Derek Gill*

“Laws are like sausages. It is better not to see them being made.”[1]

7.1 Introduction

This chapter will focus on addressing a primary question: What are the key issues and tensions facing the current New Zealand regulatory management system? Regulatory management attempts to “regulate the regulation makers”by requiring well-designed policy responses to clearly articulated problems. The chapter explores the apparent paradox that it is not clear what causes the problem of regulatory quality that regulatory management regimes in New Zealand (and elsewhere) aim to address and how the regime will achieve the desired behaviour change.

This chapter explores the key issues by applying the logic of regulatory management to the design of regulatory management itself. The development of more detailed analysis in the second phase and the regulatory toolkit in the third phase of the project, will include addressing questions about “what is to be done”with the regulatory management system in New Zealand.

7.2 Background

The policy objectives of regulatory reform are multiple and include both “better quality”regulation through more effective alignment of regulatory means to achieve policy goals, as well as “regulatory relief”through administrative simplification and deregulation to reduce the burden of regulation. In New Zealand in the late 1980s and 1990s, as in almost all other OECD countries, there was a move to deregulate product, capital and labour markets and, more recently, there has been an emergence of some re-regulation.[4]

Figure 1 below shows the spread of Regulatory Impact Assessments (“RIAs”)[7] from a handful of OECD countries in 1990[8] to include all OECD countries by 2008. There has been a rapid take-up of new approaches to regulatory management in parallel with the wave of reforms to public sectors in the 1980s and 1990s across the OECD. After the event, academics labelled the latter New Public Management (“NPM”).[9]

The intellectual impetus for deregulation came from a long scholarly tradition in law and economics which questioned whether “public interest”regulation actually served the public interest by addressing “market failures”because of problems with government failures. Government failures included problems with selection of policy instruments, asymmetric information, the gap between policy design and implementation as well as regulatory capture.[10]

Regulatory management, like NPM more generally, in New Zealand at least, appears to be primarily a practitioner-driven response to perceived problems of regulatory quality, but without coherent and well-developed analytical underpinnings. As the regulatory state pulled back from direct service provision, and increased its use of regulations, questions were asked about how well the state was managing its interests through regulation. Indeed the very notion of regulatory management did not exist in the academic literature until the 1990s and there was no corresponding tradition of scholarship.[11] Most of the academic literature on regulatory management is focused on RIAs, rather than the wider regulatory management system of which RIAs are just part. Outside the United States, most accounts start with the OECD’s 1995 publication “Recommendations on improving the quality of government regulation”. An initial literature review[12] undertaken for this chapter identified a largely descriptive literature on what is being done using RIAs, but very little that explained why or how regulatory management has emerged.

This chapter is organised to address six questions:

1 What is regulatory management?

2 What are the key elements of the regulatory management system in New Zealand?

3 How does the New Zealand system compare with arrangements in similar jurisdictions?

4 Why have regulatory management?

5 Is there a public policy rationale for regulatory management?

6 What are the issues that regulatory management policy in New Zealand needs to address?

Each of the following sections addresses one of these questions.

7.3 What is regulatory management?

Before considering regulatory management in detail, it is important to consider what is meant by regulation in this chapter. From an economic perspective, regulation is just one of a number of policy levers or interventions governments can use. In different sectors within an economy, and in the same sectors across different countries, these components are packaged together in quite different ways. As James Q Wilson observed “try to think of a government activity that has not been done or is not now being done by a private firm …it is not easy”.[13] Interventions are made up of packages that comprise five basic components:

• Purchasing of real goods and services (from both public-owned suppliers such as public schools and private providers such as private schools);
• Changing relative prices by providing consumers with subsidies (for example, installing residential insulation) or levying specific taxes (such as alcohol and tobacco);
• Redistributing incomes through transfer payments (social welfare) and taxation;
• Owning providers who sell outputs on private markets or whose outputs are publicly financed (schools);and
• Regulating private and public activities by prohibiting some activities altogether (such as drugs) and restricting how others operate (professions, for example).

Regulation broadly includes both the general framework of laws as well as specific prohibitions and privileges. As an example of general framework legislation, incorporation under the companies’legislation has played a central role in the emergence of sustained economic growth and prosperity in Europe from the 18th century onwards.[14] Specific prohibitions may sometimes enhance wealth or achieve some other important public policy purpose, but in other cases they do not. For example the Pharmacy Act 1939 was an anti-competitive measure to protect sole proprietor chemists from the expansion of a corporately owned chain (Boots the Chemist) specifically precluding corporate ownership of chemists in New Zealand. This protection benefited a specific group at the expense of all consumers and persists on the statute book to this day.

We looked in vain in the literature for a definition of the regulatory management system. Jonathan Ayto from the New Zealand Treasury[16] on an early draft of this chapter usefully provided the following definition.

The term “regulatory management system”will be used in this chapter to mean how the formal government system, documented in legislation and the conventions in the Cabinet Office Manual, is augmented with features that apply specifically to primary and secondary legislation and tertiary regulation. “Specifically”is highlighted to bring the focus onto the special measures and bespoke features of a regulatory management system that do not apply to the general business of government. These special measures would not apply in say, appointing a board member to an organisation or exhorting New Zealanders to some worthy aim through the release of a new “strategy”document.

According to OECD guidelines[17] on “good”regulatory management there are four core elements of a regulatory management framework:

• Regulatory Policies –a systematic government-wide approach to the use of regulatory instruments;
• Regulatory Tools –administrative simplification, sunset provisions, public consultation requirements, regulatory review and evaluation, compliance with enforcement guidelines, alternatives to traditional regulation, Regulatory Impact Assessments (“RIAs”);
• Regulatory Institutions –with responsibility for centralised regulatory oversight in the executive and the legislature;and
• Regulatory Procedures –administrative procedures controls, due process requirements, rules on giving notice and communication, training and so forth.

The Victorian Government[18] distinguishes four stages of the regulatory process: policy development and design, implementation, administration and enforcement, and review. In this chapter the policy development and design phase will be divided further by distinguishing “big”policy design from the “little”policy phase involving legal analysis, development of drafting instructions and the drafting of the legislation or regulations. Thus regulatory management (“regulating the regulation makers”) includes both regulatory policy making (“regulating regulation developers”) and regulatory administration and enforcement (“regulating the wielders of regulatory power”).[19]

Different actors are involved in different phases of the regulatory process in New Zealand. Bureaucrats and ministers in the executive government dominate the big policy phase, the legislature is heavily involved in the little policy phase of development of legislation (but not secondary or tertiary regulation), while the courts are involved in the administration and enforcement of regulation. The involvement of the citizens and businesses is more complex.[20] The extent to which the public affected (and the general public) are involved in the design phase varies case by case quite significantly. The involvement of different actors at different stages in the process also differs depending on whether legislation or secondary and tertiary regulation is involved. In short, mapping out the regulatory management systems applying in individual jurisdictions is quite complex (as is clear from Annex 2) and comparing across jurisdictions even more so.

