Framework Legislation

Professor Richard Whitaker[1]

University of Leicester and UK Parliament

1. Introduction

Framework bills, sometimes known as skeleton bills, are those in which the bulk of policy is made via delegated powers rather than appearing on the face of a bill. This paper assesses how much this type of legislation is used in UK legislatures, why it is used, and the consequences for parliamentary scrutiny. The paper draws on academic research and reports and debates from the UK’s legislatures.

Governments may have an incentive to use framework legislation as it enables them to delay the making of policy where details have not been worked out before a bill is introduced. It can also be used to enable ministers to respond to unknown future events by giving them flexibility to make policy changes without having to introduce new primary legislation.

Delegated powers bring problems for parliamentary scrutiny though. In many legislatures, the procedures for scrutinising delegated legislation are much less rigorous than those for primary legislation. This may not be problematic for uncontentious policy details. However, if the main elements of a policy are decided via delegated legislation, legislatures are left with much less of an opportunity – compared with primary legislation – to discuss and assess policy before it becomes law. There is also the possibility that delegated powers will be used in ways not expected when the relevant enabling legislation was passed.

Some academic research (Carey and Shugart 2009; Huber and Shipan 2002) argues that delegation should be much less likely when there is a coalition or minority government. In the first scenario parties may fear their coalition partners will use secondary legislation to shift policy away from a coalition agreement. In the second, the majority of legislators will not want to delegate power to a minority to make policy with little scrutiny. Where the executive is in a strong position though, party discipline in a governing party may assist executives in ensuring framework bills are passed, even where governing-party legislators have misgivings.

In its 2021-22 and 2022-23 annual reports, the Senedd’s Legislation, Justice and Constitution Committee (2022d, 2023c) drew attention to the use of framework bills by the Welsh Government. In the first of these two reports the committee expressed its ‘belief’ that in some cases ‘the balance of the legislation is tilted too much in favour of executive power which marginalises the democratic mandate of the Senedd’ (paragraph 18). These and other examples are covered in more detail below.

Committees in the UK Parliament have lamented the increasing use of skeleton bills at Westminster. While acknowledging that these types of bills are not new, the chairs of three House of Lords committees pointed out that in response to Brexit and Covid-19, the UK government had introduced bills that were:

‘extraordinary in terms of the extent to which they have permitted a shift of power from the legislature to the executive. In many cases they have given ministers extraordinarily wide powers, powers which have, in some cases, been conferred by primary legislation which is nothing short of skeletal’ (Hodgson et al. 2020).

The House of Lords Delegated Powers and Regulatory Reform Committee (DPRRC 2021a) went further in its report Democracy Denied, saying that the boundary between primary and secondary legislation matters for protecting parliamentary democracy. They argue if governments make too much use of delegated powers within a bill, Parliament’s integrity may be undermined.

Nevertheless, there are procedures used in other legislatures that can help to reduce some of the problems associated with framework legislation including opportunities to amend or veto secondary legislation before it comes into force. These are discussed in more detail below.

This paper will consider the significance of governments’ use of framework legislation in the UK and how this affects parliamentary scrutiny. Beyond this introduction, the paper is structured under the following headings:

2. Definitions of framework or skeleton bills

3. How prevalent is framework legislation in legislatures across the UK?

4. When might skeleton or framework bills be appropriate?

5. How do UK legislatures scrutinise framework bills and statutory instruments?

6. The drawbacks of framework legislation

7. Examples of best practice in scrutinising framework legislation

8. Conclusion

Appendix: Procedures for scrutiny of secondary legislation in the UK’s legislatures

2. Definitions of framework or skeleton bills

Framework bills, often known as skeleton bills, are defined by the House of Lords’ Delegated Powers and Regulatory Reform Committee (2021: 26) as those ‘where the provision on the face of the bill is so insubstantial that the real operation of the Act, or sections of an Act, would be entirely by the regulations or orders made under it’. The same committee offered an earlier definition of skeleton bills in its Special Report for Session 1998-99 as ‘Bills which are little more than a licence to legislate and so give flesh to the "skeleton" embodied in the Bill’. In other words, framework bills are those where the content of policy is almost entirely decided through secondary legislation.

A similar definition has been applied by the DPRRC to parts of bills where certain sections delegate extensive powers to government ministers and, for that part of the bill, provide little indication of policy content.

These definitions are qualitative and rely on judgements about how far policy content can be determined from the text of a bill. It is difficult to develop a quantitative measure of framework legislation at least for comparative purposes. In the academic literature on delegated powers, Epstein and O’Halloran (1999) calculate a delegation ratio by taking the number of sections in a bill that delegate powers and dividing it by the total number of sections in the bill. The difficulty of using this measure is that, in the UK case, there has been an increase in the number of sections in bills over time but these sections have become shorter on average (Kosti 2023: 8). In addition, there has been a decline in the number of statutes produced at Westminster per year in the post-war period but an increase in the total number of words taken up by these (Williams 2018: 46). These issues make it difficult to devise a quantitative definition of framework bills that allows for meaningful comparison to be made.

2.1 Skeleton, framework and headline bills

Some parliamentarians draw distinctions between skeleton and framework bills. Framework bills offer more detail about how delegated powers might be used than do their skeletal counterparts. This distinction was made by several legislators during the passage of the Medicines and Medical Devices Act 2021 at Westminster. Speaking for the UK government during this bill’s report stage in the House of Lords (HL Deb 12 January 2021, c.666), Lord Bethell said ‘noble Lords suggested that the Bill needed to move from a skeleton Bill to a true framework Bill.’ Later, when the House of Commons was considering Lords’ amendments (HC Deb 27 January 2021, c.487), the government minister, Jo Churchill, said the bill had been changed from ‘a skeleton Bill into a true framework Bill that makes it clear how delegated powers will be used’. Nevertheless, the terms ‘framework’ and ‘skeleton’ appear to be widely used interchangeably in many discussions of bills identified in Westminster at least, as framework bills.

Welsh Government ministers have recently referred to a further type of bill in the form of headline bills. At a meeting of the Legislation, Justice and Constitution committee on 25 September 2023, the Minister for Climate Change used this term to describe the Infrastructure (Wales) Bill: ‘This is a headline Bill, because it sets out a detailed structure of the process and what will be caught by the Bill, and also contains a number of quite narrowly drawn regulation-making powers, which allow appropriate detail to be put in. But all the detail isn't on the face of the Bill’. She distinguished headline from framework bills by defining the latter as including very little policy detail and many delegated powers, often Henry VIII powers.

3. How prevalent is framework legislation in legislatures across the UK?

Skeleton bills are not new. They are mentioned in the 1932 Donoughmore report of the Committee on Ministers’ Powers. The report refers to Acts of Parliament that are passed ‘only in skeletal form…with the result that laws are promulgated which have not been made by, and get little supervision from Parliament’. According to the report, critics at the time said skeletal acts had ‘assumed the character of a serious invasion of the sphere of Parliament by the Executive’ (p.53).

To understand how prevalent framework bills are, we need a clear definition of framework legislation. Given the difficulty of defining this quantitatively, as outlined in the previous section, qualitative definitions make this task more manageable. One approach is to defer to the expertise of the relevant committees in the UK’s legislatures and measure the amount of framework legislation by references to it in committee reports on bills.

3.1 Framework bills in the UK Parliament

Although the electronically available data do not allow us to assess the presence of these types of bills as early as the 1930s when the Donoughmore report was published, we can go back to the early 1990s. The House of Lords DPRRC’s reports are a good place to look for measuring the presence of framework bills at Westminster. When the UK Government produces a bill, it must supply a Delegated Powers Memorandum, setting out the powers taken in a bill and offering a justification for them and for the parliamentary-approval procedures (if any) through which the resultant statutory instruments will be considered. The DPRRC examines these memoranda and reports on them, offering recommendations to the government. The committee’s reports are frequently cited in debates on bills during the legislative process, especially among members of the Lords. DPRRC reports are available online back to the 1997/98 session but a special report from the 1998/99 session helpfully lists all bills the committee deemed to be skeletal from the time of its creation in 1992/93 (Delegated Powers and Regulatory Reform Committee, 1999).

To gather information about the prevalence of framework bills at Westminster, the content of all DPRRC reports from the 1997/98 session up to the 2022/23 session was searched for references to skeleton or skeletal bills or parts of bills. The committee normally uses these terms rather than ‘framework bills’ and, as noted above, provides a definition of what constitutes a skeleton bill. Figure 1 shows how many bills in each year were designated by the committee as skeletal either in full or in part from the 1991-2023. Note that this is quite a conservative measure. Searches of Hansard reveal bills not included in these data which were labelled as skeletal by MPs or peers if not by the DPRRC. In these cases, often the committee had not labelled a bill as skeletal for one of two reasons: (1) because the government had taken action to mitigate problems (discussed in more detail in the Best Practice section later) or (2) that the problem was not that a bill was skeletal but that one or more powers in the bill were very wide ranging and, in the committee’s view, inappropriate for delegated legislation.

