be elected in a way that is broadly similar to the new system for electing Members of the Senedd, the Committee perceived the same general difficulties as the author in respect of the principle of trying to apply to list Members a recall system that has been designed for constituency Members.
4. In the view of the Scottish Parliament Committee:
‘While the UK Recall Act is a useful point of reference for developing a recall system for the Scottish Parliament, its provisions do not accommodate for the additional member system for the election of regional MSPs. The limitation of the UK Recall Act in this regard has been recognised by Mr Simpson in his development of the Bill.
The Committee is of the view that, should provision be made for the recall of MSPs, the system put in place cannot be a direct replication of the UK Recall Act but rather a system for the Scottish Parliament, functioning with respect to its electoral system and accommodating for the recall of both constituency and regional MSPs.’
and
‘We ... invite Mr Simpson to take on board the evidence that has been received and to consider whether changes should be proposed if the Parliament agrees to the general principles of the Bill. We particularly invite Mr Simpson to focus on the question of the complexity and costs associated with any mechanism for recall of regional members. ... The Committee does not consider that the proposals as currently set out in the Bill have found the appropriate balance.’
5. Whilst it is true to say that the Welsh Bill has tried to devise a system that is different, in a number of ways, to the Westminster recall system, and to the system under consideration in Scotland, the author believes that the influence of the Westminster precedent has affected the Welsh Bill and that, in any case, the solutions proposed in the Bill to the challenge of applying a recall system to an electoral list system do not improve matters, but instead make them worse. This point is expanded upon below.
The root cause of the difficulty
6. When Members of a legislature are elected as individuals (for example, under a 'first past the post' system or a proportional system such as the Single Transferable Vote (STV), which also results in the election of individual Members) the seat of any Member who has been dismissed is filled by holding a by-election. The result of recalling a Member is to give the electorate the right to elect, via a by-election, a Member in their place (or even to re-elect the original Member, if they choose to stand again). The principle that voters control the membership of the legislature and that this power remains in their hands is reinforced, as the recall procedure leaves to voters the decision as to who should represent them. However, under a closed list system (such as the new system to be used by the Senedd), voters have no say in choosing their representatives. That choice is transferred to the political parties. And the system adopted by Senedd Cymru (following the regime that has been in place since 1999 for regional Members) is that there will be no by-election if a Member loses their seat. Instead, the empty seat will be filled by the person who is next on the list of the party to which the Member belonged at the time of the election. (If a Member has left the party in question since the election, this has no effect – their original party will be allowed to fill the seat.)
7. In the case of a system based on individual representation, such as the system used in Westminster, it is therefore possible to design a recall system that works fairly effectively. If a Member misbehaves, a percentage of the voters in their constituency can demand a by-election, and the voters who will express their view in that by-election can take into account the party's failure to ensure that its Members adhere to acceptable standards of behaviour. The function of a recall petition, therefore, is to trigger a democratic process for reconsidering who should represent a constituency.
8. But in the case of a system based on closed lists (or, at least, a system like that of the Senedd, where there will be no by-elections), the second stage in the process – reconsidering, through a by-election, who should represent a constituency – is not available. The Bill has been drafted, therefore, in a way that tries to find a means for the electorate to express an opinion on the misconduct of the Member, but without being able to affect the position of the party that chose the Member and to whose Senedd group the Member belonged. What the electorate is granted under the Bill, therefore, is a recall poll. This will not be the equivalent, in practice, of a recall petition under the Recall of MPs Act 2015 (or a recall petition under the Scotland Bill). Instead, it seeks to combine the two stages set out in the Scotland Bill for regional list Members in Scotland: a petition to decide whether the Member will be dismissed, and then a poll to decide whether the empty seat will be filled by the Member who has just lost it or by the person who is next in line on the party list. (We must bear in mind that the threshold for a petition to dismiss a Member is only 10% of voters, while every voter will have the right to vote in the poll. It would therefore be possible for the majority of voters to disagree with those who have demanded the Member's dismissal and wish to retain them as a representative.)
9. In respect of the Scottish Bill, it is possible to criticise Mr Simpson’s proposal for putting in place a long-winded process, but at least it tries to adhere, as closely as possible, to the principle of separating the decision on whether to recall a Member from the decision on who should succeed the Member. Of course, in the case of an individual constituency Member in Scotland, the Bill requires a by-election to take place rather than a poll to decide whether or not to retain the former Member. Therefore, there is an element of consistency that leads to having a two-stage process in both cases.
The nature of the recall poll proposed in Wales
10. Under the Bill in question, the thing that triggers the process of re-electing a Member is not a recall petition, but a 'trigger event', which refers either to a conviction for a criminal offence that includes a term of imprisonment, or a decision by the Senedd (on the recommendation of the Standards of Conduct Committee) to start the recall process. The only role for voters, therefore, would be to vote for or against the dismissal of the Member. The result of authorising the dismissal of a Member would be to replace them with a Member of the same party. Clearly, there is a real risk that voters will see this process as a confusing and pointless one designed not to give them more control over their representatives, but to deprive them of such control.
11. The most basic criticism of the Bill, therefore, is that it seeks to apply a system that can only work within an electoral system based on electing individual Members, and that it cannot give voters a better voice in relation to the Members who represent them under a list system that does not allow for by-elections.
Another weakness
12. As the Bill (unlike the Scottish Bill) does not give voters, via a petition, the ability to trigger a recall, an alternative mechanism had to be found, namely the two types of trigger event.
13. The sentencing of a Member to a term of imprisonment is an event that, on the face of it, is quite definitive. Unlike the Westminster legislation and the Scottish Bill, however, the Welsh Bill does not engage effectively with the question of appeals against a conviction or sentence. Justice demands that a person should not lose their seat in the Senedd until all appeals have failed. A successful appeal can cancel out the trigger event, of course. Therefore, in Westminster (and in respect of the Scottish Bill), the process is not triggered until the possibility of an appeal has ended. In respect of the Welsh Bill, however, it appears that a recall poll may proceed even though an appeal is pending. If this interpretation is correct, it is impossible to defend it.
Summary
14. The Bill, as it stands, does not ensure a relatively simple or effective system for achieving the aim of making Members of the Senedd more accountable to the electorate. To a large extent, this stems from the fact that the Senedd has adopted a new electoral system that, in essence, reduces accountability by transferring the right to choose representatives from the public to the political parties. The details of the system that has been adopted mean that the pattern set out by Westminster’s Recall of MPs Act 2015 cannot be emulated. However, rather than seeking to devise a completely new system (or, perhaps, reconsider some problematic elements of the electoral system itself, such as the absence of by-elections) efforts have been made through the Bill to apply elements of that pattern to a completely different regime. As a result, the system proposed in the Bill risks confusing and frustrating constituents, leading to a perception that it has been designed to give a misleading impression of better accountability.
15. In the author's view, Part 1 of the Bill should not be taken forward. Instead, after the new electoral system comes into force, the opportunity should be taken to draw up a recall mechanism (or something with the same aim) that will take advantage of the experience accrued in implementing the new electoral system, along with any opportunity that may emerge to amend some elements of this at the same time.
Keith Bush
November 2025