Regulatory Impact Statement: Increasing the Visibility of Legislative Quality Issues

Agency Disclosure Statement#

This Regulatory Impact Statement (RIS) has been prepared by, and is the responsibility of, the Treasury. It analyses a legislative option, and a range of supporting non-legislative measures, intended to encourage better quality legislation.

Limits on the Options Analysed#

The core analysis in this RIS is limited to measures based on those recommended in an earlier RIS prepared by the Treasury, entitled "Regulating for Better Legislation - What is the Potential of a Regulatory Responsibility Act?". The analysis is therefore founded on, but does not revisit all of, the matters covered by the earlier RIS.

This earlier RIS, dated 2 February 2011, considered 5 different regulatory options for encouraging better quality legislation, and explained why "Option 5" was Treasury's preferred legislative option. Subsequently, the following commitment was included in the 2011 National-ACT Confidence and Supply Agreement:

"The Regulatory Standards Bill will be included in the continuance motion for the new Parliament, and the Minister for Regulatory Reform will work closely with the Minister of Finance to achieve a mutually agreed outcome, based on Treasury's preferred option (option 5)."

The range of legislation covered by the analysis is limited to legislation proposed or made by the Executive (the government and its agencies). Extending the coverage to include other legislation, such as Local, Private or Members Bills, is not considered, as the responsibility for their preparation falls on the sponsoring member, and Parliament sets expectations for its members through internal rules, such as Standing Orders.

Limitations of the Analysis Undertaken#

The nature and rigour of the analysis of options is limited by the need to rely on subjective judgement about the nature and size of the potential behavioural impacts.

That reliance arises because:

• there is limited international experience and analysis of similar legislative requirements in similar institutional settings for us to draw upon; and
• we have no reliable way to predict:
  • the nature and extent of the Parliamentary and public use of the proposed disclosures for Bills and delegated legislation; or
  • the dynamic effects of requirements and expectations that seek to encourage, but cannot force, changes in the attitudes, behaviour and capabilities of politicians and public officials.

The attempted quantification of costs is limited to the expected costs of direct compliance with any obligations proposed. It does not include estimates of further costs that departments may voluntarily decide to incur to improve the effectiveness of their underlying QA processes, or that external parties may decide to incur to make use of the additional information published.

• Cost estimates for the production of disclosure statements are ballpark figures that draw on existing information about volumes of new legislation, and information about completion times provided by a small number of agencies that produced test disclosures for some pieces of recent legislation.
• Cost estimates for other support arrangements in Packages 1 and 2 are judgements informed by brief discussions with the agencies or people involved.
• Cost estimates for elements of Package 3 are highly speculative, as it identifies areas for development, with further work needed to define specific initiatives.

Consistency with Matters in the Government Statement on Regulation#

None of the measures considered in this RIS are likely to have a direct impact on business costs, existing property rights or market competition.

Note that, in principle at least, almost any of the suggested measures could be progressed without passing legislation. The Government is not currently prevented by law from doing any of these things, and no legal sanctions are suggested for any failure to comply with the proposed obligations. The case for a legislative response therefore rests on our judgement that:

• placing the proposed obligations in legislation provides a more credible, enduring commitment to comply that will underpin desirable behavioural change; and
• this advantage is sufficient to exceed the potential disadvantages (the risk of locking in requirements that are not as useful as hoped, and the risk of unexpected judicial involvement).

Interactions with other Legislative Initiatives#

The analysis assumes that the chosen package would effectively replace the existing Regulatory Standards Bill, presently being considered by the Commerce Committee.

Some of the legislative proposals here would make use of provisions and terms set out in the new Legislation Act 2012, such as the definition of legislative instruments and disallowable instruments, and the statutory role of the Parliamentary Counsel Office.

Further Policy Work Required#

If elements of Package 3 are progressed, further policy work will be required to identify the specific actions to give them best effect. Ideally, that further work would be integrated into the wider Better Public Services reform agenda for the state sector.