Figure 1 Number of skeleton bills or bills with skeletal sections introduced in the UK Parliament, 1991-2023 (based on DPRRC reports)

While the number of skeleton bills identified by the DPRRC is not that large, the trend is broadly upward and seems to change after the 2016 referendum on the UK’s membership of the European Union. Figure 1 shows numbers of such bills by year, rather than by session, in order to take account of the varying lengths of sessions. The average number of skeleton bills per year in the 1991-2023 period is 1. The average number per year from 1991-2015 is 0.7 but this triples to 2.1 in the 2016-2023 period. Even if we simply split the time in half chronologically, the mean number of skeleton bills in the first 17 years in the data (0.6) is lower than that for the second 17 (1.4). Nevertheless, the figures fluctuate as can be seen be the absence of any bills in 2023 labelled by the DPRRC as skeletal.

Considering these numbers as a proportion of all government bills introduced, and using parliamentary sessions as the unit of time,[2] the 2022-23 session saw 13% of bills falling into the skeleton category. The next highest figure is 9% for the 2017-19 session.[3] The time frame here is limited and so we should exercise caution as to whether this trend will continue. The data do seem to indicate though, that these bills are occurring in small numbers but more frequently, on average, over time and particularly since 2016, notwithstanding the lack of skeleton bills in 2023.

In a recent working paper, the Hansard Society (2023) argue framework legislation provides its own momentum. As more delegations of power are included in Acts of Parliament, governments can more easily argue for further delegations given precedents set in previous primary legislation. An example of this, to which the Hansard Society drew attention, is the Tobacco and Vapes Bill. The UK Government justified the delegation of powers on tobacco, nicotine and vape products and their packaging in this bill on the basis that similar powers were introduced for tobacco products and packaging in the Children and Families Act 2014 (England, 2024).

Another way of looking at this is by examining the enabling acts that have been the source of statutory instruments (SIs) passed by the UK Parliament in recent years. The data in Table 1 are drawn from all SIs passed in the UK Parliament from 2017-2023, taken from data in the UK Parliament’s data platform (api.parliament.uk).

Unsurprisingly, the European Union (Withdrawal) Act 2018 is the source of by far the most SIs since 2017. This was a mechanism by which retained EU law was transferred to the UK statute book. The European Communities Act is also high up the list as the enabling act for the implementation of much EU law while the UK was a member. Aside from the outlying case of the EU (Withdrawal) Act, there is no clear concentration of SIs from Acts passed in the most recent years. The Public Health (Control of Disease) Act 1984 was the source of many SIs made over 30 years after the act was passed, during the Covid-19 pandemic.

Academic research on the nature of delegated powers in the UK has examined how far they contain permissive (the Secretary of State ‘may’ make regulations) versus mandatory (the Secretary of State ‘shall’ or ‘must’ make regulations) delegations of power. Based on all UK legislation from 1900-2020, Kosti (2023) shows that the numbers of permissive delegations of power in bills rose from 228.7 on average per year in the 1970s to 428.2 per year in 2000s. The numbers are lower during the 2010-20 period when there was a coalition and later a minority government. Nevertheless, the upward trend from the 1970s to the 2000s is significant because it indicates delegation that gives the executive freedom to decide whether and when to use powers granted.

Table 1  Top 20 enabling acts based on number of SIs passed, 2017 - September 2023

Act	SIs	Year of Enabling Act
European Union (Withdrawal) Act 2018	807	2018
European Communities Act 1972	379	1972
Local Democracy, Economic Development and Construction Act 2009	201	2009
Public Health (Control of Disease) Act 1984	198	1984
Taxation (Cross-border Trade) Act 2018	140	2018
Social Security Contributions and Benefits Act 1992	137	1992
National Health Service Act 2006	97	2006
Sanctions and Anti-Money Laundering Act 2018	93	2018
Social Security Administration Act 1992	93	1992
Social Security Contributions and Benefits (Northern Ireland) Act 1992	68	1992
Merchant Shipping Act 1995	64	1995
Public Service Pensions Act 2013	64	2013
Welfare Reform Act 2012	61	2012
Value Added Tax Act 1994	51	1994
Legal Aid, Sentencing and Punishment of Offenders Act 2012	49	2012
Courts Act 2003	47	2003
Town and Country Planning Act 1990	46	1990
Local Government Finance Act 1988	44	1988
Scotland Act 1998	44	1998

Source: UK Parliament GitHub

3.2 Framework bills in the Senedd

There is evidence of framework bills in the Senedd stretching back to the 2007-11 term. In February 2011 the then National Assembly for Wales Constitutional Affairs Committee reported on the drafting of Welsh Government Measures in the first three years after the then Assembly gained the power to pass primary legislation. The committee explained that a number of these Measures were skeletal in nature ‘with the detail to be filled in later through subordinate legislation’ (Constitutional Affairs Committee 2011, paragraph 23). In the Fourth Senedd, the Constitutional and Legislative Affairs Committee’s inquiry on law-making during 2011-16 was partly set up due to a trend towards bills being brought forward ‘before the policy had been fully developed and leaving important details to be brought forward by subordinate legislation at a later date’ (Constitutional and Legislative Affairs Committee 2015, paragraph 12). The committee argued that much of the detail contained in the Welsh Government’s Statements of Policy Intent could have been added to the text of bills. This would provide a legal basis to the government’s plans for how delegated powers were to be used. A report of the Committee’s successor in the Fifth Senedd on the Landfill Disposals Tax (Wales) Bill (Constitutional and Legislative Affairs Committee 2017) sets out several areas of policy that were left to delegated powers but which the committee felt should be included on the face of the bill.

In the Sixth Senedd, the LJC committee’s report on the Welsh Tax Acts etc. (Power to Modify) Bill refers to the problems of skeleton bills and cites academics’ views, Senedd committee reports and reports by the DPRRC and House of Lords Secondary Legislation Scrutiny Committee as evidence of the problems. Evidence to the committee from Sir Paul Silk indicated that powers to impose taxes should normally be made by primary rather than delegated legislation (Legislation, Justice and Constitution Committee 2022b).

The LJC committee’s 2022-23 annual report identifies a further three framework bills (or bills that have framework elements): the Agriculture (Wales) Bill, Health Service Procurement (Wales) Bill and the Environment (Air Quality and Soundscapes) (Wales) Bill (Legislation, Justice and Constitution Committee 2023c). Since then, the committee has identified two further framework bills: the Infrastructure (Wales) Bill and the Local Government Finance (Wales) Bill. Taken together, these six bills constitute 43% of the fourteen bills introduced in the sixth Senedd by March 2024. This is a much higher proportion than in the UK Parliament but is based on a far smaller number of bills overall.

3.3 Framework bills in the Scottish Parliament

As in the Senedd, there is evidence of framework legislation in the Scottish Parliament from the 2011-16 term. The Delegated Powers and Law Reform committee (DPLRC) of the Scottish Parliament noted in its report on the 2011-16 term that the numbers of framework bills increased during that period. They stated that ‘in most instances they are… bills where it appears that the policy is still to be fully developed’ (Delegated Powers and Law Reform Committee, 2016, paragraph 51) and that delegated powers should not be used ‘simply to give the Scottish Ministers flexibility’ when they have not fully worked out the detail of a policy (Delegated Powers and Law Reform Committee, 2016, paragraph 53).

The same committee’s report on the Legislative Consent Memorandum on the European Union (Withdrawal) Bill explained the committee had ‘consistently raised concerns about the use of so-called framework legislation’ (Delegated Powers and Law Reform Committee, 2017a, paragraph 27). The committee drew attention to the extremely wide-ranging powers granted to ministers via this bill although they accepted the need for them given the unprecedented size of the task at hand (adjusting UK legislation to ensure a coherent statute book after the UK had left the EU). The DPLRC’s report on use of the made-affirmative procedure (Delegated Powers and Law Reform Committee, 2022a) took evidence on the use of framework bills. Witnesses pointed to the increased use of these bills, a claim that was denied by then Deputy First Minister John Swinney in his evidence to the committee.

The importance of this issue to Scottish Parliament committees, was clear from the Conveners Group’s[4] questioning of the former First Minister on 27 March 2024. They began with questions about framework bills. Points were raised about the implications of these for parliamentary scrutiny and the need to provide more detail on how delegated powers will be used.

While it is difficult to quantify the number of framework bills in the Scottish Parliament, partly due to variations in the terminology used to describe these types of bills over time, there are some prominent examples from reports of the DPLRC. They identify at least two bills from the 2011-16 term (Land Reform (Scotland) Bill and the Burial and Cremation (Scotland) Bill) as framework bills in a letter from the chair of the committee to the Commission on Parliamentary Reform (Delegated Powers and Legal Reform Committee, 2017b).

During the sixth term of the Scottish Parliament, in 2022, the DPLRC reported on the Coronavirus (Recovery and Reform) (Scotland) Bill and drew attention to the inclusion of five delegated powers to which the made-affirmative procedure had been applied (Delegated Powers and Law Reform Committee 2022b). The committee asked Scottish ministers to supply written statements giving evidence for the need to introduce measures urgently, before the relevant statutory instruments come into force. They also asked that each measure have a sunset provision and that ministers include an assessment of the impact of each measure on those it would affect. The Scottish Government conceded on all these points except for the impact assessments.

The National Care Service (Scotland) Bill, introduced in 2022 is described by the Health, Social Care and Sport committee as a framework bill. The committee’s summary of responses to its call for views on the bill, says that among respondents ‘There is widespread frustration at the overall lack of detail because the Bill is a framework bill, and this…poses challenges for scrutiny’ (Health, Social Care and Sport Committee, 2022, p.4). Respondents also argued that the lack of policy detail in the bill meant there were wide margins of error in the estimates of costs, adding uncertainty to the outcomes that might result from use of delegated powers.