Before developing the main lines of argument in this chapter it is important to explore the role of interests in regulation. Public interest theories start from the idea that regulations are developed in the public interest (that is, based on efficiency problems due to market failures or equity concerns about income distribution).[21] There are a number of objections to this approach including the problem of defining the public interest and the assumption that regulators are disinterested. The economic theory of regulation starts from the opposite proposition;that regulation is developed not as a response to the public interest, but to serve the interests of a particular group. As the benefits of regulation are often concentrated, and the costs are diffuse, the regulation will be captured by the regulated industry. The introduction of the Pharmacy Act in 1939, discussed above, is an instructive New Zealand example.

Neither public interest nor capture theory stand up well as general positive theories of regulation: “The claim that agencies are systematically biased in favour of regulation finds little support in public choice theory the political science literature or elsewhere.”[22] James Q Wilson[23] uses case studies drawn from public administration to identify four types of relationships between regulatory regimes and their agencies shown in Box 1 below. Even in cases where benefits are highly concentrated and where costs are dispersed, one observes both client-captured agencies and entrepreneurial agencies. James Q Wilson uses the Federal Drug Administration as an example to suggest “agencies born of entrepreneurial politics are at risk of capture but capture is not inevitable”.[24]

In the light of this introduction to regulatory management and the inconclusive role of interests in regulation, the next section will explore the practice of regulatory management in New Zealand.

7.4 What is the formal regulatory management system in New Zealand?

New Zealand was an early adopter of regulatory management approaches, introducing a requirement for Cabinet papers to include business compliance cost assessments being introduced in the early 1990s, and RIAs in 1998. The regulatory management regime has grown over time as the focus shifted from the reduction of compliance costs facing business to a broader objective of improving regulatory quality. Box 2 below discusses how New Zealand rates relatively highly on a range of international comparisons on the quality of regulation. The OECD, in a recent comparative analysis, found that, relative to other OECD countries, New Zealand (and Australia) showed high emphasis on policy coherence and much less emphasis on simplification.[25]

Responsibilities for the overall regulatory management system transferred from the Ministry of Economic Development (“MED”) to the Treasury in 2008, although MED retains an ongoing role in regulatory reform. After the election of a new government in 2008, regulatory management took on an increased emphasis with the creation of a new Ministerial portfolio allocating responsibility for regulatory reform. At the same time there was a strengthening of regulatory process requirements and institutional strengthening.

• Regulatory Policy –The 2009 Government Statement “Better Regulation, Less Regulation”, a regulatory review work programme;
• Regulatory Institutions –Treasury RIA team (“RIAT”) providing strategic co-ordination of the regulatory quality system, the Legislation Advisory Committee (“LAC”), and Legislation Development Committee (“LDC”), parliamentary scrutiny, Regulations Review Committee (“RRC”), RIAT training;
• Regulatory Procedures
• Ex ante –an enhanced Regulatory Impact Analysis (“RIA”) regime including a greater role for RIAT, agency disclosure statements, ministerial certification requirements;and
• Ex post –Regulatory scanning by agencies, annual regulatory plans, annual regulatory reporting, post implementation reviews, an Omnibus Regulatory Reform Bill, Regulations (Disallowance) Act 1989, RRC reports.

There are significant differences in the regulatory management system applying to different types of regulation –primary, secondary or tertiary. Annex 2 maps out the regimes applying to the different types of regulation in New Zealand and highlights the special measures in the regulatory management system. The special measures applying to primary legislation, which are covered in the first section of Annex 2, are mainly focussed at the front end “big policy”design phase. By contrast there is relatively little focus on the legal development or administration and enforcement phase and an emerging focus on review.

The situation with secondary and tertiary regulation is more complex as it depends upon whether the Treasury RIA Team includes the regulations within the RIS process. Regulations outside of the RIS process, which includes much tertiary regulation, are subject to very limited ex ante screening or ex post review. Secondary regulations subject to the RIS process are subject to more ex ante review within the executive but without the process of parliamentary scrutiny other than the Regulations Review Committee.

The Bill as proposed has attracted considerable criticism, in particular from the legal community,[26] but also interestingly in the RIS prepared by the Treasury. The latter’s RIS on the Regulatory Standards Bill provides a long list of “symptoms”or limitations of the existing regulatory management regime.[27] This assessment also contains an impressive list of processes underway in the policy development, executive government decision-making and legislative consideration phases, which correspond to the design and development phase in the Victorian Government schema in Figure 3 below. What is striking is the relative shortness of the list after the development and design phase. The RIS only lists regulatory review, judicial review and the recently introduced systematic process of legislative scanning. As will be discussed in At [7.7], however, the list of “symptoms”does not provide a “clear diagnosis”of the “disease”which caused the symptoms.

7.5 How does the New Zealand approach compare with system components in comparable jurisdictions?

This section will briefly compare the New Zealand regulatory management system with that of the Australian federal government and the states and territories. The review has occurred at two levels –at the macro level of the institutions and tools and at the micro level of how individual regulations are developed.

Table 1: A Comparison of the Features of Regulatory Management Regimes

Starting with the macro level, Table 1 provides a comparison the key regulatory institutions and tools in New Zealand with those of the Australian federal government and all eight Australian states and territories combined. A key finding shown in the Box 4 is the degree of convergence or institutional isomorphism. For example, RISs, RIAs and some consultation requirements for the development of new regulation were common to all jurisdictions. Almost all jurisdictions have introduced some form of review of the stock of existing regulation. Unlike a number of Australian jurisdictions, the New Zealand regime does not include a sunset clause provision, a business consultation website, or a requirement that all regulations be held and disclosed on a formal centralised register.

Reviewing the institutions and tools, however, is only part of what is required. Regulatory management is meant to be a system where the whole is greater than the sum of the parts. Understanding the system requires an exploration of how the components fit together. As a result, the chapter explores how regulatory development occurred in New Zealand compared to the Australian Federal Government by briefly looking at the process from beginning to end.

The comparison of the Australian federal system to that in New Zealand suggests:

• Australia has a stronger bureaucratic centre – the Australian federal system has a more formal central agency-driven system of policy co-ordination.

• Public consultation occurs in different phases –in Australia there is more emphasis on extensive formal public consultation with selected stakeholders in the policy design and legal development phase through the use of exposure drafts. Only contentious legislation was referred to Select Committee in Australia. By contrast, virtually all legislation is referred to Select Committee in New Zealand (the default setting is a six-month submission process), but the use of formal consultation in the design phase was more uneven.

Thus at a macro level, while there is considerable similarity on the features of the regimes, at the more micro level of how individual regulations are developed, there are important differences in the systems across different jurisdictions. The implicit assumption behind regulatory management systems appears to be that problems experienced in the quality of policy analysis are undermining the effectiveness of the design of regulations. Overall, the main focus of regulatory management regimes in all Australasian jurisdictions appears to be an attempt to “regulate the regulation makers”by strengthening the “big”policy design phase at the start of the regulatory process.