Jonathan Ayto
Principal Advisor,
Regulatory Quality Team,
The Treasury
29 January 2013

Executive Summary#

• This Regulatory Impact Statement (RIS) analyses a legislative option and various supporting or complementary measures intended to encourage better quality legislation. The legislative option is limited to one recommended in an earlier Treasury RIS on the potential of a Regulatory Responsibility Act.
• The legislative measure considered is an obligation to publish a disclosure statement for government Bills, substantive government SOPs, and delegated legislation drafted by PCO, at the time that legislation is introduced or published.
• The disclosure statement would be prepared by the administering department. It would flag certain significant powers or unusual features the legislation contains, and provide information about the nature of the quality assurance work undertaken in the process of developing the legislation.
• This disclosure is expected to enhance Parliamentary and public scrutiny of the legislation, by highlighting matters that may be linked to legislative quality. It is also expected to increase departmental attention to quality assurance standards.
• In order to limit risks, it is proposed to legislate only for a basic set of disclosures that we are confident will be workable for almost all Bills and covered delegated legislation. Some extended disclosures would be trialled administratively.
• Incomplete or wrong disclosures will not give rise to legal remedies or sanctions. The incentives for compliance will come from public service respect for the law and reputational effects, supported by clear expectations and timely reminders.
• The proposed content of the disclosure statement has been informed by consultation and testing undertaken by Treasury with departmental and targeted external contacts, supported by a discussion document that incorporated a set of indicative legislative provisions drafted for Treasury by PCO.
• The expected compliance costs of the disclosure statement requirement are low. The support measures give rise to more costs but, in many cases, these could be managed within departmental baselines. Having departments absorb these costs is likely, however, to affect the timeliness and levels of expected compliance.
• Three illustrative packages of measures (plus the status quo option) are considered, which vary in their level of ambition and resource cost. The legislative element is the same for each package, and so the differences between packages relate to the scope of the non-legislative measures proposed.
• While the measures in each package have a common theme and are intended to be mutually reinforcing, it is quite possible to pick and mix individual measures. The key measures in each package are outlined in the summary table below.
• No preferred package has been explicitly identified, but we doubt that the status quo is a durable long-term option. Choices about the range of measures to adopt come down to a judgement about appetite for change, though the most ambitious package would take a number of years to implement.
• Although the outcomes are uncertain, we believe there is enough potential flexibility in how the measures in Packages 1 to 3 can be implemented to ensure costs do not exceed the realised benefits whatever measures are chosen.

Summary Table#

Context:  The Regulatory Standards Bill#

1. The Regulatory Standards Bill was introduced as a Government Bill in March 2011. Its stated purpose is to improve the quality of legislation by:

• specifying principles of responsible regulation that are to apply to new legislation and, over time, to all legislation;
• requiring those proposing new legislation to state whether it is compatible with those principles and, if not, the reasons for the incompatibility; and
• granting courts the power to declare existing legislation to be incompatible with those principles.

2. The content of the Bill reflects the recommendations of the Regulatory Responsibility Taskforce, a group of external experts appointed by the Government to assess and propose changes to an earlier private members Bill.

3. The Taskforce's report and proposed Bill created sharply divided opinion even before the Bill itself was introduced. As a generalisation, it has attracted support from businesses and business organisations, and criticism from legislative experts, academic commentators, and departments. The Commerce Committee, to which the Regulatory Standards Bill was referred, then produced an interim report in September 2011 to disclose advice received from the Regulations Review Committee, which identifies a range of concerns with the Bill.

4. Following the 2011 election, the National and ACT parties concluded a Confidence and Supply Agreement that included the following commitment:

"The Regulatory Standards Bill will be included in the continuance motion for the new Parliament, and the Minister for Regulatory Reform will work closely with the Minister of Finance to achieve a mutually agreed outcome, based on Treasury's preferred option (option 5)."

5. All these references can be accessed online from links in the Endnotes on p.20.

Scope of Analysis:  Policy Measures Consistent with Treasury's Option 5#

6. Option 5 was Treasury's alternative, preferred option for improving the quality of legislation, as identified in the 2011 RIS prepared by Treasury to help the Government decide whether to support the Taskforce's proposed Bill.

7. The core features of Option 5, as described in the 2011 RIS, are reproduced again in Annex 1. Rather than require certification of all legislation against a set of pre-specified principles, and allow the courts to test the compatibility of legislation against those principles, the approach proposed in Option 5 is to:

• require the disclosure, sometimes with explanations, of key features of a piece of legislation to help readers decide where scrutiny is warranted; and
• strengthen Parliamentary processes for the scrutiny of legislation.