The DPLRC report on the Circular Economy (Scotland) Bill says ‘The Bill will primarily deliver enabling powers that will set a framework for taking action into the future’ (Delegated Powers and Law Reform Committee 2023, paragraph 4). Although this appears to be a framework bill, the committee agreed with the bulk of the delegated powers. This acceptance results from a range of factors including that sufficient restrictions on the powers are stated on the face of the bill, enough information is provided about how the power will be used, there are precedents for similar use of delegated powers elsewhere or there is a requirement to consult affected parties before bringing forward policy detail.

In its Policy Memorandum, the Scottish Government labelled the Agriculture and Rural Communities (Scotland) Bill, published in January 2024, as a framework bill (Scottish Government 2023). The DPLRC’s (2024) report on the bill drew attention to the problems of this, noting that ‘the total amount of funding to be allocated [to the rural support plan proposed by the bill] and the breakdown of this funding is not known ahead of the legislation being passed’ (paragraph 29). In its report on the bill, the Rural Affairs and Islands Committee (2024) accepted the need for a framework bill on this issue but drew attention to the implications for parliamentary scrutiny, given the lack of detail on how powers in the bill might be used.

This is not an exhaustive list so it is difficult to be precise about the proportion of bills that fall into the framework category in Scotland. At the time of writing, 40 government bills had been introduced in the sixth term, with four identified as framework. Roughly speaking, this percentage of framework bills in Scotland since 2021 (10%) is lower than in the Senedd (43%) and slightly higher than that at Westminster for the 2021-22 and 2022-23 sessions combined (9%).

3.4 Framework bills in the Northern Ireland Assembly

Framework bills in the Northern Ireland Assembly (NIA) are not expected to occur frequently on the basis that the NIA’s decision-making rules require broad support for any legislation to be passed. Delegating many policy-making powers to a minister in a single bill is not likely to be an attractive option in this scenario.

4. When might skeleton or framework bills be appropriate?

The academic literature on delegation of power sets out several reasons why legislators might delegate policy-making power to executives. One theme in studies of this subject is time constraints. If there is an urgent need to implement legislation, primary legislative procedures may be too slow and so parliamentarians may be more willing to delegate (Carey and Shugart 2009: 18, Epstein and O’Halloran 1999: 49).

Levels of trust among members of a government may also affect how far legislators are willing to delegate policy-making power. According to this approach, the motivation to include policy detail in a bill will be at its highest when members of government do not trust each other to implement their desired policies (Huber and Shipan 2002: 185). More broadly, this is particularly important for coalition governments. Coalition partners will want to make sure they each stick to agreements reached about policy. According to this approach, we should  see more policy detail in laws when there are coalitions in government. This explanation would fit well with the case of the Northern Ireland Assembly (as noted in the previous section) where governments are always coalitions of parties from the unionist and nationalist communities.

If there is a minority government in office, the academic literature suggests we should  see more detail in primary legislation because non-governing parties will not want to delegate policy-making power to ministers who could use this power to shift policy away from what a majority in the legislature wants to see (Carey and Shugart 2009; Huber and Shipan 2002).

Beyond the academic literature, the Welsh and UK governments state the circumstances under which they argue delegated powers may be acceptable. The Welsh Government’s principles of when suchpowers might be acceptable are set out in its Legislation Handbook on Assembly Bills (paragraph 10.1). These include that:

· the matters in question may need adjusting more often than it would be sensible for the Assembly to legislate for by primary legislation;

· there may be rules which will be better made after some experience of administering the new Act and which it is not essential to have as soon as it begins to operate;

· the use of delegated powers in a particular area may have a strong precedent and be uncontroversial;

· there may be transitional and technical matters which it would be appropriate to deal with by delegated powers.

The UK Cabinet Office Guide to Making Legislation sets out similar but slightly wider-ranging advice on when the UK government sees the delegation of powers as appropriate. These circumstances include:

· Filling in the detail of minor, technical or administrative matters.

· Situations when amendments to legislation might be needed more frequently than could reasonably be carried out by Parliament through the primary legislation process.

· When consultation is needed on the detail of a policy such as the level of a fee.

· When operating in a new area of policy to give ministers “an acceptable level of flexibility” to make changes in the light of experience.

· To allow flexibility for policy to be made differently for different groups or areas.

· Where there are precedents for using delegated powers and where it is uncontroversial to do so.

Some of the circumstances set out above are evidenced in reports from the DPRRC. For example, part of the Land Registration Act 2002 sets out a legal framework for the introduction of electronic conveyancing. In this case, stakeholders recommended that the government needed flexibility to take account of future developments in implementing this policy. Given this and the detailed and uncontroversial nature of the delegated powers to introduce electronic conveyancing, the DPRRC felt this bill was acceptable (Delegated Powers and Regulatory Reform Committee 2001).

Similarly, when considering the Pension Schemes Act 2015, the DPRRC noted that there were strong precedents for delegating many powers in this area of policy and did not dismiss the government’s arguments about the need for flexibility to accommodate changes in the pensions market (Delegated Powers and Regulatory Reform Committee 2014b). This is surely an example of a complex policy area that requires expertise on the part of policy makers and where the delegations are on matters of detail. Given this, the DPRRC was more concerned with the scrutiny procedures to be used for regulations made under the Act than with the amount of delegated powers it contained.

There are also examples of more policy detail being provided by governments under pressure from parliaments as a framework bill passes through the legislative process. The UK government made changes in response to DPRRC reports on the problems of introducing a Community Infrastructure Levy as part of the Planning Act 2008 (Greenberg 2011: 210). This levy is paid by owners or developers of land, the value of which might increase as a result of housing developments. According to the DPRRC’s first report on this bill, too many of the features of the levy were to be established by powers delegated to ministers. The UK Government responded by introducing amendments providing detail on how the levy would work in practice. The DPRRC noted in two reports (13^th and 15^th reports of the 2007-08 session, 2008b, 2008c) that these amendments made the relevant part of the bill much less skeletal in nature, making it more acceptable to the committee.

Nevertheless, parliaments do not have to agree with government guidance on when the use of skeleton bills or delegated powers more broadly might be acceptable. In its report on the Infrastructure (Wales) Bill, the LJC rejected the Welsh Government’s argument about some of the delegated powers in the Bill. Specifically, ministers were delegated the ability to define the kinds of infrastructure developments for which consent might be sought. The Government argued this was needed to allow for flexibility to respond to future situations not yet anticipated (LJC Committee 2023d, paragraph 138). While this might have been consistent with some of the guidance in the Legislation Handbook on Assembly Bills, the committee argued the Government had not provided ‘robust reasoning’ for the delegation and that it was not ‘appropriate that the Senedd will have no role in approving its use’ (paragraph 154).  

A similar point was made by the LJC’s predecessor with reference to the Counsel General’s guidelines on procedures for scrutinising delegated legislation. In its report on the Higher Education (Wales) Bill in 2014, the Constitutional and Legislative Affairs Committee (2014) discussed the scrutiny procedures to be used for statutory instruments enabled by the Act. Paragraph 25 of the committee’s report stated:

We acknowledge that the Welsh Government will use the Counsel General‘s guidelines on the choice of affirmative or negative procedure in drafting legislation. While they are extremely helpful, we will come to our own view on a particular provision based on a range of factors, including our own analysis of the Bill and the context of each provision.

Similarly, the DPRRC in the UK Parliament, has its own guidance on how skeleton legislation should be dealt with and the limited circumstances in which it might be acceptable (see section 7 below).

The following section assesses how UK legislatures scrutinise framework bills and statutory instruments in general.

5. How do UK legislatures scrutinise framework bills and statutory instruments?

5.1 Procedures for framework bills

The Senedd’s Legislation, Justice and Constitution committee considers all bills introduced in the Senedd. This includes examining ‘the balance between the information that is included on the face of the Bill and that which is left to subordinate legislation’.[5] This mechanism allows the committee to establish how skeletal each bill is. The committee’s evidence sessions with relevant ministers as part of Stage 1 of the legislative process can be used to question the government on this point. The LJC also draws attention in its annual reports to the use of framework bills.

The Welsh Government’s (2019) Legislation Handbook on Assembly Bills does not refer explicitly to framework or skeleton bills but does so implicitly, stating that when drafting legislation, ‘Careful thought will need to be given to whether the matters, although detailed, are so much of the essence of the Bill that the Assembly ought to consider them along with the rest of the Bill’ (paragraph 10.2a). The Handbook provides advice on the provision of draft statutory instruments and states in chapter 11 that ‘it is helpful’ for the Senedd to see draft regulations that are ‘central to a bill’s effect’ (11.1). Ideally the relevant committee should be able to consider draft regulations at Stage 1 of the legislative process. If this cannot be done, then a statement of policy intent ‘must be prepared’ (11.7).