Looking across different jurisdictions in Australasia, there are important and subtle variations that warrant explanation. Exploring these differences is beyond the scope of this chapter, but the author suggests that regulatory management regimes have largely been “welded”onto existing machinery of government arrangements. These machinery of government arrangements have their origins in the unique constitutional evolution of the different jurisdictions. The next section will explore the role of select committees in providing technical quality assurance in the New Zealand system.

Parliaments in all Australasian jurisdictions conform to Ladley’s law. In brief, Ladley’s law states that Oppositions “do not criticise government policy in order to improve it, they attack the policy in order to overthrow the government”.[28]

New Zealand’s experience is that opposition reserve policy criticisms to debates in the House, consistent with Ladley’s law. In Select Committee this party political dynamic is temporarily suspended. Under Standing Orders (the rules Parliamentarians set to bind themselves) a Select Committee may recommend amendments to a Bill provided these are consistent with the Bill’s principles and objectives. As a result while Select Committees in New Zealand can recommended “big”policy changes, the process of review generally results in significant technical improvements to the legislation at the “little”policy level. McDowell and Webb observe “while most amendments are concerned with specific details of the Bill, there have been occasions where substantial amendment has occurred. However, Select Committees rarely alter or affect the policy initiatives themselves”.[30] In a similar vein, Palmer and Palmer in Bridled Power observe:[31]

MPs feel that their most constructive legislative role occurs in Select Committees, as indicated by the remarks Dr Martyn Finlay made …in 1978 opposition there [in Select Committees] ceases to be fruitful, and destructive criticism should give way to constructive suggestions.

7.6 What explains the adoption of regulatory management regimes?

The previous section discussed the New Zealand regulatory management regime and briefly compared it with state and federal systems in Australia. It suggested that despite the macro system level similarities in the regulatory institutions and processes, there were important differences at the micro level of the design and implementation of regulations. This section of the chapter surveys the factors that might explain the adoption of regulatory management regimes by a number of OECD countries. The next part turns to normative concerns about the public policy rationale for the introduction of regulatory management regimes. The next section will follow the literature in focussing on RIAs as the central regulatory management tool employed in OECD countries.

Figure 1 below shows the spread in the adoption of RIAs to include almost all OECD members by 2008. RIAs are also being adopted in non-OECD countries.[32] The spread and diffusion of regulatory management regimes generally, and RIAs, in particular is not necessarily the same as a process of international policy convergence.[33] As Radaelli observes:[34]

RIA appears to be a typical solution in search of its problem. In fact, the problems with which RIA is associated differ widely across countries. RIA is an attempt to tackle the problem of competitiveness in Australia. It becomes a solution to the problem of credibility in the process of liberalisation and economic integration (via NAFTA) in Mexico. It certainly was a solution to the problem of ‘rolling the state back’in the early days of compliance cost assessment in the United Kingdom. It is an instrument geared towards the general aim of simplification and the ‘slim state’in Germany. It is a way the EU tries to cope with the problem of legitimacy of its regulatory system.

There are a number of possible positive explanations for the international emergence of the institutionalisation of regulatory management regimes (including RIAs) in executive and legislative decision-making processes.[35]

One potential line of explanation is the association between the growth of regulatory management and new public management more generally. Performance-based budgeting emerged in a range of OECD countries as a largely practitioner-driven development in response to frustration with the limits of “line item”input-based budgeting. The incremental nature of budgeting and the tendency of the “base”funding of established agencies and programmes to be rolled over each year prompted concerns about the lack of responsiveness of budgeting to changing political and societal needs. Similarly, the introduction of the RIA can be seen as aiming at improving management of the stock of existing regulations by attempting to ensure that increments of scarce political capital and legislative time are allocated to the highest overall priorities. At [7.4] above, we discussed how more recently in Australasia increasing emphasis is being placed on improving the quality of regulatory enforcement and administration. There is another parallel as well. Both public sector management reform and regulatory management regimes have to be grafted onto existing institutional and constitutional arrangements (the role of the courts, supreme audit institutions, legislative processes, merit-based public services and so forth). In that sense it could be argued that regulatory management is unfinished business arising from the emergence of NPM.[36]

The argument above emphasised the origins of regulatory management in individual nation states as a tool for improvement of regulatory quality. The growth of regulatory management regimes has occurred across a range of OECD countries. An additional explanation is the international diffusion of regulatory management, like public management more generally, through “policy transfer”[37] and/or “policy learning”.[38]

To illustrate this international diffusion argument, the United States was the first country to introduce an institutionalised regulatory management regime by establishing the Office of Information and Regulatory Affairs (“OIRA”) in the Office of Management and Budget (“OMB”). The OECD then acted as a key channel of diffusion. The techniques and practices developed in OIRA were then adopted as “best practice”when key OMB staff (such as Scott Jacobs) moved to the OECD. Through the channel of the OECD, RIAs and other regulatory management techniques were then diffused to other OECD countries.[39] Figure 1 looks like a classic curve for the take-up of an innovation or the spread of a contagious disease through a herd (depending upon the perspective on regulatory management).

The term in institutional sociology for this sort of diffusion is “institutional isomorphism”.[40] Institutional isomorphism is one of those academic concepts that, while hard to use in everyday discussion, is nevertheless commonly found in everyday life. Organisations mimic similar organisations, by adopting policies and practices that appear successful even if the context and the initial conditions are not the same. According to DiMaggio and Powell convergence can occur through normative pressures brought about by professions or epistemic communities (as norms developed by groups enter into organisations);through miming (models can be diffused through employee migration or by international organisations);or coercion (conformity with EU requirements and other international obligations).

Figure 1: RIA adoption by OECD countries

Source: Bounds (OECD 2011)

But like NPM, the convergence in regulatory management is more apparent that real –beneath the common rhetoric of “smarter better”regulation or “better regulation, less regulation”,[41] the different country contexts have resulted in a wide variety in the practice of regulation. Radaelli uses nine country case studies (plus the EU) to explore diffusion of RIAs. He concludes that “RIA appears to be a typical solution in search of its problem”[42] and the spread of RIAs is a case of “diffusion without convergence”.[43]

A coherent explanation of policy convergence would also need to explain why the availability of new regulatory management instruments, such as RIA, would result in their adoption and utilisation. So it is to the demand for regulatory management regimes that this section now turns.