8. Back in 2011, the policy features of Option 5 were described only in indicative terms, and in some cases included alternative options. The purpose of this RIS is to examine the possible features of Option 5 in more detail. Moreover, noting that the 2011 National-ACT Confidence and Supply Agreement refers only to an outcome "based on Option 5" we also consider the potential to refine or expand on the features originally identified. What remains consistent, however, is:

• the goal of improving the quality of government legislation; and
• the focus on transparency and support for Parliament's legislative scrutiny role, without bringing the courts further into the legislation-making process.

The Nature of the Problem and the Case for Further Action#

9. The nature of the problems with the quality of legislation in New Zealand, and the case for taking further action, was more fully canvassed in Treasury's 2011 RIS. For convenience the main points from the 2011 RIS are summarised below.

Objectives and Assessment Criteria#

10. Our 2011 RIS suggested that the key objectives for any additional measures to support legislative quality should be to promote one or more of the following:

• the delivery of better policy analysis and advice into the decision-making process for potential legislative changes (because major problems with legislation originate in choices made at the policy phase and it is a lot harder to get agreement to significant policy changes later on);
• the translation of policy into legislation that conforms to generally accepted attributes of good legislation (because these attributes reflect key values that support individual rights and respect for the law); and
• more systematic identification and prioritisation of problems or potential improvements in the existing stock of legislation (because we know legislative mistakes can be made; that legislation can become less effective or redundant over time as the policy environment changes; and assessing the cost and performance of legislation is a neglected and under-resourced task compared with assessing government spending).

11. This does not mean that we expect proposals based on Option 5 to target all three key objectives, or to offer a complete response to the problems and deficiencies described above and in Annex 2. As noted in the 2011 RIS, the core features of Option 5 are expected to complement existing measures, and to primarily target the 2^nd objective - the translation of policy into good legislation.

12. Our main assessment criteria are the plausible benefits, expected costs, and anticipated risks of any proposed measures. The key issues here are likely to be:

• the ability to manage risks and costs - since the benefits are uncertain, it is desirable that costs can be cut if the benefits turn out to be limited; and
• the nature of the incentives created - as the benefits will come mainly from the ability to induce behavioural changes among the key participants.

13. Important indicators associated with the control of risks and costs include:

• alignment with existing constitutional norms, as constitutional changes can alter the power balance between key roles, and should not be made lightly or without careful consideration of the potential dynamic consequences;
• compatibility with different regulatory philosophies and frameworks, because broad political and public buy-in is required for enduring success;
• alignment with current arrangements for the development of legislation, as this offers potential economies and facilitates system integration; and
• the ability to adapt and amend requirements, in light of both practical experience and the inevitable changes in the legislative environment, supporting institutions, and concepts of good practice.

The case for favouring an "Option 5" approach#

14. Before assessing the merits of different packages "based on Option 5", we should recap the reasons for favouring the "Option 5" approach. As noted earlier, the core characteristics of the Option 5 approach are a focus on transparency and support for Parliament's legislative scrutiny role, without bringing courts further into the legislative process. The arguments that support this focus include:

• Disclosing and explaining policy and process choices made in developing legislation is more accommodating of different regulatory philosophies than reporting on compliance with selected principles stated in fixed form;
• Enhancing Parliament's role as a guardian and promoter of good legislative policy is squarely in line with existing constitutional norms, whereas this would be an essentially new role if given to a court;
• The strong public service norm to obey the law, and other reputational effects, provide incentives to comply without creating legal remedies or sanctions that could bring courts further into the legislative process;
• The provision and scrutiny of additional information on Bills and delegated legislation can be readily integrated into Parliament's existing select committee and Regulations Review Committee processes;
• The scrutiny of legislation able to be provided by Parliament is far more timely, frequent and predictable than that able to be provided by a court;
• By building on Parliament's existing role and processes, we reduce the risk of unintended and unexpected outcomes (particularly from court involvement) that could arise from having requirements in legislation, and we are better able to manage the additional costs that may be incurred.

How will this public disclosure improve the quality of legislation?#

15. The rationale for the public highlighting of selected features and the quality assurance processes for new legislation is that it will increase the attention paid to the matters disclosed, by both:

• interested MPs and external parties (because information of potential interest to them will now be far more readily accessible); and
• legislative decision-makers (because they will have to explicitly identify and explain some of the choices they have made in developing the legislation)

16. This increased attention is, in turn, expected to increase the likelihood that those matters will be addressed in a way consistent with the attributes of good legislation (see page 6), because:

• for interested MPs and external parties, easier access to information relevant to legislative quality makes it more likely they will comment, and, hence, more likely it gets Parliamentary attention; and
• for legislative decision-makers, disclosure increases the visibility, and thus reputational cost, of inadequate quality assurance or poor justifications.