In the UK Parliament, the process of scrutinising framework bills is fundamentally the same as for any other type of bill. These bills do not have any special status in the Standing Orders of either House. Nevertheless, the DPRRC asks the government to make a declaration when a bill is a skeleton bill. The Cabinet Office Guide to Making Legislation offers a definition of what it calls ‘framework (or ‘skeleton’)’ bills or provisions. It states that ‘legislation in this form will need to be fully justified to the DPRRC. In these cases it is helpful to explain the steps taken to include policy detail, limitations on the power and appropriate safeguards on the face of the bill’ (Cabinet Office, 2022, paragraph 15.19). Similar to the advice given in the Legislation Handbook cited above, the Cabinet Office Guide also points out that MPs and peers ‘will find it helpful’ (paragraph 15.35) to see draft versions of regulations at the committee stage in the legislative process.

5.2 Scrutinising statutory instruments in UK legislatures: differences from Senedd procedures

The Appendix provides summaries of procedures used to scrutinise statutory instruments (SIs) in the legislatures of the UK. This section draws attention to two procedures that differ from the Senedd’s approach.

5.2.1 Consideration of Scottish Statutory Instruments under the affirmative procedure in the Scottish Parliament

When the Scottish Parliament considers a Scottish Statutory Instrument (SSI) under the affirmative procedure, the relevant Scottish Minister is required to propose a motion to the lead committee (meaning the committee responsible for the relevant policy area) that the SSI be approved. The minister gives evidence to the committee which then reports to the Scottish Parliament. This procedure is significant because it differs from how affirmative SIs are considered in the UK and Welsh parliaments in that the lead (subject-specific) committee has the opportunity to question the relevant Minister as part of the standard procedure. More details of this are in the Appendix.

5.2.2 The SL1 Letter procedure in the Northern Ireland Assembly

The distinctive aspect of the Northern Ireland Assembly’s (NIA) procedure for considering what are called statutory rules (SRs rather than SIs) is that committees consider proposals for SRs ahead of them being laid before the Assembly. Proposed SRs are sent to the policy-relevant committee (rather than a delegated legislation committee) via what is known as an SL1 Letter. This letter must explain the procedure to be used for the SR and its purpose as well as the results of any consultation. The committee can scrutinise the statutory rule from a policy as well as a legal perspective. This process is derived from the statutory duty of NIA committees to ‘advise and assist each Northern Ireland minister in the formulation of policy’ (Northern Ireland Act 1998, section 29(1)(a)).

Ministers do not have to accept changes suggested by the committee and may be less likely to implement them if they are certain of support from a broad range of Members of the Legislative Assembly (MLAs) for their version of the SR. In addition, committees have wide remits and limited resources and this can mean that SL1s may proceed quite quickly through committees. The technical aspects of scrutiny are delegated to the Examiner of Statutory Rules who uses similar tests of the SR’s legality to those used by staff in the JCSI.

6. The drawbacks of framework legislation

These descriptions of statutory instrument procedures hint at the problems of limited opportunities for scrutiny compared with that of primary legislation. This section looks in more depth at these and other drawbacks of framework legislation and the extensive use of delegated powers.

6.1 Inability to scrutinise policy detail

The principal drawback of framework legislation is that it results in much policy being made via statutory instruments, the scrutiny of which is far more limited than that of primary legislation. This fundamental problem was set out in a letter from the chairs of three House of Lords committees (SLSC, DPRRC and the Constitution Committee) in 2020 to government ministers. The committee chairs argued that that skeleton bills mean policy development is limited at the point bills are being scrutinised and that this is ‘detrimental to good government as well as effective parliamentary scrutiny’. They also said:

Without substantive provision on the face of the Bill, Parliament is being asked to pass legislation without knowing how the powers conferred may be exercised by ministers and so without knowing what impact the legislation may have on members of the public affected by it. It is not enough for the Government to give assurances to Parliament about how the powers will be exercised…powers delegated to ministers have to be assessed not by how the Government say they will be exercised but by how they could be exercised by future administrations (Hodgson et al. 2020).

In its 2014 study of delegated legislation based on interviews with MPs, peers, and staff from both Houses of the UK Parliament, the Hansard Society explain how framework legislation sometimes comes about due to a combination of disagreements over policy and time pressure on governments. This can lead drafters to write delegated powers into a bill rather than including policy detail. If there is a lack of clarity about how the power might be used, then it may be difficult to draft anything other than a wide-ranging power. This would likely slow down the passage of the bill because legislators may object to the breadth of the powers (Fox and Blackwell 2014, p.59) and may reason that widely drawn powers introduce the possibility of the delegated legislation being used in an unanticipated way.

Elsewhere, the Hansard Society has argued that the advice set out by the Cabinet Office on framework bills is not followed consistently. This advice, set out earlier, is to explain how delegated powers in framework bills have been limited and to provide draft versions of regulations for parliamentarians to see during the legislative process. The Hansard Society argues that if this advice was closely followed, many of the problems described above would be greatly reduced (Hansard Society 2023, p.9).

6.1.1 Examples from the Senedd

The Agriculture (Wales) Act 2023 provides a prime example of the inability to scrutinise policy detail in framework bills. In its report on the bill, the LJC (2023a) committee pointed to a range of problems. First, they noted that significant elements of two parts of the bill ‘fall into the category of an enabling bill’ (paragraph 64). Second, concerns about how delegated powers might be used were heightened by the Welsh Government’s justification of a framework approach on the grounds that the bill will be in place for several decades. These powers could be used to make major policy changes well into the future without adequate scrutiny (paragraphs 66 and 71). An example of these concerns was raised by the Economy, Trade and Rural Affairs Committee. In its report on the bill, attention was drawn to a section that allows the definition of agriculture to be altered by statutory instrument. This meant that ‘the entire scope of this framework Bill for Welsh agriculture policy could be changed by Welsh Ministers in future by subordinate legislation’ (Economy, Trade and Rural Affairs Committee 2023, paragraph 23).[6] Third, the LJC committee rejected the Welsh Government’s argument that the affirmative procedure for statutory instruments provides for ‘significant’ or ‘full scrutiny’. This is on the grounds that the procedures for scrutinising delegated legislation are far more limited than those for primary legislation (paragraphs 83-84), as the next section goes on to discuss.

LJC committee reports on other bills highlight the issue of the Welsh Government declining to provide draft versions of regulations, despite the advice in the government’s Legislation Handbook on Assembly Bills (chapter 11). The LJC committee requested that the Welsh Government publish a draft of the ‘public services outsourcing and workforce code’ prior to the Social Partnership and Public Procurement (Wales) Act 2023 reaching Stage 3 of the legislative process (Legislation, Justice and Constitution Committee 2022c: paragraph 67). The Government declined to do so stating that it needed to consult with stakeholders first (Blythyn 2022).

Similarly, the LJC committee requested the Welsh Government publish drafts of regulations enabled by the Tertiary Education and Research (Wales) Act 2022 so that these could be considered at Stage 2 of the legislative process (Legislation, Justice and Constitution Committee 2022a, recommendation 3). The Welsh Government declined to do so for similar reasons to those given in the Social Partnership and Public Procurement example above (Miles 2022).

6.2 Limited scrutiny of statutory instruments

The LJC committee considers all statutory instruments laid before the Senedd. It assesses them in technical terms and with regard to their legal or political importance (see Senedd Standing Orders 21.2 and 21.3). LJC reports on SIs must be published no later than 20 days after an instrument has been laid. Committees in the Senedd and academic researchers have pointed to the problems of scrutinising the policy content of SIs given committees’ workload, which includes both legislative and oversight functions. Stirbu explains that the size of the Senedd means members often need to take on roles in multiple committees. The resulting high workloads along with turnover in committee membership can make it difficult at times for committee members to give as much attention as is needed for detailed scrutiny of legislation (Stirbu 2021: 42-44). In its 2015 report Making Laws in Wales, the Constitutional and Legislative Affairs committee stated, ‘It is particularly difficult for committees with a heavy policy and Bill workload to factor in such scrutiny at short notice following the laying of an instrument and given that there is only a relatively short period in which to report’ (2015, paragraph 331). They suggest this strengthens the argument for expanding the size of the Senedd. They recommend that when committees do have sufficient time to look at delegated legislation, they do so during the government consultation process before an SI is laid. This gives more time and an opportunity to suggest amendments before the content of the SI is fixed. Nevertheless, the problem remains of the limited time and resources needed to monitor government consultations.

Research on delegated legislation in the UK case indicates that scrutiny of SIs is minimal. One problem is that they are unamendable and therefore present legislators with a take-it-or-leave-it decision. There is theoretically a possibility of an SI either being annulled (if subject to the negative procedure) or not being approved (if subject to the affirmative procedure). Defeats are very rare though. Hansard Society data from 2021 in the UK case show only six SIs have been rejected since 1950 and none have been rejected in the Commons since 1979 (Hansard Society 2021). The House of Lords tends not to reject (although does debate) SIs, conscious of its position as an unelected chamber.

While MPs can attempt to trigger a debate on an SI subject to the negative procedure by tabling an Early Day Motion (EDM) to ‘pray’ that the SI be annulled, this happens infrequently. In the 2016-17 session, for instance, only 21 of 537 negative SIs were prayed against (Pywell 2019, p.110). Even when an MP tables an EDM, the government controls whether a debate takes place or not.

When affirmative procedures are used at Westminster, after consideration in a Delegated Legislation Committee (DLC), votes on the floor of the House tend to take place without debate (as noted above). This means that many MPs may know little about the SI on which they are voting (Pywell 2019, p.111). Data from the 2016-17 session indicate the Commons spent only 0.5% of its sitting time on motions to approve SIs and that the average duration of a DLC meeting was 26 minutes (Pywell 2019, p.111). These limitations on scrutiny are compounded by the fact that the Joint Committee on Statutory Instruments does not consider the merits of delegated legislation.