The spread of tools such as RIAs, and regulatory management regimes more generally, is not always perceived in the literature as a neutral technical development. Some political economy explanations suggest that the emergence of regulatory management is part of the global diffusion of “regulatory capitalism”. Levi-Faur suggests that as the deregulation agenda ran out of steam, the emphasis shifted from regulatory relief to smarter regulation.[44] Jane Kelsey makes a similar argument in the New Zealand context that development by the ACT party of the Regulatory Responsibility Bill is a continuation of the neo-liberal agenda of deregulation.[45]

An alternative political economy argument is about differences in access to justice by ordinary citizens as opposed to business interests. An argument can be developed about the growth in volume and complexity of regulation in part associated with the growth in the extent of the regulatory state. Unless offset by a regulatory management regime, access to justice becomes restricted to corporates who face more concentrated impacts and have deeper pockets than ordinary citizens.[46]

Both these lines of argument suggest that the increased demand for regulatory management came from different points in the political spectrum, whether neo-liberals attempting to re-energise the regulatory reform agenda once re-regulation ran out of steam, or liberals (in the American sense) concerned with access to justice. This also brings out the point that the regulatory management, like regulation itself, is not a neutral technical improvement device that operates in a political vacuum. In the case of the United States, RIAs were used “as an instrument to pursue the regulatory paradigm of the President.... RIA is introduced to foster deregulation ... [under Reagan and Bush] and shift policy towards a pro-regulation stance as shown by the Clinton and, perhaps, Obama administrations”.[47] Similarly, Bagley and Revesz[48] argue that the OIRA has placed greater emphasis on cost reduction and regulatory relief than on regulatory co-ordination.

A final meta political economy argument is that the spread of regulatory management regimes reflects similar political dynamics to the regulatory phenomenon that it seeks to manage. That is, regulations are often adopted because of an imperative for political decision makers to be “seen to be doing something”. In New Zealand’s case the Companies Special Investigation Act 1989 introduced after the 1987 share-market crash provides a good example. This legislation did not arise from a quality policy process in which the best intervention was selected to address a clear market failure, based on dispassionate calculus about the balance between social cost and social benefits.[49]

Similarly the adoption of regulatory management regimes may occur because of a political imperative to “do something”rather than based on a dispassionate rational calculation that the benefits of a regulatory management regime exceed the costs. Radaelli and De Francesco comment:[50]

For a politician, adopting a general provision on how regulatory proposals should be empirically assessed has low cost and high political benefits ... there is an incentive to opt for symbolic adoption.

This is consistent with the lack of impact from introducing RIAs, for example. The OECD, based on a range of country reviews, have concluded the “relative lack of integration of RIA in the policy process, assessments are done too late, consultation is often not robust …oversight needs more teeth ... Late timing of impact assessments is a widespread issue”.[51] What this highlights is that regulatory management, like any other public policy arguments for intervention, must be supported by an adequate causal theory of how the policy will achieve the desired behavioural change. It is to these public policy arguments that the chapter now turns.

7.7 Is there a public policy rationale for regulatory management?

Regulatory management regimes attempt to “regulate the regulation makers”by requiring an articulation of the policy problem that the regulatory intervention is designed to solve and what the alternatives are. A review of the regulatory management system, like this one, should impose the same discipline. This chapter now considers normative questions concerning the public policy rationale for the introduction of a regulatory management regime. Somewhat surprisingly, the review conducted for this chapter has not identified a rigorous, concise and compelling public policy case for special measures required as part of a regulatory management regime. This is ironic, as regulatory management requires regulators to articulate the case for public policy actions requiring regulations. Without a clear exposition of the cause or causes of poor quality regulation, it is difficult to prescribe an intervention or package of interventions that will be focused on the source of the problem.[52]

The Regulatory Impact Statement the Treasury provided to accompany the Regulatory Standards Bill currently before the House provides a long list of “symptoms”. The statement indicates that “informed domestic opinion consistently suggests that legislation [and regulation] could be much better than it is”[53] and the diagram (reproduced in Annex 1) highlights the limits of existing arrangements which suggests that the quality problem has a number of dimensions. This leaves unclear whether quality issues predominantly arise in the design phase relating to policy analysis, in legal drafting, or in the consideration by the legislature. The scope of the RIS does not extend to how the law and regulations are implemented. It also does not address whether the same problems occur with primary legislation, or secondary and tertiary regulation.

Paragraph [7.5] discussed how the main thrust of the regulatory management regime in New Zealand and in Australasia generally appears to be an attempt to “regulate the regulation makers”by strengthening the “big”policy design phase at the start of the regulatory process. The implicit assumption behind regulatory management systems appears to be that problems experienced in the quality of policy analysis are undermining the effectiveness of the design of regulations. There are a number of normative arguments for a regulatory management regime to improve regulatory design that are explored below. These include information quality, instrument selection, the growth of regulation, special features and the quality of policy advice. The different arguments point to different features required in a high-quality regulatory management regime.

7.7.1 Quality of information

The first normative argument for a regulatory management regime starts from the quality of the advice and information available to decision makers in the executive and the legislature. There are two potential arguments at the policy design phase and the legal drafting phase. The former is addressed here;the latter under the discussion of the quality of regulatory organisations (see [7.8.2]). The Treasury, in its 1987 briefing to the incoming government, asserted “proponents of a particular form of regulation often discount the difficulties that a regulatory solution to a particular problem will encounter”.[54] It is suggested that these arise from difficulties in obtaining information on costs, incentive alignment problems, and the insufficient adaptation of regulatory rules of thumb to complex situations.

In economic theory, quality of information problems are generally best met by direct information provision through disclosure and labelling or indirect provision through registration and licensing. The RIS regime, when it was first introduced, placed great emphasis on providing Ministers with a short (2-3-page) problem definition and assessment of options and consequences so that Ministers could make more informed decisions.

While in movies like Field of Dreams the promise “if you build it they will come”is realised, in the real world things don't work out the same way. In practice, in a range of endeavours, experience suggests that increased supply of information does not necessarily result in increased demand and use. Experience with evaluations shows that they are commissioned and then the results are routinely and systematically ignored.[55] Similarly, a recent survey of the use of performance information found that in every jurisdiction for which studies were available, legislatures have largely ignored the new information.[56] There was little use by the executive in budget making either.[57]

A range of empirical studies found a relatively modest impact of RIS regimes on the regulatory outcomes with studies in a range of jurisdictions repeatedly finding poor quality RIAs[58] and non-compliance with RIA procedures.[59] Like the other studies, the RIA studies are consistent with the proposition that adoption and incorporation of regulatory management regimes do not necessarily result in use by decision makers.

The experience with RISs in particular, and other NPM developments such as performance information more generally, all suggest that supply of better information and requiring it be incorporated in governmental processes does nothing to ensure that this information will actually be used. Moreover, if the information is used, it may be used indirectly or symbolically to legitimate an existing decision rather than directly used to make a better informed decision.

7.7.2 Intervention selection

The second public policy argument for a regulatory management regime relates to intervention selection. The introductory section discussed how, in addition to regulation, the government has a range of other policy interventions such as subsidies, specific taxes, spending on transfers and public ownership. The New Zealand Treasury has argued “the Government process itself tends to bias intervention towards the use of regulations rather than other policy instruments”.[60] Thus the argument runs;while the costs of subsidies and transfers are appropriated annually and subject to review by the executive and the legislature, the ongoing costs of regulation are not scrutinised in the same way.

One could argue that the costs of tax subsidies are not disclosed and the costs of transfers are significantly understated –the full cost of taxation imposed deadweight costs over and above revenue raised that have been variously estimated at between 32-48 per cent (using partial equilibrium models) and 15-50 per cent (using general equilibrium models) of the taxation.[61] But leaving aside the quibble about accounting versus economic costs of taxation and spending, the argument has some merits.