17. The increased attention is also likely to encourage relevant oversight agencies to develop better expert guidance on the targeted legislative features, and better guidance and design improvements for the targeted quality assurance processes, which in turn is likely to improve their use and effectiveness.

The case for a legislative response#

18. The case for a legislative (rather than administrative) initiative to improve the quality of legislation comes down to a matter of judgement.

• At least in principle, legislation is not essential because there is no immediate proposal to establish a new statutory body, confer a new function on an existing body, or provide for legal sanctions. It would be possible for the Government to voluntarily agree, or for the House to amend Standing Orders to require Ministers, to provide the relevant information.
• Legislation has the further disadvantage that it takes more effort to put in place, makes it harder to fix any unanticipated problems; is less able to accommodate or allow for unusual situations; and it is always possible that the courts will find an unexpected role or give an unexpected interpretation.
• However, an obligation imposed by Parliament on departments delivers a far more credible and enduring commitment to proper disclosure on behalf of the government, and also has a much greater status and profile with the public and government agencies, than administrative requirements or Standing Orders, even without legal remedies or enforcement mechanisms.
• There is plenty of precedent, both domestically and internationally, for using legislation to require the government to disclose information to Parliament.
• Legislation is also the more appropriate vehicle where Parliament wishes to reach down past Ministers, and impose obligations directly on departments.

Possible Packages of Measures for Giving Effect to an Option 5 Approach#

19. There are a number of measures that could be adopted to encourage improvements in the quality of legislation that are consistent with the Option 5 approach. Rather than attempt to assess the impact of every permutation, we have identified three illustrative packages of measures (in addition to the status quo) as a way to represent the range of options available.

• Each package has a theme, but most measures are discrete options that can be added to, or subtracted from, any package in a modular way.
• The essential differences between the illustrative packages lie in their level of ambition and, correspondingly, their likely resource costs.
• The legislative core of all packages is the requirement for a disclosure statement to accompany Government Bills, major amendments to Bills, and significant delegated legislation (see Annexes 3, 4 and 5 for details).
• Each package is a mix of legislative and non-legislative measures, but the differences between packages relate to the scope of the non-legislative measures (see Annex 6 for more details of these measures).
• There are, however, definite limits on the range of disclosures we think are robust enough to be implemented as legislative requirements at this point. We have done some testing of key disclosures, but not enough to yet be confident that all suggested disclosures are workable in all likely situations.

20. The packages are:

"Maintain the Status Quo"

• Continue with the existing regulatory management tools at this time

Package 1: "Legislate but Keep the Cost Low"

• A basic set of disclosures for Bills and delegated legislation, identifying key legislative features and information on quality assurance processes
• Provision of guidance and admin support to departments from PCO/Tsy, and provision of extra select committee support from the Office of the Clerk
• A one-off independent review to commence within 5 years of enactment

Package 2: "Legislate and Reinforce with Other Changes"

Incorporates all the features of Package 1, but also ....

• Some extended disclosures for Bills, and extended coverage of delegated legislation, to be trialled administratively
• Supporting measures to encourage more timely and rigorous identification of potential issues earlier in the legislative development process, such as:
  • Tightening RIA expectations for reporting on expected costs, possible economic losses, likely compliance rates, and implementation plans;
  • Reviving the Legislation Design Committee, with a more proactive role to give advice to departments on complicated legislative projects;
  • A revamped Cabinet Legislation Committee, with a greater mandate to consider legislative quality issues and test implementation plans.

Package 3: "Legislate and Lift Attention to Existing Legislation"

Incorporates all the features of Package 1 and 2, but also ....

• A series of complementary measures intended to significantly lift the attention also given to the management of existing legislation, such as:
  • Clearer expectations of government agencies for the implementation, administration, enforcement, monitoring and review of legislation;
  • Reporting requirements for regulators and administrators of legislation that are better tailored to a regulatory role or perspective;
  • A possible new Office of Parliament role to review how well different regulatory regimes perform against their stated policy objectives.

Overall Conclusions and Recommendations#

21. The indicative packages of measures outlined above represent different levels of ambition. In each case the level of achievable benefits is uncertain, but we think the risks and costs can be managed so they do not exceed the benefits achieved.