6.3 The use of delegated legislation to make guidance and tertiary legislation

Further problems with the scrutiny of delegated legislation identified by the DPRRC (2021a) in its Democracy Denied report and in the SLSC’s (2021) Government by Diktat report include the use of delegated powers that allow ministers to issue guidance, with no parliamentary procedure for scrutiny of the power. This is not unusual in itself but the DPRRC draw attention to powers allowing ministers to issue mandatory guidance or guidance to which the person to whom it is addressed ‘must have regard’. The committee points out that such guidance is, in effect, legislation that the UK Parliament cannot scrutinise (paragraphs 91-95). In its Government by Diktat report, the SLSC gives an example of how guidance can be used to ‘amplify’ legislation in the form of the Health Protection (Coronavirus, Restrictions) (No. 3) and (All Tiers) (England) (Amendment) Regulations 2021. Here the definition of a ’critical worker’ was left to guidance and yet this was central to deciding who was entitled to benefit from childcare during the pandemic (Secondary Legislation Scrutiny Committee 2021, paragraph 56).

The DPRRC and Constitution Committee reports also refer to the use of delegated powers to create tertiary legislation. This is when ministers are delegated the ability to sub-delegate power to an institution of some kind. The European Union (Withdrawal) Act 2018 allowed for tertiary legislation not subject to any parliamentary procedure. The DPRRC’s report on the Bill said some of the powers within it ‘could, for example, be used to create new bodies with wide powers to legislate in one of the many areas currently governed by EU law, including aviation, banking, investment services, chemicals and medicines’ (Delegated Powers and Regulatory Reform Committee 2017, paragraph 28). Such bodies could make law without being subject to scrutiny in the UK Parliament.

6.4 Using framework legislation and delegated powers for financial and fiscal matters

The problems of poor scrutiny of SIs noted above are particularly significant when it comes to the financial implications of legislation made in this way. The Renewable Heat Initiative scheme in Northern Ireland, which proved to be a very costly policy was established using delegated legislation and ultimately led to the resignation of the Deputy First Minister and the collapse of the Northern Ireland Executive.

In Wales, the change to the general speed limit for restricted roads from 30 to 20 miles per hour was brought about through a statutory instrument.[7] This SI was considered by the LJC Committee and the Climate Change, Environment and Infrastructure Committee but not by the Finance Committee. The financial implications of the change are potentially widespread. The Welsh Government argued it would save costs due to the lower likelihood of accidents, while others argued costs would increase for those transporting goods by road and for the bus industry due to longer journey times (Howorth, 2023). The opportunity for scrutiny of these potential financial implications was considerably more limited than it would have been had the change been made through primary legislation.

The Welsh Tax Acts etc. (Power to Modify) Act was discussed briefly in Section 3 above. Debate on the delegated powers in this piece of legislation was partly concerned with whether tax changes should be possible through delegated legislation. For the purposes of comparison, in the UK case, changes to taxation are made via primary legislation in the form of the Finance Bill although the procedure is different from that used in other areas of law-making. The 1968 Provisional Collection of Taxes Act allows for some tax changes to take effect by 6pm on budget day via a Provisional Collection of Taxes motion, which Parliament must approve. Other resolutions must be passed by Parliament within ten sitting days of the budget to give effect to tax changes. The Finance Bill must then be passed within a fixed time period in order to give permanent effect to the resolutions (Seely 2023, p.8). This procedure is specific to the budget but crucially it does not involve delegated legislation.

For some, taxation powers are so fundamental that they must only be made by primary legislation. Some witnesses giving evidence to the LJC committee on this bill made exactly that point. The Welsh Government’s argument was that it needed the powers to respond quickly to changes in UK tax law to protect Welsh revenues. This was particularly the case for tax avoidance schemes where changes might be needed swiftly to ensure revenues are protected. The LJC committee report (2022b) engages in questions of whether a system similar to that used in the UK Parliament (through the Provisional Collection of Taxes Act 1968 and an annual Finance Act) should be adopted in Wales with views for and against this set out by witnesses. Ministers argued that this would not directly address the reasons why they included delegated powers in the Welsh Tax Acts etc. (Power to Modify) Act. This was on the basis that these powers were designed for rapid changes being made outside of the annual budgetary cycle.

6.5 Unintended consequences of framework legislation and the process of futureproofing

One possible effect of framework bills is that the extensive use of widely drawn delegated powers may mean they are used in ways not anticipated when the powers were drawn up. An example from the UK is the power given to ministers via the Social Security Contributions and Benefits Act 1992 to determine the amount of Housing Benefit that can be received by claimants. This Act allows ministers to set the ‘appropriate maximum housing benefit’. This begins at a particular level from which deductions are made depending on the regulations made by ministers via delegated legislation. In 2012 the Conservative-Liberal Democrat coalition government introduced a change that reduced the amount of benefit paid if claimants, in the social rented sector, had spare bedrooms in their dwellings.[8] The government referred to this as the removal of the spare room subsidy, while opponents labelled it a ‘bedroom tax’ (Tucker 2018, p. 351). It led to tension within the coalition, evidence of which was the introduction by a Liberal Democrat backbench MP of a private members’ bill to alter the policy, although the bill did not become law.

A further example is the laying of a statutory instrument (SI) by the UK government in 2015 to make over £4bn worth of cuts to tax credits.[9] The House of Lords voted to delay consideration of this SI. The House of Lords Constitution Committee argued that this was an example of an SI being used in a very different way from that intended by the authors of the enabling legislation (Tax Credits Act 2002), which was to vary these tax credits rather than to make huge cuts to them (Constitution Committee, 2016, paragraph 41).

Alternatively delegated powers may be included in bills where there is no intention of using them at the time the bill is being passed through the legislative process. Often known as futureproofing, this gives ministers the option to introduce policy changes in future without having to make primary legislation. The LJC has drawn attention to this practice in its reports on Senedd bills. A recent example is the inclusion of a power in the Environment (Air Quality and Soundscapes) (Wales) Bill 2023 expanding the circumstances under which Welsh Ministers can introduce trunk road charging. Under the new law, tolls could be introduced for the purpose of reducing air pollution. The Welsh Government’s explanatory memorandum for the bill indicated there were no plans to introduce such road charges. The then Minister for Climate Change, Julie James defended the policy to the Climate Change, Environment and Infrastructure Committee (CCEIC) on 29 March 2023 saying:

‘I know it does seem counterintuitive that we’re taking powers to do something we’re not planning to do. But it’s because we want to futureproof this Bill; we don’t want to have to come back with more primary legislation in circumstances where, actually, we haven’t been successful in reducing airborne pollutants as a result of vehicle emissions by the other measures that we’re taking forward’

In its report on the bill the LJC (2022b) said ‘As a matter of principle, we do not believe it is appropriate for the Welsh Ministers to take powers “just in case” they may be needed in the future, particularly when the policy may be more appropriate for primary legislation’. The CCEIC’s report (2023) added to this point by referring to unintended consequences that might follow from introducing road charging, such as higher levels of traffic on residential roads. On this basis, the committees argued that it would be better to introduce these changes through primary legislation which provided an appropriate opportunity for scrutiny.

UK governments use futureproofing as a motivation for the inclusion of delegated powers in some bills. This term can be found in delegated powers memoranda that the government must produce at the point of introducing a bill. In response to this, the House of Lords’ Constitution Committee’s 2018 report on The Legislative Process: The Delegation of Powers, referred to the problems of scrutinising skeleton bills and argued that the Government ‘cannot rely on generalised assertions of the need for flexibility or futureproofing’ as justification for framework legislation (Constitution Committee 2018, paragraph 58). These similar responses from committees in the UK Parliament and Senedd indicate the problematic nature of futureproofing in the eyes of some parliamentarians.

7. Examples of best practice in scrutinising framework legislation

In the specific case of framework bills, the Cabinet Office Guide to Making Legislation(paragraph 15.19) advises ministers they will need to justify the use of this type of legislation in full to the DPRRC. This means explaining what has been done to include policy detail in the bill, explaining limitations to the powers included and pointing to appropriate precedents for the use of delegated powers.

One of the things that governments can do to improve a parliament’s ability to scrutinise framework bills is to provide drafts of the regulations that ministers plan to make with the powers delegated to them in a bill. The Cabinet Office Guide advises that this should be done at committee stage in Westminster when MPs and peers consider the detail of a bill (paragraph 15.35). If draft regulations cannot be produced at that stage, UK government departments are asked to consider providing policy statements that indicate how delegated powers would be used (paragraph 15.36).

The DPRRC (2021b) goes further in its guidance for Departments on skeleton bills. This asks governments to use skeleton bills only in exceptional circumstances and to indicate their use with a “skeleton legislation declaration” in the delegated powers memorandum produced with each government bill. This memorandum must list all the delegated powers in a bill, indicate the proposed parliamentary procedure and justify the need for the delegation as well as the choice of procedure.