Around $70 million is spent each year on auditing public spending to ensure that it was spent lawfully and to explore whether it was effective. The fiscal responsibility provisions in the Public Finance Act (“PFA”) have also increased the focus and transparency on the future implications of current fiscal decisions. By contrast, the regulatory scanning and planning process has revealed that public agencies in New Zealand struggle to even identify the secondary and tertiary regulations they are directly responsible for as the administering department. No corresponding resource and effort is devoted to managing and reviewing the stock of legislation and regulations to see whether it is operating as expected or achieving the intended purposes. As George Tanner, former Chief Parliamentary Counsel, observed,[62] the lack of any systematic process for post-enactment scrutiny means that routine maintenance of some very major pieces of legislation rarely happens. We paint our houses and service our cars, but we don’t look after our laws in the same way”.

A related argument on bias in intervention selection in favour of regulation, is the lack of transparency. Compared to explicit expenditure, a bias in favour of regulation exists because of the lack of transparency in the costs of regulation and who bears those costs. This is an extension of the public choice notion that fiscal illusion, which arises with complex tax structure, makes judging the tax burden more difficult.[63] The OECD develops a multifaceted argument for regulatory management based on the need for selection of the best policy instrument.[64]

The bias in intervention selection argument leads to consideration of the doctrine of “takings”discussed elsewhere in this project.[66] Suffice it to say here that the “takings”doctrine applies to only a limited range of government regulatory interventions.

Another potential solution to the problem of intervention selection bias in favour of regulation is the introduction of a regulatory budget. To date no jurisdiction has successfully introduced a regulatory budget, although their introduction has been announced in at least two Budgets –by the Reagan Administration in the mid-1980s and by the Blair Government in the 2008 Budget. The notion of regulatory budgets, while appealing in principle, raises a number of practical issues such as the absence of an agreed unit of account to measure cost and the absence of effective measurement of costs incurred that can be compared with budget.

The intervention selection argument leads to unanswered questions about why the regulatory management regimes in New Zealand apply to expenditure proposals such as the family assistance in the 2011 Budget (Working for Families).

The main thrust of the regulatory management regimes reviewed in the last section has been to strengthen the policy design and development phase of the regulatory process through the use of RIAs in particular. Strengthening the policy phase would seem an appropriate response to problems of information asymmetry between executive decision makers and bureaucrats. Other than the provision and disclosure of more information, it is unclear how strengthening the policy phase changes the “instrument selection bias”that decision makers face that encourages the use of regulatory instruments over other interventions with a direct fiscal impact.

7.7.3 Special character of regulation

The third public policy argument for a regulatory management regime is that regulation has special characteristics that set it apart from other government interventions and, therefore, regulation warrants special measures. With specific regulations the state acts as coercer in reducing the set of choices open to business and citizens by precluding certain activities or requiring others. By contrast with fiscal interventions the state acts more as an enabler (although still underpinned by the coercive power to tax);[67] expanding the set of choices open to citizens and business by changing the income constraint or the relative prices they face in making those choices. Thus there is an argument that the special character of regulation poses a risk to liberty.

This is a strong prima face argument but does little to explain many of the common features of regulatory management regimes. It would suggest that general empowering regulation which provides for the broad framework of the rule of law, such as companies legislation, should be exempt so that the focus of scrutiny would be on regulations that impose specific prohibitions and privileges.[68] It would argue, for example, for bill of rights-type vetting of specific regulations on the grounds of protection of liberty. This does not seem consistent with the imposition of cost benefit analysis particularly associated with the Office of Information and Regulatory Affairs as part of the regulatory management in the United States and proposed for New Zealand under the Regulatory Standards Bill.

7.7.4 The quality of policy advice

An alternative line of argument can be developed suggesting that there is a more general problem with the quality of policy advice received by Ministers.[69] This is an argument for improving the quality of policy advice generally rather than for special measures applying to regulation. Indeed, in the United Kingdom context Hallsworth and Rutter suggest that specific regulatory assessments such as the RIA should be replaced by more general “policy assessments”.[70] Concerns with the quality of policy advice, however, can be used as a secondary argument to buttress the case for special measures to improve regulatory design, based on primary concerns such as intervention selection bias, regulatory growth and the special characteristics of regulation.

7.7.5 The growth of regulation

The final possible public policy argument for a regulatory management regime is that the volume and burden of regulation has grown rapidly and therefore regulation warrants special measures to get the growth under control. Whereas the previous argument was predominantly about the quality of regulation, this argument concerns the quantity of regulation.

On the face of it there is an issue with the growth of the quantity of regulation. Measurement of this trend is difficult, as unlike other government interventions, such as spending there is no unit of account that enables a simple comparison of regulation across time or across countries. Nonetheless, most surveys of the field conclude that the reduction in economic regulation associated with product market liberalisation was exception to the rule. Over the last 20 years, regulation appears to have increased in OECD countries. The steady growth recorded in measures such as the number of statutes, the number of regulatory agencies or length of regulatory rule books all suggest that the much vaunted economic deregulation has been more than offset by the growth in social and environmental regulation. In a provocative summary, John Braithwaite concludes that the triumph of neo-liberalism is a “fairytale”, deregulation is a “myth”and instead we are witnessing the emergence of “regulatory capitalism”.[71]

Regulation as Shleifer observes “is ubiquitous around the world yet standard economic theories predict it should be rather uncommon”. [72] This has led a number of economists to the view that regulation is driven not by efficiency and welfare enhancing concerns, but by politics and rent seeking. As argued above at [7.3], however, rent seeking, like the public interest theory of regulation it replaced, does not provide an adequate basis for a general theory of regulation. Shleifer in his article on efficient regulation goes on to argue “ubiquity of regulation is explained not so much by the failure of markets, or by asymmetric information as by the failure of courts to solve contract and tort disputes cheaply predictably and impartially”.[73]

Regardless of the merits of Shleifer’s argument, the growth in the quantity of regulation does not seem to explain the regulatory management regime in New Zealand which is focussed on improving the quality of regulation.

This section of the chapter poses an apparent paradox, because it is not clear what causes the regulatory problem that regulatory management regimes in New Zealand (and elsewhere) aim to address. In order to apply the logic of regulatory management to the design of regulatory management itself, the next section will explore whether there are other issues relating to the special characteristics of regulation that need to be considered.

7.8 What are the issues that regulatory management policy in New Zealand needs to address?

At [7.5] we identified the main focus of regulatory management regimes in all Australasian jurisdictions which appear to “regulate the regulation makers”by strengthening the “big”policy design phase at the start of the regulatory process. This part will explore the implication for both regulatory design (“regulating regulation developers”) and regulatory administration and enforcement (“regulating the wielders of regulatory power”) of complexity and uncertainty, quality of regulatory organisations, heterogeneity of regulatees and enforcement.

7.8.1 Complexity and uncertainty

Complexity and uncertainty affect both regulatory design and regulatory administration and enforcement.