22. It may not matter too much which combination of measures is pursued, but we do recommend a conservative approach to specifying requirements in legislation initially. By trialling more ambitious disclosures as administrative requirements in the first instance, we minimise the risk of locking in unworkable or low value provisions, and improve our ability to manage the level of costs incurred if the disclosure statement approach is less successful than hoped.

23. Legislation offers not only a credible and durable commitment, but also a good foundation around which a range of non-legislative measures can be introduced.

24. Sticking with the status quo is an option, but the underlying concerns that have driven demand and support for a Regulatory Standards Bill would remain, and would be likely to reappear in the form of new proposals before too long.

Consultation#

25. Information on consultation that occurred in the lead-up to Cabinet's decision to introduce the present Regulatory Standards Bill can be found in our 2011 RIS.

Treasury discussion document with indicative legislation

26. In mid August 2012, Treasury circulated a discussion document, together with indicative legislative provisions, for a targeted consultation process. It sought early views on the merits and workability of a set of possible provisions for a revised Bill, to help inform Treasury advice. Feedback was sought from:

• key regulatory or legal contacts in departments and in the Office of the Clerk of the House of Representatives;
• a variety of external public law experts, including Law Society committees and lawyers from several major law firms,
• former Regulatory Responsibility Taskforce members, and
• several bodies representing business interests that have previously shown a strong interest in the Regulatory Standards Bill.

27. Treasury also sought the cooperation of a few departments to attempt test disclosures for a few of their recent Bills or disallowable instruments, and give us feedback on any problems they identified and on the level of effort required.

28. Feedback provided by officials and public law experts elicited a lot of useful comments on the form or scope of suggested provisions. These have influenced the selection, and will inform any drafting, of the legislative provisions suggested. On the whole, the feedback suggested to us that a Bill requiring a set of core disclosures for government Bills and significant delegated legislation is workable, and should impose only small additional compliance costs.

29. Among the public law experts, there was also a general consensus that a Bill along these lines had the potential to establish a useful impact, and was worth progressing. By contrast, former Taskforce members and business bodies made few comments on particular disclosures but, in general, thought the proposal was far too weak and indicated their continued preference for the current Bill.

30. Interestingly, the matters most likely to attract suggestions from external parties for greater disclosure (the LAC guidelines, NZ's international obligations, rights and interests affirmed by the Treaty of Waitangi, and regulatory takings) were basically of the same type – that is, existing quality assurance expectations lacking a specific process or product (though in the case of regulatory takings, not even an existing expectation). As noted in Annex 4, we believe that this type of matter does not make a great disclosure requirement because their scope is broad and often not widely understood, which creates a high risk of either idiosyncratic interpretations or carefully worded, limited responses designed solely to minimise reputational risk.

31. Nonetheless, since they are clearly seen as important, we have made a margin call to include basic disclosures relating to the first two matters – consistency with NZ's international obligations and with the principles of the Treaty of Waitangi – and a disclosure relating to material economic losses, which would cover regulatory takings. But we still reject a disclosure relating to consistency with the LAC guidelines because we think the advisory nature and extremely wide scope of the LAC guidelines means there is no prospect of receiving either a meaningful Yes/No response, or a meaningful description of the process gone through to assess consistency. If a specific quality assurance process or product were to be agreed and introduced by a future government for any of these matters, the likelihood of providing more useful disclosures would be greatly increased.

Formal departmental consultation

32. Formal departmental feedback was sought on a draft Cabinet paper and an earlier version of this RIS. Departments did not identify any substantive concerns with the proposed measures, which contrasts with their earlier comments on the Taskforce's Bill. However, a number of departments noted that the proposals would add to pressure on their resources and, in the absence of additional funding, it will either take departments more time to prepare regulatory advice or require greater prioritisation of their regulatory work.

Implementation#

33. For any of the Packages other than the status quo, some lead-in time will be required to develop guidance, templates and systems for the preparation, delivery and publication of the proposed disclosure statements, as well as make consequential changes to Cabinet's LEG requirements and to Standing Orders.

34. Under Packages 2 and 3, further work would be required to revamp or extend existing administrative requirements, and develop new performance expectations for government agencies. These can, however, be introduced progressively and need not precede the implementation of the disclosure statement requirement.

35. There are some resource costs associated with these measures. Most measures can be implemented with or without additional funding to meet those costs but, in areas like the rollout of new expectations, the absence of additional financial support will affect the speed of implementation and likely levels of compliance.

