MAB04 Professor Alistair Clark

Pwyllgor y Bil Atebolrwydd Aelodau | Member Accountability Bill Committee

Bil Senedd Cymru (Atebolrwydd Aelodau ac Etholiadau) | Senedd Cymru (Member Accountability and Elections) Bill

Introduction

1. My expertise is in electoral integrity and administration, with numerous published research articles and reports on these themes. I have also researched and published on issues around parliamentary conduct and standards. I write in a personal capacity (http://www.ncl.ac.uk/gps/staff/profile/alistairclark.html#background).

General Concerns

2. I have concerns about the speed with which this legislation is being brought forward. This is significant legislation which needs sufficient time to be both scrutinised, and, if it is successful, to be implemented. As it stands, implementing this legislation for use in the 2026 Senedd election seems to contravene the long-established Gould principle, that electoral legislation must be passed 6 months prior to the first electoral event that will be held under that legislation.[1] This is to ensure that candidates, political parties, electoral administrators and any other relevant actors are fully apprised of the legislation and its implications for that electoral event. The Gould principle is well recognised as best practice, and accepted by electoral administrators and the Electoral Commission.

3. Both the practice of recall, and the introduction of the issue in part 3 about making or publishing false or misleading statement are important pieces of legislation. It is more important to get these right than to have rushed and problematic legislation on the statute book in advance of the 2026 Senedd election. I therefore recommend that this legislation be paused and, should a new administration be minded, be reintroduced as part of the new Welsh government’s legislative programme in 2026.

Recall System – General Points

4. The Committee’s first question, Will the Recall system proposed in the bill be effective in achieving its aims of enhancing the accountability of Senedd members?, is difficult to answer. Ideally, recall should act as a deterrent to members’ misconduct. If this is the case, there should be no misconduct to regulate or to hold members accountable for. In this case, judging whether it might have been the crucial aspect in determining members’ (mis)behaviour and thus been effective is like proving a negative, and impossible to do.

5. What can be said is that the proposed recall legislation potentially gives the Senedd a further tool to hold members accountable. However, its use should only be seen as a last resort, after all other potential avenues for redress and accountability have been exhausted. This includes not only measures in the Senedd, but also disciplinary measures by political parties where appropriate.

Recall system - Practicalities

6. There are significant concerns around the second potential trigger event. Leaving the potential for a Recall Poll to depend on the future publication of ‘recall guidance’ seems ill-advised. While this may be a consequence of the speed which this legislation is being brought forth, leaving recall to depend on such ill-defined guidance would seem to potentially open the door to the Committee publishing different guidance for different cases, or, given that the Committee size will change post-2026, for a Committee with a particular composition to argue for recall guidance which suits political, rather than standards, interests.

7. This may be mitigated in the longer term by adding lay members to the Committee, as proposed in Part 2 of the Bill (See also para 23 below on this). However, the logic proposed remains circular. As paragraphs 92 and 93 of the Bill’s Explanatory Memorandum note, this is the Committee issuing guidance about the matters which the Committee itself must take into account. This seems ill-conceived.

8. Paragraph 92 of the explanatory memorandum then observes that the Committee must conduct a public consultation on the draft guidance. While public views will be undoubtedly welcome, this will take time if it is to be done properly. This will add to a lack of timeliness in dealing with any immediate case before the Standards Committee, and only lengthen the time before this guidance can be used.

9. This process seems unsatisfactory. Better to take the time to do this properly. I recommend that this second trigger mechanism be given much clearer consideration, and based in more objective criteria which is then enshrined in legislation instead of some indefinite guidance which may be forthcoming at some undefined future point. This should help to ensure that any decisions, or guidance, by the Committee in this regard is robust and might survive any challenge. The speed with which the Bill is being brought forward militates against this, but for this legislation to be well considered more time is required to weigh up the options for more objective criteria.

10. I have no immediate additional suggestions for such triggering criteria at this stage, except to say that these must be well linked to, and a clear progression of, the Senedd’s Code of Conduct and its menu of potential sanctions to begin with, rather than seeing recall as something vital and necessary in its own right. The Standards and Conduct Committee is currently considering wider aspects of the Standards process separately.

11. There seems to be an underlying assumption in the Bill and its accompanying documentation that political parties and third parties (as defined by the Electoral Commission) will not campaign in any recall poll. This seems somewhat optimistic. There will be incentives particularly for those campaigning to remove a member, but also for those who wish to retain their services, to campaign, and quite possibly to do so energetically. Even if political parties do not campaign in a particular recall poll, there would seem still to be potential for third parties to campaign for a particular outcome.

12. Whether it is right that regulation of such campaigns should be made by Welsh ministers, as the Bill currently provides, is an open question. Part 1, Section 11 seems widely drawn and delegates Ministers some important powers in this regard. Closer consideration of how recall polls might be regulated under Welsh circumstances is therefore necessary, and also whether these are powers that should be allocated to Ministers without oversight. 

13. The Committee’s sixth questions asks about whether the bill strikes an appropriate balance between the recall system in the Bill and what has been left to be filled in by regulations or subsequent guidance. Following from my points above, I would suggest that too much has been left to be done either by regulations or future guidance. I would recommend that a better balance be struck, with more detail contained in primary legislation.