Examples of good practice from Westminster include those where the skeletal nature of bills is reduced by the provision of information about how delegated powers will be used. For example, in its report on the Health and Social Care Act 2008 the DPRRC pointed to a new health-in-pregnancy grant that was to be made available. All aspects of the rules concerning entitlement to and the amount of the new grant were to be set out in regulations. These therefore appeared to be skeleton clauses but, in line with the Cabinet Office Guide to Making Legislation, the Government provided draft regulations. The committee responded that, on this basis, they did not consider the power to be ‘inappropriate’ (Delegated Powers and Regulatory Reform Committee 2008a, paragraph 17).

In its report on the Infrastructure Bill in the 2014-15 session, the DPRRC (2014a) drew attention to part of the Bill that is ‘entirely enabling in character’. It grants powers concerned with allowing individuals and groups to purchase a stake in local renewable electricity generation facility. This was acceptable to the DPRRC because Schedule 5 of the Bill provides much detail about what provisions can be made using these powers and therefore limits how the power can be used. Again, this is consistent with what the Cabinet Office Guide asks legislative drafters to do.

Based on discussions in the House of Lords’ second reading debate of the Criminal Defence Service Bill in the 2005-6 session, the DPRRC (2015) considered whether this was a skeleton bill. The Bill deals with the provision of legal aid. It amended the Access to Justice Act 1999 which had set up the Criminal Defence Service. The committee concluded that when read with this 1999 Act, there was sufficient clarity about the policy that the delegated powers would implement, that they did not deem this to be a skeleton bill. The lesson from these examples is that bills that delegate considerable numbers of powers to ministers may be acceptable if enough detail is provided about the limits to and the policies to be implemented by those powers. Such information reduces the scrutiny deficit in that legislators are more aware when assessing a bill, of the policies that could follow from it.

Academic literature on delegated powers offers some pointers as to good practice from outside the UK. Fleming and Ghazi (2023) look at how delegated legislation is scrutinised in Australia, Canada, India, Ireland, New Zealand and South Africa. Although they find that merits-based scrutiny is weak in all six of these cases, they find strengths in other aspects. In the Australian case, backbenchers are in a better position than those in the UK’s legislatures in forcing debate on disallowing a piece of delegated legislation. In the Australian Senate, such motions take precedence over other business. Australia also has a system whereby motions to disallow an SI will automatically take effect if not dealt within fifteen sitting days. This means that, in the House of Representatives (the lower house in the Australian Parliament) motions of that type are normally debated in government time in order that the government does not lose the SI. Canada and New Zealand have similar procedures although these are limited in that they are not open to all members to move. For instance, in Canada, such motions can only be moved if based on a report from the committee that looks at the legality of secondary legislation. These procedures might allow for more attention to be drawn to SIs than at present in UK legislatures and might generate change through government anticipating reactions to particular SIs.

8. Conclusion

Framework or skeleton legislation is used in some cases to outline a policy to be introduced and for stakeholders to be consulted before policy detail is worked out. Framework bills enable governments flexibility to deal with future scenarios, the details of which are not yet known. This was particularly the case in the run up to the UK leaving the EU

Framework legislation, however, brings problems for parliamentary scrutiny. Statutory Instruments that follow from framework legislation cannot be amended by the Senedd. Furthermore,  opportunities for scrutinising the policy detail implemented through delegated legislation are far more limited than those for scrutinising primary legislation. This may mean that significant policy changes are introduced without much scrutiny. As noted in earlier, such changes might have widespread financial implications or could involve making changes to taxation. As pointed out in 2015 by the Constitutional and Legislative Affairs committee and by Stirbu (2021), the limited size of committees in the Senedd, combined with their heavy workload and their dual legislative and policy functions means that finding time to scrutinise SIs within the time allowed can be difficult. This problem is not limited to the Senedd. The UK Parliament’s procedures mean that very little time is spent in the chamber of the House of Commons scrutinising SIs. Even Commons committees dedicated to discussing delegated legislation may deal with it rapidly.

Granting powers to ministers that they have no intention of using in the foreseeable future – often described as futureproofing – can add to the problems of framework bills. This generates uncertainty by creating possibilities for powers to be used in unintended ways by future governments, who may have different intentions from those introducing the delegated powers.

These drawbacks are significant given the prevalence of framework legislation. 43% of the bills introduced in the sixth Senedd by March 2024 were framework bills. Although it is difficult to be precise, the equivalent figure was 10% for the sixth Scottish Parliament and roughly 9% for the 2021-22 and 2022-23 sessions combined in the UK Parliament.

Some of these problems might be reduced by following the advice contained in the Legislation Handbook on Assembly Bills or the Cabinet Office’s Guide to Making Legislation about the use of delegated powers and framework bills (covered in section 4 above). Nevertheless, the Senedd has repeatedly pointed to the problems of framework bills where policy detail cannot be scrutinised through the primary legislative process and where – even if government guidance is followed – delegated powers are open to be used by future governments without sufficient scrutiny and in ways not envisaged at the time enabling bills are passed.

Some of the procedures used outside of the Senedd and UK Parliament for scrutinising statutory instruments may help to reduce the scrutiny deficit caused by framework bills. One example is the requirement for ministers to move a motion in the relevant policy-based committee in the Scottish Parliament for statutory instruments considered under the affirmative procedure. Another, in the NIA, is the SL1 Letter procedure, meaning amendments can be suggested before secondary legislation formally begins its passage through the legislature. The system, in Australia, for giving priority to votes disallowing a statutory instrument may help to improve scrutiny in a procedure that is otherwise balanced in favour of the executive. Nevertheless, there would be less need for these innovations if the use of framework legislation was reduced.

Appendix: Procedures for scrutiny of secondary legislation in the UK’s legislatures

Scrutiny of statutory instruments in the Senedd

When proposing a statutory instrument, the Welsh Government must publish an explanatory memorandum, a Regulatory Impact Assessment and a report from the Auditor General for Wales if needed, with views given the appropriateness of any expenditure that follows from the instrument.

There are several procedures for parliamentary scrutiny of statutory instruments in the Senedd. Senedd Acts that delegate powers will state the procedure assigned to each. This choice, along with the instrument itself, will be considered by the Legislation, Justice and Constitution committee. Some SIs have no parliamentary procedure assigned to them and so are not considered by the Senedd. This is often the case for commencement orders which bring into force all or part of an Act.

Affirmative procedure

This procedure, in essence, means that an SI must be approved by the Senedd before it can come into force. In this case a minister must table a motion that the SI be approved. Before this can be done, the committee considering the instrument must have reported or 20 days need to have passed since the motion was tabled. SIs normally go to LJC committee which considers them on the basis of legality and whether their aims are best achieved via secondary rather than primary legislation. Any SIs involving taxes are also looked at by Finance Committee. Other committees can choose to report on an instrument although they must give the Welsh Government notice of this within seven days of the SI being laid. The SI comes into force only if approved by the Senedd.

Negative procedure

This is the most frequently used procedure. In this case, the Senedd has the power to annul an SI within a particular period after it is laid. First, the minister makes the policy by signing the instrument. Once it is made, the relevant minister will lay the SI before the Senedd which then has up to 21 days to annul it. The LJC committee cites breaches of this rule in its annual reports, in other words, it points out cases where an SI comes into force fewer than 21 days after it was laid. If no motion to annul is tabled, or if a motion is tabled but defeated, the SI remains law. As with affirmative-procedure SIs, the LJC committee reports on these instruments.

Watkin and Greenberg (2018, pp.203-204) point out that if there is a vote in the chamber to annul an SI and the vote is a tie, the Presiding Officer must vote against the motion thereby allowing the instrument to remain valid or, in other words, for the status quo to persist. Significantly, this is the opposite of what happens with primary legislation when, in a tied vote, maintaining the status quo means voting against the legislation. Governments anticipating tied votes (for instance, if they were to have 50% or close to 50% of the Senedd seats) may therefore have an incentive to use secondary legislation under the negative rather than affirmative procedure.

Super-affirmative procedure

Under this procedure, a draft of the SI must be available some time before it is laid. It is possible for MSs to suggest amendments during this period but there are no requirements for ministers to take them on board. In some cases, ministers may be required to report on how they respond to the amendments when they lay the instrument in its final form.

Enhanced-negative procedure

SIs subject to this approach must be published in draft before being laid by the minister. Amendments can be suggested but, as with super-affirmative SIs, there is no requirement that ministers act on these except that they might be required to explain any changes made when laying the instrument.

Scrutiny of statutory instruments in the UK Parliament

In the UK Parliament, if an SI has been assigned a procedure for parliamentary scrutiny, it will normally be considered by the Joint Committee on Statutory Instruments (JCSI). This includes members of the Commons and the Lords although much of its detailed work is done by parliamentary lawyers (Page 2001, p.159). The JCSI deals with the legality of SIs and not with their policy content. Among other things, the committee looks for drafting errors and assesses whether the content of the SI is within that permitted by the enabling act from which it originates.

The House of Lords’ Secondary Legislation Scrutiny Committee (SLSC) considers the policy content of SIs. It decides whether to draw the House of Lords’ attention to particular SIs it views as legally or politically important. Its decisions are not binding.

As in the Welsh case, some SIs have no parliamentary procedure assigned to them. These tend to be uncontentious. For those that do have a procedure, the options are similar to those in the Senedd. More detail can be found in the Commons Library Briefing on SIs (Kelly, 2016).