Monitoring, Evaluation and Review#

36. Packages 2 and 3 propose explicit trialling of some extended disclosure requirements that will require the Treasury and PCO to undertake ongoing monitoring of departmental experience in producing disclosure statements. Monitoring is an essential part of the risk management strategy for the Packages.

37. More generally, the proposed disclosures and supporting measures are new requirements that have not been extensively tested, and the nature and size of the behavioural impacts are necessarily speculative. Consequently, an eventual review of experience with the new arrangements will be essential to determining whether they are truly fit-for-purpose and cost-effective and whether amendments or additional measures may be desirable to further improve legislative quality.

38. Our 2011 RIS originally proposed a statutory independent review of the operation of the whole regulatory management system every 5 years as part of Option 5. Experience with statutory review clauses in other legislation, however, has shown them to be a mixed blessing. It has the merit of ensuring an actual review takes place (when intentions and promises are often forgotten or ignored in other circumstances), but anticipating the best timing up-front is hard, and the scope and quality of the review will depend on the enthusiasm of, and other pressures faced by, the responsible Minister or agency at that particular time.

39. On balance, we now suggest that any legislation include provision for a single independent review of the operation of that legislation, which must commence within 5 years of the legislation coming into effect and be completed and tabled within 12 months. Whether the review also addresses wider issues, and the timing of any further reviews, would be left to administrative discretion.

Endnotes#

Regulatory Standards Bill, 2011 No.277-1,

http://www.legislation.govt.nz/bill/government/2011/0277/latest/DLM3601205.html

Report of the Regulatory Responsibility Taskforce, 30 September 2009

http://www.treasury.govt.nz/economy/regulation/rrb/taskforcereport

Interim Report on the Regulatory Standards Bill, Commerce Committee, 30 September 2011

http://www.parliament.nz/en-NZ/PB/SC/Documents/Reports/8/1/c/49DBSCH_SCR5309_1-Interim-report-on-the-Regulatory-Standards-Bill.htm

Confidence and Supply Agreement with ACT New Zealand, 5 December 2011

http://www.parliament.nz/NR/rdonlyres/5634F13B-7744-4D03-A9F9-EBB0E4F83621/207925/NationalACT_Confidence_and_Supply_Agreement0512201.pdf

Regulating for Better Legislation - What is the Potential of a Regulatory Responsibility Act?, Regulatory Impact Statement, The Treasury, 2 February 2011

Annex 1#

Annex 2#

Annex 3: Disclosure of Information to Support the Scrutiny of Legislation (included in all 3 Packages)#

The core of Option 5 is the proposal that information that might help in assessing the quality of a piece of government legislation should be made readily available to interested external parties at the time a Bill or substantive SOP is introduced, or delegated legislation is enacted.

This annex briefly explains the core features of the proposed disclosure mechanism. The selection of matters to be disclosed is discussed and explained in Annex 4 and 5.

Annex 4:  Matters Proposed for Disclosure about Bills  (included in all 3 Packages)#

The objective of the proposed disclosure requirement for Government Bills is to focus more attention on whether a Bill has the attributes of good legislation. It would do this by providing members of Parliament and the public with information to help support their scrutiny of a Bill while it is being considered for enactment by Parliament.

The proposal is limited to Government Bills, because the intention is to legislate for the proposed disclosures and this is only appropriate where the obligations are to be imposed on persons or bodies that are not members of Parliament.

It would, however, be open to the House to agree through amendments to Standing Orders that the same or similar disclosure requirements should apply to Local Bills, Private Bills or Members Bills, which are not the responsibility of the government.

Desired characteristics of matters for disclosure#

There are many matters that could be disclosed that might potentially help the reader assess the quality of a Bill. However, the more matters that must be routinely disclosed, or the more detailed the disclosure required, the greater the time and cost of its preparation, and the less accessible the material may become for the average reader. Some selectivity is therefore essential.

Consequently, the choice and form of matters proposed for disclosure has been influenced by a preference for disclosures that are:

• significant for quality;
• clear in concept;
• easy to produce;
• capable of concise expression; and
• likely to minimise potential duplication.

On the whole, this favours legislative characteristics that are well understood, and quality assurance processes built on established government expectations.