The regulatory failure causing leaky buildings is a case study discussed elsewhere in this project,[74] but it provides a good example of the complexities involved.[75] Leaky buildings were not due to a single simple cause like technology failure, insufficiently sophisticated standard setting or because of bad builders. Rather, they were the outcome of a complex interaction of new technologies, new regulatory standards, installation practices (the use of sealants), lack of awareness of the 10cm clearances, new products (untreated timber), a lack of owner maintenance and so forth.[76] Layton suggests that the RMA also contributed to the problem, by raising land prices which encouraged the construction of houses without eaves to get the most out of each parcel of land.[77]

The regulatory failure arising from leaky buildings nicely illustrates the problem of complexity in Figure 2 below. On the left-hand side of Figure 2, cause and effect are only apparent in retrospect and solutions are unknown in advance. Responding to a complex problem requires learning the way forward. This requires a critical mindset that is continuously searching for anomalies and observations that don't fit.[78] This process of critical reflection can enable some complex problems to be transformed into “knowable problems”. A knowable problem occurs where cause and effect are identifiable, but are separated in time and space which means that patterns are difficult to detect other than by experts.

If the policy domain is characterised by uncertainty and complexity, this has implications both for regulatory design and regulatory administration and enforcement.

Dealing with design first, in the face of uncertainty and complexity there is no evidence-based gold standard or best practice. Majone observes “the key concept in decision making under uncertainty is not optimisation but consistency. If no generally accepted unique solution exists then decision-making procedure acquires special significance. This is the basic insight on which the classical theories of judicial and legislative procedures are based;the reason why procedures play such an important legitimating function in the decisions of courts and legislatures”.[79]

Figure 2: Kurtz and Snowden’s sense-making framework[80]

Majone observes:[81]

[B]y definition, uncertainty is pervasive in risk regulation. What is less well understood, however, is that in many cases scientific uncertainty cannot be significantly reduced. In controversies over the analysis and management of risk, the issues over which the experts disagree most vehemently are those that are, in Alvin Weinberg’s terminology, trans-scientific rather than strictly scientific or technical. Trans-scientific issues are questions of fact that can be stated in the language of science but are, in principle or in practice, unanswerable by science.

In the face of uncertainty and complexity, regulatory effectiveness crucially depends on quality of the implementation and enforcement phases. Legitimacy is brought to centre stage as the degree of compliance depends in a sense on the consent of the governed. The effectiveness in achieving the results desired by regulations are co-produced by the regulatees and the regulators.[82]

Eppel, Turner and Wolf have summarised a number of conditions facing implementation managers that are applicable to regulation when complexity and uncertainty are important. Among those conditions, they note:[83]

• regulatory outcomes depend upon influencing independent actors which cannot be directly controlled;
• knowledge is distributed and not all knowledge is known at the point of decision;and
• in order to function actors simplify situations but the things that are overlooked may grow in relevance over time.

In domains where the exact cause of the problem and the solution are not known in advance, different ways of working are required. This necessitates moving beyond the stylised cycle of big policy design, detailed design, implementation and review shown in Figure 3 below. Where the problem is known but the solution is not, techniques are required that involve learning the way forward. This “learn to build”[84] approach developed in the private sector works iteratively with real user experience with quick response times and measurement of key characteristics. It allows for “fast failure”by promptly reversing “bad”change and reinforcing “good”change.[85]

In domains characterised by uncertainty and complexity, one can question the usefulness of tools to screen and scan new regulations such as cost benefit and cost effectiveness analysis. This is because it is not possible to divine in advance the “one best way”to address a policy problem. It suggests that in a world where the impact of regulations will evolve in ways that are impossible to predict in advance, more is to be gained by more effectively learning the way forward by monitoring and managing the stock of existing regulations rather than screening the flow of new regulations. To the extent the screening tools act as a means to encourage exploration of alternative paths and strategies, or encouraging monitoring and evaluation to be put in place, then they can play a useful role as a part of ex ante screening. To the extent they encourage a focus on the best ex ante solution and lead to a minimisation of what is unknown or unknowable, they may in fact be counter-productive.

7.8.2 Organisational quality

The problems of uncertainty and complexity are accentuated by consideration of organisational quality. Majone observes that “a serious problem of EU regulation in general, and of risk regulation in particular, is the mismatch between the growing complexity of the tasks and the inadequacy of the existing regulatory institutions”.[86]

Just as the EU struggles in terms of the quality of its regulatory organisations, so does a small country such as New Zealand.

The quality of regulatory organisations becomes increasingly important as the complexity of regulation continues to increase. Complexity increases both from rising expectations of citizens for the government to “do something”and from pressures from the internationalisation of policy associated with globalisation, policy harmonisation and the implementation of international agreements.[87]

In the face of the internationalisation of government and New Zealand’s size and capability limitations, as a country we face regulatory choices analogous to the “make”or “buy”decision. New Zealand may more effectively be able to exercise jurisdiction through working with others to develop regulation rather than going it alone and attempting to develop regulation specifically for New Zealand. Great working together across national boundaries can occur through integration at different levels: policy design, administration and enforcement, and adjudication.[88]

Looking at the future of the New Zealand state, Gill and others suggest that “[s]erious questions have been raised about New Zealand’s ability to sustain a full range of credible public institutions and the need for more public regulatory bodies such as the proposed Trans-Tasman Therapeutics Agency”[89] that will be transnational rather than national regulatory bodies.

Even if the “big”policy problems can be adequately managed, there are “little”policy problems associated with legislative drafting. In the legislation drafting phase there are innumerable technical and legal decisions to give effect to the policy design. As an example, New Zealand enacted a Crown Entities Act to introduce a unified governance and accountability regime for arms’-length public entities, in part because of the problems caused by inconsistent and sometimes apparently random governance provisions applying to different entities.[90] While the quality of legislative drafting per se may not be an issue,[91] there is an issue of the lack of “little”policy legal analysis to underpin the drafting instructions. This will be explored further in the next phase of the project.

7.8.3 Heterogeneity of regulatees

To date the discussion has focussed almost solely on problems of regulatory policy design without much consideration of the impact on citizens and businesses who are expected to comply (the regulatees)[92] or the bureaucrats supposedly tasked with ensuring compliance (the regulators).[93] Turning first to regulatees, one of the central ideas in Ayres and Braithwaite’s[94] work on responsive regulation is that regulatory design and implementation should be responsive to the motivations of the regulated. These motivations are not uniform. The government is only one of the actors, as the outcomes achieved by regulation do not just depend upon the quality of the regulatory design. Regulatory outcomes also depend upon the nature and extent of voluntary compliance by citizens and businesses, and the way the regulations are implemented. The regulated consist of a heterogeneous mix of rational, incompetent and virtuous actors. Each will require different approaches that place greater emphasis on deterrence, incapacity and restorative justice respectively. Ayres and Braithwaite suggest an enforcement pyramid with the intensity of enforcement escalating if there is non-compliance.[95]

The heterogeneity in the population of the regulated, like the discussion of uncertainty and complexity above, highlights the limits of ex ante scanning of regulatory design and the potential role for ongoing monitoring of the stock of regulation. This monitoring needs to review whether the expected outcomes from the regulation are being achieved, as well as whether unexpected results are emerging as enforcement of the regulations evolves.