Affirmative procedure

Under this procedure, SIs require approval by Parliament to become law. SIs considered through the affirmative procedure will be assigned to a Delegated Legislation Committee in the Commons. The committee can consider the instrument but cannot make any amendments to it. Following this, a vote to approve the SI will take place on the floor of the Commons without debate. The House of Lords will not debate an SI under this procedure until the Joint Committee on Statutory Instruments and the Secondary Legislation Scrutiny Committee have considered the instrument. A vote on the SI takes place in the chamber of the House of Lords.

Made-affirmative SIs come into force before being seen by Parliament. These SIs require approval within a fixed time period in order to remain in force. For example, 77 Covid-related SIs were made using this procedure during the pandemic (Judge 2021, p.289). These allow for legislation to be made quickly but such law is only scrutinised after it has come into effect.

Negative procedure

SIs considered under this procedure become law but can be annulled if either the Commons or Lords passes a motion to that effect within a certain time period, normally 40 days. In the Commons, any member can put down a motion to annul an SI but the motion is not likely to be debated unless it attracts many signatories. It is up to the government to decide whether to give time for a debate.

Super-affirmative procedure

In this case, Parliament, through its committees, can consider a proposal for an SI before it is laid. Amendments can be suggested but do not have to be acted upon. After this, the SI will be laid by the minister under the affirmative procedure.

Scrutiny of statutory instruments in the Scottish Parliament

SIs laid before the Scottish Parliament (Scottish Statutory Instruments (SSIs)) to which a parliamentary procedure has been assigned are considered by the Delegated Powers and Law Reform Committee (DPLRC). This committee, like the JCSI in the UK Parliament and the LJC Committee in the Senedd, carries out a legalistic assessment of SSIs. The DPLRC then reports to the parliamentary committee responsible for considering the substance of a particular SSI (known as the lead committee). This committee assesses the merits of the SSI in policy terms. The details of scrutiny vary depending on the procedure assigned. The Scottish Parliament’s Guide to SSIs offers more detail on this.

Affirmative procedure

In this procedure, the relevant Scottish Minister must propose a motion to the lead committee (which considers the substance of the SSI) that the committee recommends the SSI be approved. The relevant Minister will attend a meeting of the lead committee and give evidence, after which the committee will vote on the motion. The lead committee then reports to the Scottish Parliament on its decision within 40 days of the SSI being laid. If the committee recommends that the SSI be approved, Parliament then votes on it. If the committee recommends otherwise, the Parliamentary Bureau, which includes the Presiding Officer and representatives from political parties represented in Parliament, decides whether MSPs should vote on it in chamber. The SSI must be approved by Parliament to become law.

Negative procedure

Under the negative procedure an SSI will be considered by DPLRC and the lead committee. Within 40 days of the SSI being laid, any member can put down a motion before the lead committee to annul the instrument. That committee then votes on whether to recommend that the SSI is annulled. If there is a committee vote in favour of annulling the SSI, Parliament will then vote on a motion to annul. The SSI ceases to have legal effect if Parliament votes to annul it.

Provisional affirmative procedure

This is normally only used in emergency situations. As with made-affirmative SIs in the UK Parliament, SSIs under this procedure come into effect before the Scottish Parliament has seen them. They must be approved within a fixed time period in order to remain in force.

Super-affirmative procedure

Super-affirmative SSIs are made available in draft form and are normally referred to the DPLRC and the lead committee, which can take evidence from those affected by the SSI. There are normally 60 or 90 days to consider the SSI in draft. The DPLRC and lead committee then report to the Scottish Government. After this point the affirmative procedure applies.

Scrutiny of statutory instruments in the Northern Ireland Assembly

The distinctive elements of how statutory instruments (known as statutory rules (SRs) are scrutinised in the Northern Ireland Assembly (NIA) are set out in the main text above (section 5.2.2). Beyond this, procedures used for considering SRs have the same principles and names as those in the other legislatures of the UK except that the super-affirmative procedure is known as the draft-affirmative procedure and the made-affirmative procedure is entitled ‘confirmatory procedure’.

The Chairpersons’ Liaison Group, which brings together committee chairs in the NIA, carried out an inquiry into strengthening committee scrutiny during the 2017-2022 term. This occurred in light of the controversy surrounding the Renewable Heat Incentive scheme,[10] which was introduced through a Statutory Rule. Among their recommendations on the scrutiny of SRs, the group suggested committees might have power to amend SRs although the detail of the process for this was not set out. They also suggest more time should be given to committees if they deem that consultation on an SR is needed with those likely to be affected by it.

References

Blythyn, Hannah (2022) Letter to Huw Irranca-Davies, 15 December 2022, online at: https://business.senedd.wales/documents/s132332/LJC6-01-23%20-%20Paper%2039%20-%20Letter%20from%20the%20Deputy%20Minister%20for%20Social%20Partnership%2015%20December%202022.pdf

Cabinet Office (2022) Guide to Making Legislation, online at: https://www.gov.uk/government/publications/guide-to-making-legislation.

Carey, John and Matthew Shugart (2009) Calling out the tanks or filling out the forms? In John M. Carey and Matthew Soberg Shugart (2009) (eds) Executive Decree Authority, Cambridge, Cambridge University Press.

Climate Change, Environment and Infrastructure Committee (2023) Environment (Air Quality and Soundscapes) (Wales) Bill: Stage 1 Report, July 2023, online at: https://senedd.wales/media/cxuaxybj/cr-ld15956-e.pdf.

Constitution Committee (2016) Delegated Legislation and Parliament: A response to the Strathclyde Review, 9^th report of session 2015-16, HL Paper 116, 23 March.

Constitution Committee (2018) The Legislative Process: The Delegation of Powers, HL Paper 225, 16^th report of session 2017-19, 20 November.

Constitutional Affairs Committee (2011) Inquiry into the Drafting of Welsh Government Measures: Lessons from the first three years, online at: https://senedd.wales/media/dpvdbr4l/cr-ld8393-e-english.pdf.

Constitutional and Legislative Affairs Committee (2014) Report on the Higher Education (Wales) Bill, October 2014, online at: https://senedd.wales/media/43ick0ka/cr-ld9938-e.pdf.

Constitutional and Legislative Affairs Committee (2015) Making Laws in Wales, online at: https://senedd.wales/laid%20documents/cr-ld10379/cr-ld10379-e.pdf.

Constitutional and Legislative Affairs Committee (2017) Report on the Landfill Disposals Tax (Wales) Bill, online at: https://senedd.wales/laid%20documents/cr-ld10973/cr-ld10973-e.pdf.

Delegated Powers and Law Reform Committee (2016) Legacy Report, 29^th report, 2016, online at: https://webarchive.nrscotland.gov.uk/3/archive2021.parliament.scot/S4_SubordinateLegislationCommittee/Reports/DPLRS042016R29.pdf.

Delegated Powers and Law Reform Committee (2017a) Legislative Consent Memorandum on the European Union (Withdrawal) Bill, SP Paper 231, 54^th Report, 2017, online at: https://bprcdn.parliament.scot/published/DPLR/2017/11/17/Legislative-Consent-Memorandum-on-the-European-Union--Withdrawal--Bill/DPLRS52017R54.pdf.

Delegated Powers and Law Reform Committee (2017b) Letter to John McCormick, Chair of the Commission on Parliamentary Reform, 9 February 2017, online at: https://webarchive.nrscotland.gov.uk/3/archive2021.parliament.scot/S5_Delegated_Powers/CoPR2.pdf.

Delegated Powers and Law Reform Committee (2022a) Inquiry into the use of the made affirmative procedure during the coronavirus pandemic, SP Paper 110, 12^th report, 2022, online at: https://digitalpublications.parliament.scot/Committees/Report/DPLR/2022/2/10/564c1832-ed18-49e5-83e4-3ea890131e07#24902f72-2864-4458-b2ca-4bfda80746e4.dita.

Delegated Powers and Law Reform Committee (2022b) Coronavirus (Recovery and Reform) (Scotland) Bill: Stage 1, SP Paper 147, 22^nd report, 2022, online at: https://bprcdn.parliament.scot/published/DPLR/2022/3/29/5c98006e-d8d9-4ec6-a0ee-9d5b2cad5aac/DPLRS062022R22.pdf.

Delegated Powers and Law Reform Committee (2023) Delegated Powers in the Circular Economy (Scotland) Bill at Stage 1, SP Paper 461, 64^th report, 2023, online at: https://bprcdn.parliament.scot/published/DPLR/2023/11/2/ab81cc16-48bd-4ebe-b2f9-421f0feba827/DPLRS062023R64.pdf.

Delegated Powers and Law Reform Committee (2024) Delegated Powers in the Agriculture and Rural Communities (Scotland) Bill at Stage 1, SP Paper 515, 6^th report, 2024, online at: https://digitalpublications.parliament.scot/Committees/Report/DPLR/2024/1/26/da95779b-8433-412c-b109-795bef958bfa#Introduction.

Delegated Powers and Regulatory Reform Committee (1999) Special Report for Session 1998-99: The Committee’s Work, 29^th report of session 1998-99, 27 October 1999.

Delegated Powers and Regulatory Reform Committee (2001), First Report of Session 2001-02, 4 July 2001.

Delegated Powers and Regulatory Reform Committee (2005) First Report of Session 2005-06, 15 June.

Delegated Powers and Regulatory Reform Committee (2008a) Sixth Report of Session 2007-08, HL Paper 76, 13 March.