The sources we have drawn upon#

For the proposed matters for disclosure we have drawn primarily on:

• existing Cabinet requirements and expectations relating to the development of legislation, for disclosures relating to quality assurance products and processes
• the Legislation Advisory Committee (LAC) guidelines, for disclosures of significant powers and other features warranting careful scrutiny; and
• matters specified for inclusion in the explanatory notes accompanying Bills and delegated legislation in the United Kingdom and Australia.

The merits and risks of the different types of proposed disclosures#

The matters suggested for disclosure below fall into seven different categories or types, with the following general merits and risks:

1:  A general policy statement for the legislation

A general policy statement provides a succinct statement of the legislation's policy objectives and the way in which it seeks to meet those objectives. It helps the reader interpret the intent of the provisions in the legislation, and is a good starting point for considering the merits and performance of the legislation.

This is a longstanding requirement for Bills, currently included in the explanatory note. Unlike the rest of the explanatory note, however, the prime responsibility for its drafting rests with the administering department rather than the legislative drafter. There is a risk that Parliament will not want the general policy statement moved into the disclosure statement.

2:  Established quality assurance products and associated background documents

Established quality assurance products, and other background documents that can inform the consideration of the merits of the legislation, make excellent disclosure requirements because they are easy to identify, usually take a standard form, and do not require the generation of new material.

The fact that they exist as established products tends to mean they cover issues with significant implications for legislative quality. Unfortunately this does not guarantee the product itself is of a high standard. But particularly where it concerns the underlying policy, a poor QA product is usually obvious enough that the reader can use that to form a judgement on the quality of the policy work undertaken, which in turn may inform their view of the likely quality of the legislation on which it is based.

3:  Established quality assurance processes, lacking a specific product

Established quality assurance processes, involving consultation with a mandated expert agency, can make good disclosure requirements because it is usually easy to identify whether consultation was required and undertaken, and this generates a pretty simple and concise disclosure.

The disadvantage of not having a specific product to disclose is that the reader will not know what the process threw up, at least without new information having to be generated, and must otherwise rely on the confidence the reader has in the care, expertise and influence of the expert agency consulted.

4:  Existing quality assurance expectations, lacking a specific process or product

Existing quality assurance expectations without a standard process or product do not make ideal disclosure requirements because the only informative disclosure involves outlining the process gone through to identify any potential problems. The more specific the expectation the better, as it reduces the length of the disclosure, and is more amenable to the reporting of quite specific actions taken.

The more broadly framed the expectations, such as assuring consistency with NZ's international obligations or with the principles of the Treaty of Waitangi, the greater the risk of idiosyncratic interpretations and levels of disclosure, or alternatively of routine resort to very carefully worded, standardised responses designed to minimise reputational risk rather than to inform a reader.

We do support a disclosure for the two cited expectations above, based solely on their potential significance for legislative quality, but this is a marginal call overall. We have, however, rejected a disclosure for the even more general Cabinet expectation that legislation will be "consistent with the LAC guidelines", as we believe the advisory nature and wide scope of the LAC guidelines means there is no prospect of receiving either a meaningful Yes/No answer or a meaningful description of the process gone through to assess consistency.

5:  Significant powers conferred by the legislation

Significant powers conferred by a piece of legislation make a good disclosure because they highlight features that have huge potential to affect the rights and freedoms of individuals. Explicitly acknowledging the relevant powers not only prompts the reader to think about whether the extent of the power is appropriate and necessary, but also think about the nature of the safeguards that should apply to the exercise of the power.

The challenges with this type of disclosure are, first, to set criteria that will consistently capture the most significant powers, and second, to extend the disclosure to include useful explanations without significantly adding to disclosure length or cost.

6:  Unusual features of the legislation warranting careful scrutiny

Unusual features make a good disclosure if they indicate a potentially significant risk to individual rights and can be clearly and readily defined and identified. Most such features are "unusual" because they depart from well-established legal presumptions.

A disadvantage of this type of disclosure is precisely that the feature occurs infrequently and so leads to a huge number of "No" disclosures. Another challenge is targeting the feature of most interest, because the underlying legal principles cannot be easily reduced to a simple and stable rule.

7:  Potential key impacts of the policy being given effect by the legislation

Singling out potential key impacts of the proposed legislation does not provide an ideal disclosure because impacts are a matter of judgement, and their relative significance a matter of values. Our strong preference would be to refer readers to any RIS provided, which ought to identify all the key policy impacts.