7.8.4 Administration and enforcement by regulators

There is a large, and not very coherent, literature on policy implementation starting with Jeffrey L Pressman and Aaron B Wildavsky. Their classic 1973 book Implementation: Great Expectations;How Great Expectations in Washington are Dashed in Oakland[96] (read Wellington for Washington and Auckland for Oakland) started an extensive applied literature that highlighted bureaucratic pathologies and ambiguities as “street level”[97] bureaucrats struggled to make sense of a world in which they were charged with applying multiple and often conflicting rules in complicated contexts.

One strand of the implementation literature starts from the view that implementation generally, and regulatory administration and enforcement in particular, can be framed as a top down rational planned activity. In this approach policy objectives and decisions can be cascaded down through the choice of instrument into detailed plans and standard operating procedures and risks can be anticipated and managed. Figure 3 shows a stylised policy cycle which corresponds to the world of the known and the knowable in Figure 2 above.

Figure 3: Stylised regulatory life cycle[98]

The implication for regulatory management of this top down approach is that effort should be invested into identifying and managing ex ante the “risks”that arise in the implementation of regulation. Risks can arise in policy design such as when policies change behaviour in unexpected ways.[99] Risks can arise from changes in the external environment which work against the operation of the policy. Risks also arise from execution including:

• the selection of the appropriate enforcement strategy and style of enforcement;
• co-ordination problems between layers of public and private agencies as responsibility for the policy package is divided into secondary and tertiary rule-making bodies, outputs from public agencies and contracts with private suppliers;
• slippage within agencies between the policy objective and the translation of that objective into standard operating practices;
• slippage within agencies between standard operating procedures and the practices of front line operators;
• resource constraints as implementation tasks are fitted to the available budget;
• organisational inertia and risk aversion which resists role redefinition;and
• organisational drift as practices are refined over time.

The implication for the regulatory management regime of the top down approach is to try harder by applying “one more heave.”This suggests strengthening the processes in the regulatory design phase so that implementation risks are proactively anticipated and managed so that the regulatory enforcement actually achieves the policy objectives.

The alternative to the top down rational policy model starts from the bottom up where implementation is an emergent process of experimentation and learning –the world of complexity and chaos in Figure 2 above. This is a world of learning the way forward in which policy is a work in progress which is revised and amended as one goes along. In this frame regulatory implementation is viewed as a process of policy learning and experimentation where the policy instruments used and indeed sometimes the goals themselves morph and change in response to emergent experience. Mumford describes this way of working as “thinking ahead”and “thinking along the way”. “Thinking ahead”means being aware of the many things that could go wrong based on an understanding of the strengths and weaknesses of regulatory approaches. “Thinking along the way”means critically examining issues that emerge from practice, and being prepared to adjust the regime as you go.[100]

The learning and experimentation approach has significant implications for the regulatory management regime. It suggests that the focus of the regulatory management regime needs to shift away from the design of regulatory interventions and focus more on the design of regulatory institutions –their role, mandate, power, capabilities, resources, ability to learn from experience and so on It also suggests the emphasis needs to shift from reviewing the flow of new regulations to more effectively monitoring and managing the stock of existing regulations.

There are a number of challenges from the authorising environment to working in ways consistent with a learning and experimentation approach. The first is the political imperative driven by Ladley’s law to announce regulatory reforms as solutions to problems rather than policy experiments. As a result, emergent problems are often seen more as threats to credibility of the policy than as opportunities to learn by adjusting the policy. Another is the threat of administrative law review with the increased activism of the courts.[101] All these forces drive an ossification of rule making as agencies become “reluctant to issue new rules, revisit old ones and experiment with temporary rules”.[102]

When regulatory agencies or staff within those agencies do not behave in the way expected, the standard response is the “retreat into rules”.[104] This defensive response to replace discretion with rules may reduce the regulation’s effectiveness by discouraging innovation and experimentation and removing the ability to tailor responses to the context facing particular stakeholders. The retreat into rules generates a culture of “doing things right”by following standard operating procedures rather than using discretion to “do the right thing”to achieve the goal of the policy.

The retreat into rules is also observed applying to the regulatees as well as the regulators. Regulatory compliance generates conformance with the rules rather than a focus on achieving the outcome sought. The literature is rife with examples of disasters that occurred (the collapse of the Piper Alfa platform,[105] friendly fire shooting down Blackhawk helicopters in Iraq)[106] because of goal displacement. The goal becomes compliance and gets focused on following detailed rules rather than the overall outcome of greater safety. Disasters happen not because personnel didn’t follow the rules but because people followed rules to the letter.

Sabatier suggests six necessary and sufficient conditions for the effective implementation of policy objectives including: clear and consistent objectives, adequate causal theory of how the policy will achieve change, administration and implementation structures that can enhance compliance, committed and skilful implementation, support from interest groups and no adverse changes in the authorising environment.[107] These are stringent conditions indeed and are unlikely to be met in many policy domains such as those characterised by chaos or complexity in Figure 2 above. A number of these conditions are lacking from the current design of the regulatory management regime including clarity of objectives and an adequate causal theory of how the policy will achieve change.

7.8.5 Theories of governance

Looking at different OECD countries, Radaelli and De Francesco identify three different logics for the adoption of RIAs that link theories of regulation with administrative procedures. The first logic delegation –explains the rise of RIAs in the United States as an attempt by the President to achieve greater control over rule making by executive agencies in order to pursue a de-regulatory agenda. The second logic –democratic governance –argues that by making administrative rule making procedures more transparent, it will become more open to diffuse interests and more accountable to citizens. The third logic –based on rational policy making –is that the more systemic use of economic analysis will improve the quality of policy decision making.

Traces of all three can be found in OECD documents and in official thinking in New Zealand, although the rational policy making logic is the dominant rationale used for the introduction of the regulatory management system. For example the OECD argues that it is important to ensure that RIA “is undertaken at the inception of policy proposals, when there is an opportunity and interest in identifying the optimal approach and alternatives to regulation can be given serious consideration”.[108] The logic is a multifaceted one buttressed by appeals to improved transparency and improved control of delegation problems as well.[109]

Implicit in the different logics are different theories of how the world works, the nature of bureaucracy and the nature of politics. For example, in the delegation model, politics is framed as conflict for control between the executive and legislative branches, while in the democratic governance logic it is in the conflict between elites. The next section –the conclusion –raises the often un-surfaced assumptions on the implicit theory of how the world works that underpin policy positions on regulatory management.