Delegated Powers and Regulatory Reform Committee (2008b) Thirteen Report of Session 2007-08, 15 October 2008.

Delegated Powers and Regulatory Reform Committee (2008c) Fifteenth Report of Session 2007-08, 12 November 2008.

Delegated Powers and Regulatory Reform Committee (2014a) Second Report of Session 2014-15, HL Paper 15, 27 June.

Delegated Powers and Regulatory Reform Committee (2014b) Twelfth Report of Session 2014-15, 17 December 2014.

Delegated Powers and Regulatory Reform Committee (2017) Third Report of Session 2017-19, HL Paper 22, 28 September 2017.

Delegated Powers and Regulatory Reform Committee (2021a) Democracy Denied? The urgent need to rebalance power between Parliament and the Executive, HL Paper 106, 19 November, online at https://publications.parliament.uk/pa/ld5802/ldselect/lddelreg/106/106.pdf

Delegated Powers and Regulatory Reform Committee (2021b) Guidance for Departments on the role and requirements of the Committee, November 2021.

Economy, Trade and Rural Affairs Committee (2023) Agriculture (Wales) Bill: Committee Stage 1 Report, January 2023, online at: https://senedd.wales/media/d4befhvr/cr-ld15627-e.pdf.

England, Matthew (2024) Creeping ministerial powers: the example of the Tobacco and Vapes Bill, Hansard Society Blog, 15 April, online at: https://www.hansardsociety.org.uk/blog/tobacco-and-vapes-bill.

Epstein, David and Sharyn O’Halloran (1999) Delegating Powers: A Transaction Cost Politics Approach to Policy Making under Separate Powers, Cambridge: Cambridge University Press.

Fleming, Thomas G. and Tasneem Ghazi (2023) Parliamentary Scrutiny of Delegated Legislation: Lessons from Comparative Experience, The Political Quarterly, 94(3): 412-419.

Fox, Ruth and Joel Blackwell, (2014) The Devil is in the Detail: Parliament and Delegated Legislation, London: Hansard Society.

Greenberg, Daniel (2011) Laying Down the Law: A discussion of the people processes and problems that shape Acts of Parliament, London: Sweet and Maxwell.

Hansard Society (2021) Delegated legislation: the problems with the process, online at: https://www.hansardsociety.org.uk/publications/reports/delegated-legislation-the-problems-with-the-process.

Hansard Society (2023) Proposals for a New System for Delegated Legislation, online at: https://www.hansardsociety.org.uk/publications/reports/proposals-for-a-new-system-for-delegated-legislation-a-working-paper

Health, Social Care and Sport Committee (2022) Summary of responses to the committee’s call for views, online at: https://www.parliament.scot/-/media/files/committees/health-social-care-and-sport-committee/ncs-call-for-views-summary.pdf.

Hodgson, Taylor and Blencathra (2020) Letter from committee chairs to Michael Gove MP and Jacob Rees-Mogg MP on skeletal bills and skeletal provision, online at: https://committees.parliament.uk/publications/2750/documents/27198/default/

Howorth, Francesca, (2023) Gearing up for 20mph speed limits, Senedd Research, online at: https://research.senedd.wales/research-articles/gearing-up-for-20mph-speed-limits/.

Huber John D. and Charles R. Shipan (2002) Deliberate Discretion: The Institutional Foundations of Bureaucratic Autonomy, Cambridge: Cambridge University Press.

Judge, David (2021) ‘Walking the Dark Side: Evading Parliamentary Scrutiny’, The Political Quarterly, 92(2): 283-292.

Kelly, Richard (2016) Statutory Instruments, House of Commons Library Briefing Paper, No. 06509, online at: https://researchbriefings.files.parliament.uk/documents/SN06509/SN06509.pdf.   

Kosti, Nir (2023) Conceptualization and measurement of regulatory discretion: Text analysis of 120 years of British legislation, Regulation and Governance, Early view.

Legislation, Justice and Constitution Committee (2022a) Report on the Tertiary Education and Research (Wales) Bill,March 2022, online at: https://senedd.wales/media/eazdi3dz/cr-ld14992-e.pdf.

Legislation, Justice and Constitutional Affairs Committee (2022b) Report on the Welsh Tax Acts etc. (Power to Modify) Bill, April 2022, online at: https://senedd.wales/media/t02ctq5w/cr-ld15085-e.pdf.

Legislation, Justice and Constitution Committee (2022c) Report on the Social Partnership and Public Procurement (Wales) Bill, November 2022, online at: https://senedd.wales/media/zp1g3y5r/cr-ld15466-e.pdf.

Legislation, Justice and Constitution Committee (2022d) Annual Report 2021/22, October 2022, online at: https://senedd.wales/media/gjfnzsff/cr-ld15390-e.pdf.

Legislation, Justice and Constitution Committee (2023a) Report on the Agriculture (Wales) Bill, January 2023, online at: https://senedd.wales/media/miodmivf/cr-ld15632-e.pdf.

Legislation, Justice and Constitution Committee (2023b) Report on the Environment (Air Quality and Soundscapes) (Wales) Bill, July 2023, online at: https://senedd.wales/media/sz4fekki/cr-ld15959-e.pdf.

Legislation, Justice and Constitution Committee (2023c) Annual Report 2022/23, November 2023, online at: https://senedd.wales/media/thmdd2p2/cr-ld16186-e.pdf.

Legislation, Justice and Constitution Committee (2023d) Report on the Infrastructure (Wales) Bill, November 2003, online at: https://senedd.wales/media/rvxh0c0o/cr-ld16163-e.pdf.

Miles, Jeremy (2022) Letter to Huw Irranca- Davies, 29 March 2022 online at: https://business.senedd.wales/documents/s124474/Letter%20from%20the%20Minister%20for%20Education%20and%20Welsh%20Language%20to%20the%20Chair%20of%20the%20Legislation%20Justice%20a.pdf.

Page, Edward C., (2001) Governing by Numbers: Delegated Legislation and Everyday Policy-Making, Oxford: Hart Publishing.

Pywell, Stephanie (2019) Something old, something new: busting some myths about Statutory Instruments and Brexit, Public Law, January, pp.102-120.

Rural Affairs and Islands Committee (2024) Stage 1 report on the Agriculture and Rural Communities (Scotland) Bill, SP Paper 553, 5^th report, 2024, online at: https://bprcdn.parliament.scot/published/RAI/2024/3/18/5c333ef7-082b-4555-bf65-ce2006c5fcd4-1/RAI-S6-24-05.pdf.

Scottish Government (2023) Agriculture and Rural Communities (Scotland) Bill – Policy Memorandum, online at: https://www.parliament.scot/-/media/files/legislation/bills/s6-bills/agriculture-and-rural-communities-scotland-bill/introduction/policy-memorandum-accessible.pdf.

Secondary Legislation Scrutiny Committee (2021) Government by Diktat: A call to return power to Parliament, HL Paper 105, 20^th report of session 2021-22.

Seely, Antony (2023) The budget and the annual Finance Bill, House of Commons Library Briefing, 813, online at: https://commonslibrary.parliament.uk/research-briefings/sn00813/.

Stirbu, Diana (2021) Power, Influence and Impact of Senedd Committees: developing a framework for measuring committees’ effectiveness, Cardiff: Welsh Parliament, online at: https://senedd.wales/media/xtqk0ojr/gen-ld14672-e.pdf.

Tucker, Adam (2018) Parliamentary Scrutiny of Delegated Legislation, In Alexander Horne and Gavin Drewry (eds) Parliament and the Law, 2^nd edition, London: Bloomsbury.

Watkin, Thomas Glyn and Daniel Greenberg (2018) Legislating for Wales, Cardiff: University of Wales Press.

Welsh Government (2019) Legislation Handbook on Assembly Bills, May 2019, online at: https://www.gov.wales/sites/default/files/publications/2019-05/legislation-handbook-on-assembly-bills.pdf.

Williams, Matthew (2018) How Language Works in Politics: The Impact of Vague Legislation on Policy, Bristol: Bristol University Press.

---

[1] I am grateful to Hedydd Phylip and to officials in the Senedd, Northern Ireland Assembly, Scottish and UK Parliaments for much helpful advice on the topics discussed here. Any errors remain my own responsibility.

[2] Data on numbers of government bills are published by session, hence the use of sessions as a time unit for these calculations.

[3] These data are calculated by session as Parliament’s website provides detail of all government bills introduced on a sessional basis.

[4] The Conveners Group brings together conveners of Scottish Parliament committees.

[5] This statement appears at the start of LJC committee reports on bills. See for example, paragraph 10 of the LJC’s report on the Local Government Finance (Wales) Bill, March 2024.

[6] For more on futureproofing, see section 6.5.

[7] The Restricted Roads (20 mph Speed Limit) (Wales) Order 2022, enabled by the Road Traffic Regulation Act 1984.

[8] This change was made through the Housing Benefit (Amendment) Regulations 2012.

[9] This SI was the Tax Credits (Income Thresholds and Determination of Rates) (Amendment) Regulations 2015. The House of Lords’ vote to delay consideration of this SI led the UK Government to launch the Strathclyde Review into Secondary Legislation and the Primacy of the House of Commons.

[10] The report of the Independent Public Inquiry into the scheme can be found here: https://cain.ulster.ac.uk/issues/politics/docs/rhi/2020-03-13_RHI-Inquiry_Report-V1.pdf.