Unfortunately, not all legislative proposals have RISs, the policy may have changed or the policy further fleshed out since the RIS was prepared, and the RIS itself may be based on poor or inadequate analysis. In these circumstances it may be worth asking whether subsequent analysis has occurred that further illuminates key impacts such as costs, any potential significant economic losses, and levels of expected compliance and enforcement.

Specific Matters Proposed for Disclosure#

The tables below identify the specific matters for which a disclosure could be sought under one of the seven different types of disclosures identified above.

Basic and extended disclosures

For some of the seven types of disclosures, there are choices to be made between:

• a "basic" disclosure, which is relatively concise, low cost and easy to answer, but may only hint at a possible issue (requiring the reader to investigate further); and
• an "extended" disclosure, which provides additional information useful to the interested reader, but requires more time and effort to produce, and might be difficult to frame in a way that is simple or universally applicable.

As a general rule, we believe:

• the "basic" disclosures will be workable in almost all Bills, and hence are suitable for inclusion in legislation;
• the "extended" disclosures will further lift the attention to good practice, but really need to be trialled or tested for a period to give confidence that each matter is framed in an appropriate and cost-effective way, before any consideration is given to their incorporation in legislation.

At the bottom right hand side of each table below we indicate which requirements we think could be included in the proposed legislation, and which we think should be tested further.

Government Changes to Government Bills#

Bills are frequently changed to a greater or lesser extent after they have been introduced to Parliament. The select committee process is a major source of changes. Another is amendments presented in the form of Supplementary Order Papers (SOPs), which may be introduced by either government or non-government Members.

Since changes post-introduction can have a significant impact on the content and effect of legislation, it makes sense in principle to also expect a government agency to flag and explain the relevant features of any government amendments proposed outside of the select committee process - just as it would have been required to do if the amendments had been included within the relevant Government Bill on introduction.

However, many changes made by government SOPs are very minor, for example they might correct cross-references and definitions. Requiring disclosures for minor or technical SOPs would result in little useful information for the effort and time involved. This could be overcome by imposing a supplementary disclosure obligation only when the amendments would have materially changed the original disclosure statement.

There are two potential approaches to presenting such a disclosure. One is to provide a supplementary disclosure statement to the original, and the other is to provide a new revised version of the full disclosure statement. The best option may depend on the nature of the proposed changes, which suggests allowing for some flexibility in the approach that may be adopted to presenting the effect of the changes.

Annex 5:  Matters Proposed for Disclosure about Delegated Legislation (included in all 3 Packages)#

The objective of the proposed disclosure requirement for delegated legislation is to provide the public and members of Parliament (especially members of the Regulations Review Committee) with information to support their scrutiny of the legislation.

Standing Orders permit the Regulations Review Committee to examine and report on all regulations, and to investigate complaints from the public on the operation of regulations. Members of the committee can also give a notice of motion to disallow a regulation, which must then be dealt with by the House.

Coverage of Delegated Legislation#

There is a huge and extremely diverse range of delegated legislation for which a disclosure statement could in theory be contemplated. But if the intention of providing a disclosure is to inform opportunities for scrutiny, then it makes practical sense to limit coverage to those instruments that are subject to Parliamentary disallowance.

Under the new Legislation Act 2012, the range of delegated legislation that will now be subject to disallowance is likely to increase, including not just Orders in Council but also other instruments defined as having "significant legislative effect". Feedback from departments suggests that this new definition of ‘disallowable instruments" is likely to include a range of instruments for which a requirement for separate disclosure statement may not always be useful or cost-effective (e.g. fisheries notices).

Consequently it is proposed that the disclosure obligation should be limited to those disallowable instruments that are drafted by PCO in accordance with the Legislation Act. Further consideration can be given under Packages 2 and 3 to determining whether there might be other types of disallowable instruments that might routinely justify the cost of preparing an accompanying disclosure statement, and testing this administratively.

Matters Proposed for Disclosure#

We have similar ambitions for the quality of delegated legislation as we do for primary legislation. Nonetheless, the key concerns for delegated legislation are not always the same as for Bills. It is far less likely, for example, that delegated legislation will be inconsistent with important legislative principles, independently of its empowering Act. However, quality assurance products and processes take on more importance because delegated legislation does not have to go through a challenging select committee process that is informed by public submissions.

The following subset of proposed disclosures for Bills, relating almost entirely to quality assurance processes, seems the most highly relevant to delegated legislation.

Annex 6: Outline of Administrative Support Measures#