14. Part 1, section 12 is clear that under the two-answer recall ballot proposed, a majority of votes cast, or 50%+1, will be the winning option. Recall polls are likely to be relatively low turnout electoral events. There is the possibility that a member is recalled on the basis of a relatively small vote in the constituency/region which they have been recalled. Whether this might be desirable would seem to be worth consideration since, on principle, the idea that a member can be removed by 50%+1 of a low turnout ought to be of concern to members. Although unusual, one option might be for a recall poll to achieve a certain level of turnout before a recall poll is successful. This would, however, be a policy matter for consideration by the Welsh government and Senedd.

15. Equivalent legislation currently before the Scottish parliament, the Scottish Parliament (Recall and Removal of Members) Bill, identified the distribution of the vote as a potential problem in recall processes being conducted under proportional representation rules. Closed list proportional representation, by necessity, operates at a larger scale than first past the post constituencies. Under the previous additional member system in the Senedd, the PR tier was based on regions, consisting of a number of constituencies. The new system in use from May 2026 re-labels regions as constituencies, with lists based on those areas. Regardless of the terminology, these are still much larger electoral areas than the previous first past the post constituencies.

16. The difficulty foreseen by the Scottish Parliament (Recall and Removal of Members) Bill is one where a strong local campaign in one area of a Scottish parliament proportional representation region was sufficient to get a recall petition over the threshold (10%) to begin to unseat a member. The solution offered was that three constituencies within the broader PR region would be needed to vote in favour of the recall petition to avoid local rivalries playing a part, and to ensure that there was wider demand for the removal of the MSP in question.[2]

17. While the Bill currently before the Senedd does not have a recall petition process, a recall poll will nevertheless be conducted at larger scale than, for example, the equivalent Recall constituency election under the House of Commons’ 2015 legislation. In a low turnout process, which I would expect a Senedd recall poll to be, there may be a potential danger of a localised campaign being effective in removing a member. Whether such an eventuality might be likely or how it should be dealt with requires some consideration by stakeholders, in conjunction with the points about campaigning made above.                

18. The one-day electoral process is welcome, and preferable to a two-stage process including a recall petition. Its design echoes the more familiar by-election process in first past the post constituencies, albeit with a binary choice before voters. These recall processes will still however be conducted at larger scale, and need more resources – funding, polling and count staff, locating polling stations, postal votes etc – than the by-elections which are more familiar. Local authorities already struggle to staff electoral events. Consideration is necessary, with input from the Electoral Management Board, of how Constituency Returning Officers may do so, given that the potential requirements and disruption (e.g. where Schools might be used as polling places, more Schools will be involved) are greater than previously.  

19. Ballot paper design for a Recall poll will need careful consideration, beyond the wording that is already in the Bill. There is political science evidence that voters complete ballot papers from top to bottom. This can lead to bias in results, with options towards the top of the ballot paper more likely to be chosen. A vertical design, with the retain option above the remove option (or vice versa) would seem to imply a preference for whatever option is placed above the other. Avoiding this difficulty would require the two options to be placed horizontally, side by side.

Committee Membership

20. The proposal to introduce Lay Members to the Senedd’s Standards Committee is welcome. Lay Members have benefited standards processes in several Committees in the UK parliament by bringing an element of independent expertise to standards regulation and oversight.[3] The general provisions around Lay Members in the Bill seem sensible.

21. The key question of how many Lay Members to appoint has been left unaddressed by the legislation. To some degree this is inevitable, given the institutional changes that will happen from the 2026 election. It is however possible to anticipate potential outcomes in this regard, since the overall size of the future Senedd is already known, and the Committee is likely to be larger than it currently is. This should have been done.

22. It is correct that the number of Lay Members should not exceed the number of elected members (Para 118 explanatory memorandum). At the same time, it is important that Lay Members do not feel pressurised or outnumbered by elected members into taking one course of action over another, which could happen if there are fewer lay members than elected members. This speaks to an equivalent number of Lay Members and elected members in the future Standards Committee.

23. The question of Lay Members’ role in the recall guidance process (trigger event 2) envisaged by the Bill is unclear. This is currently a key process in the recall process that would benefit from independent oversight. Consequently, assuming the Bill is proceeded with, any such guidance must be informed by Lay Members’ views. This suggests either that initial guidance is reissued after the appointment of Lay Members, or alternatively the Committee waits until Lay Members are appointed before drafting such guidance. This seems to underline the difficulties with the recall guidance process in trigger event 2 highlighted above which seem to be caused by the haste to legislate for recall.

24. Extending the power for the Commissioner for Standards to undertake their ‘own initiative’ investigations seems reasonable, and in keeping with developing practice elsewhere in the UK.          

25. One curiosity of the Senedd’s Standards of Conduct Committee is that Senedd Standing Orders (SO. 22) do not prevent the Chair of the Standards Committee from being from the same party as the Welsh government or largest party. This is out of keeping with practice in the other legislatures in Great Britain. In the House of Commons, the Chair of its Standards Committee is selected from the largest opposition party (currently held by a Conservative MP). This is also the case in the Scottish parliament (currently held by Labour). This adds a further degree of independence to the Standards process in both institutions.

26. The addition of Lay Members to the Senedd’s Standards and Conduct Committee presents an opportunity to consider the functioning of the Committee more widely. I would therefore recommend either the current Bill consider legislating for the Standards Committee Chair to be selected from the main opposition party, or, if more applicable, for Standing Orders to be amended accordingly. Combined with the inclusion of Lay Members and the power for ‘own initiative’ investigations for the Commissioner, this will add to both the practice and the perception of the independence of the Senedd’s Standards regulation.