7.9 Conclusion

This chapter has briefly compared the regulatory management regime in New Zealand with those of the Australian Federal Government and the state governments before outlining the positive explanations and normative policy rationales for these regimes. It has shown how regulatory management has expanded rapidly across OECD and increasingly non-OECD countries. This diffusion is in turn part of practitioner-driven attempts to achieve public sector reform –which academics label new public management (“NPM”). Regulatory management attempts to “regulate the regulation makers”by requiring well-designed regulatory policy responses to clearly articulated problems. The chapter explored the apparent paradox that it is not clear what causes the problem of regulatory quality that a regulatory management regime aims to solve and how the regime will achieve behaviour change.

The policy implications are the focus of Stages 2 and 3 of the project overall. New Zealand, along with other OECD countries, has put considerable effort into strengthening the front end of the regulatory management system involving the “big”policy design of new regulation. The chapter suggests that there are important roles for consultation (to identify trip wires) and for development of more sophisticated “little”policy frameworks. The analysis in this chapter suggests the problem of regulatory quality may be better addressed by focusing less on cost benefit analysis and more on regulatory effectiveness in achieving behavioural change. Based on the notion of regulation as an emergent problem of policy learning and experimentation, it argues for less emphasis on review of policy design of regulation interventions and greater emphasis on the design of regulatory organisations and their role, mandate, power, capabilities and resources.

This final section briefly draws out the some issues from the key strands of the argument developed in this chapter that are relevant to questions of regulatory reform addressed in other chapters in this volume.

7.9.1 Issue 1 –Regulation as policy learning and experimentation

An often un-surfaced assumption underpinning policy debates is the implicit theory of how causation operates in the real world.In traditional policy and service design, it is often assumed that the world is operating in the known and knowable quadrants (on the right-hand side of Figure 2 above) where problem and solution are known in advance so the key challenge is to select the best practice intervention.

If, however, the policy domain is characterised by uncertainty and complexity, there is no gold standard or evidence base for best practice. The utility of tools like cost benefit and cost effectiveness analysis to screen the flow of new regulations is particularly questionable as a means of divining in advance the one best way to address a policy problem. Instead it emphasises the importance of more actively managing the stock of existing regulations, and building monitoring and evaluation and mechanisms to enable “fast failure”. Centralised systems and standardised approaches embodied in RIAs are not tolerant of decentralised experimentation and learning. Because of risk aversion, together with the retreat into rules, centralised processes are likely to have the opposite effect and discourage experimentation and learning.

7.9.2 Issue 2 –The gap between regulatory design and enforcement

The gap between policy developed in capitals like Washington or Wellington and what is implemented in Oakland or Auckland is well documented. If regulatory standards do not adequately take into account the resources required to allow the desired enforcement style to take place then a predictable gap will emerge. An assumption that is less often clearly surfaced is that regulation can usefully be framed as a top down rationally planned approach whereby policy objectives and decisions can be cascaded down through choice of instrument into detailed planned and standard operating procedures and risks can be anticipated and managed. The alternative model starts from the bottom up where implementation is an emergent process of experimentation and learning. Indeed the policy instruments used and indeed sometimes the goals themselves morph and change in response to emergent experience.

The bottom up approach has quite different implications for the regulatory management regime, as it suggest greater emphasis should be placed on more effectively monitoring and managing the stock of existing regulations. The discussion covered multiple sources of slippage between the objective of the design, the translation of that design into standard operating practices (“SOPs”) and between SOPs and what operators actually do. This raises questions about the design of regulatory institutions (not interventions) and their role, mandate, power, capabilities and resources.

7.9.3 Issue 3 –International government integration

This chapter discussed the diffusion of RIAs in particular and regulatory management approaches as examples of institutional isomorphism. In turn, the diffusion of regulatory management is just an example of the wider phenomenon of greater international government integration in the face of the internationalisation of policy. In the face of New Zealand’s size and capability limitations, we face regulatory choices analogous to the “make”or “buy”decision. Can New Zealand more effectively exercise jurisdiction by working with others to develop regulation or by going it alone and developing regulation regimes specifically for New Zealand?

7.9.4 Issue 4 –Ex ante prescription and ex post adaption

All regulation inevitably involves the exercise of discretion and judgement by regulatees and regulators. This is particularly true as the regulatory context evolves over time in ways not anticipated by the original regulatory policy design. How much should the regulatory regime attempt to prescribe in advance the rules that should apply and how much should the regulatory regime allow for discretion to adapt as circumstances change?

While the high level policy objectives from regulation can be clearly stated, sometimes the outcomes from regulation are not knowable in advance when the outcomes depend upon complex interactions of a range of actors. As there is often no gold standard or best practice, in these cases good practice involves learning the way forward through experimentation. How can regulatory regimes be designed that allow for experimentation and learning, where it is required?

7.9.5 Issue 5 –Rule of law principles

There are a number of principles that guide development of good regulation embodied in LAC Guidelines including consistency with legal principles and existing legislation, comprehensiveness, clarity, certainty, enforceability and so forth.[110] The New Zealand Treasury has developed a set of attributes of best practice regulation which include proportionality, certainty, flexibility, durability, adaptability, capable regulators and growth being focussed.[111] How are the inevitable trade-offs such as those between certainty and durability and adaptability to be resolved? The Regulatory Responsibility Bill proposes one solution –the threat of court review will improve upstream incentives to address these trade-offs in a serious way. Are there others?

7.9.6 Issue 6 –Better alignment between design of the regulatory management regime and the causes of poor quality regulation

Paragraph [7.7] of this chapter identified a number of possible normative arguments for the role of a regulatory management regime including information quality, instrument selection, and special characteristics of regulation. It went on to show how the different arguments point to quite different features required in a high quality regulatory management regime.

The implicit assumption behind regulatory management systems in most jurisdictions is that problems are experienced in the quality of policy analysis underpinning the design of regulations. Is the problem the detailed design at the micro level of the individual regulatory intervention, or at the more sectoral level in the interaction between the regulation and other interventions, or at the overall policy system? Did the problem with leaky buildings arise in part as a by-product from the RMA forcing up urban land values thus encouraging the construction of buildings without eaves?

Alternatively, would decision-making be improved by more careful and systematic ex ante policy analysis? While this may be the case, it risks introducing a new source of error –by focusing attention on the known risks, diverting attention away from the unknown.

Fundamentally the design of a regulatory management regime must address the problem of the value proposition. The evidence from studies on the effectiveness of RIAs provides limited overall positive support for regulatory management regimes’ability to reduce the cost per life saved or to improve the consistency of decision-making. New Zealand’s own experience with the use of RIS is salutary as they are “often produced too late so it becomes a compliance exercise”.[112]

Regulatory management attempts to “regulate the regulation makers”. This chapter argues regulatory management should itself be subject to the same tests as are imposed on other regulations. Is the regulatory management intervention itself a well designed response to a clearly articulated problem? This chapter is a small contribution to clarifying the value proposition to ensure that attempts to manage the regulatory power of the state do in fact achieve valid public policy purposes.

Annex 1: The Treasury Regulatory Impact Assessment for the Regulatory Standards Bill 2008

Annex 2: Mapping the New Zealand Regulatory Management System