Working paper

Encouraging Quality Regulation: Theories and Tools (WP 03/24)

Formats and related files


Achieving good regulatory outcomes normally requires high quality design, implementation and review of the regulatory regime. Major regulatory theories focus on concepts such as the public interest, the role of interest groups, and regulatory capture to explain why regulations come into existence. Regulatory design, however, exists at two levels. Downstream design involves creating a regime to give the appropriate incentives to firms and consumers. Upstream design seeks to incentivise regulators themselves to create and operate high quality regulatory regimes. This paper focuses on the latter.

The OECD has undertaken a major programme on regulatory governance to ensure quality in the design and implementation of regulations. Such measures are now widespread. New Zealand has gradually implemented these approaches including Regulatory Impact Analysis (RIA) in its decision-making processes. These measures are supported, and to some degree required, by increased interaction with Australian practices through institutions such as the Council of Australian Governments and the obligations of Trans-Tasman Mutual Recognition Arrangement. Achieving full integration of best practice that creates an environment for consistently delivering high quality regulation requires a broad and sustained focus on design, capability, incentives and follow-up.

New Zealand has attempted over the last decade to improve regulatory outcomes by focusing on the incentives on regulators. There is still scope for further improvement. Sustained progress on a number of mutually supporting initiatives, with continued reinforcement of the underlying messages and careful building of the necessary institutions and practices is required for continued improvement.


The views expressed in this Working Paper are those of the author(s) and do not necessarily reflect the views of the New Zealand Treasury. The paper is presented not as policy, but with a view to inform and stimulate wider debate.

1  Introduction#

This paper focuses on improving the quality of all forms of regulation imposed by government. It surveys basic concepts surrounding the regulatory activities of government, focusing on incentives faced by regulators, and reviews New Zealand practice to identify priorities for future actions to improve those incentives. The choice between different types of regulation, such as prescriptive versus performance-based or rate of return versus CPI – X, is not therefore addressed here.

Regulation is defined as the employment of legal instruments for the implementation of social-economic policy objectives and broken down into economic and social regulation (den Hertog, 1999). Economic regulation can be structural, through restrictions on entry and exit and qualifications, or conduct, such as price control, advertising restrictions and quality standards (den Hertog, 1999).

In the New Zealand context, regulation includes statutes (legislation), statutory regulations (Orders in Council) and quasi-regulation (a wide range of standards, guidelines etc issued by Ministers or departments under the authority of statute or statutory regulations). Quasi-regulation is subject to less scrutiny than the other forms of regulation.

The paper begins by discussing, and then briefly reviewing, basic taxonomies of the benefits and costs of regulation; major theories of regulation, the concept of “regulatory illusion”, the incentives faced by regulators and regulatory failure in Section 2. Section 0 examines how regulations are implemented, discussing institutions and instruments, comparing performance-based and prescriptive approaches to regulation and outlining the components of an integrated environment for ensuring high quality regulation. The paper then addresses past approaches to modifying the process of forming regulation within government overseas and in New Zealand in Section 4, before considering future paths to improve outcomes both domestically and trans-Tasman. Concluding remarks are made in Section 5.

The paper assumes that any changes to regulatory systems would have to operate within existing constitutional, political and judicial institutions. The aim therefore is to identify how, within existing constitutional constraints, to best ensure that whatever regulation is put in place in future achieves the maximum excess of benefits over costs. As a general survey, rather than an in-depth investigation, it also does not seek to identify any particular level or type of regulation to be desirable or not, or to identify any specific regulatory regimes as being priorities for reform.

2  Reasons for regulating#

This section reviews the seminal theories of regulation to understand the incentives faced by regulators, the costs and benefits of regulation, the concept of “fiscal illusion” and theories of why regulations are made.

2.1  Major theories of regulation#

Theories of regulation try to explain why regulation is adopted. The traditional public interest theory “regarded market failure as the motivating reason for the entry of regulation” with regulation correcting the inefficiency (Peltzman, 1989).

2.1.1  Public interest

Public interest theory suggests that government regulation is a response to public demands for government to rectify situations of market failure through imperfect competition, market disequilibria, missing markets (caused by hidden or asymmetric information, high transaction costs, externalities, public goods) or market outcomes that are undesirable for social reasons (den Hertog, 1999).

This of course assumes that (1) the market outcome represents a “failure” of some sort, and the market is not capable of fixing the problem itself, (2) that the government is capable of fixing that failure so that the optimal efficient outcome will be achieved (given constraints from institutions, technology and information) and (3) that the benefits of doing so will outweigh the additional costs created by the intervention (after taking into account administrative costs and any new allocative inefficiencies).

In summary, public interest theory can be said to assume that the regulatory regime will both aim for and achieve economic efficiency. Relaxing these assumptions leads to the concept of regulatory failure (see below), a concept that is not always considered in regulatory decisions. Public interest theory also fails to predict how the public interest is translated through political institutions into a decision, who will be regulated and who will receive the benefits or bear the costs, or the form of the regulation (den Hertog, 1999).

2.1.2  Economic theory and regulatory capture

This altruistic view of regulation was not seen as universally convincing and in the 1960s alternatives began to emerge. These included capture theory where regulations comes to serve the interests of those regulated, the economic theory or Chicago theory of regulation where it serves as a response to “interest group” demands, and public choice theory which focuses on rent seeking behaviour (Posner, 1974).

Regulatory capture occurs where, due to industry control of information, the effects of repeated interactions and career opportunities, the regulator comes to serve the interests of the regulated (Posner, 1974). This can be through direct subsidies, entry restrictions or tariffs, controls on substitutes, or price fixing (Stigler, 1971). Issues include why the industry cannot prevent the creation of the regulator in the first place, why regulation imposes burdens on industry in favour of others, or why costly regulation is accepted (although this could be to shield against more effective regulation) (den Hertog, 1999).

An extension of this approach, the economic theory of regulation or the Chicago theory of government, suggests that a regulatory regime may itself be “acquired by the industry and is designed and operated primarily for its benefit” (Stigler, 1971: p3). Regulation is sought through the supply of votes and resources to politicians with the cost reflecting factors such as the cost of the proposed regulation and the geographic concentration of benefits or costs.

Regulatory capture could occur, for example, where an agency was established to conduct occupational regulation for quality reasons and became captured by that same profession to achieve benefits for incumbents through entry restriction.[1] This is one example of regulation as a means of taxation of one group on behalf of others. Such regulation which can go as far as explicit reduction of some property rights in favour of rights held by others, sometimes referred to as government takings (Guerin, 2002b). This can also be described as “taxation by regulation” and used to explain regulated services provided below cost though “internal subsidies” from other profitable services (Posner, 1971:22). It can be argued that this is a deliberate choice by the state to ensure provision of the service without direct fiscal impact, effectively substituting a specific narrowly-based tax for funding from general taxation revenue with reduced scrutiny and less pressure to balance against competing uses of the funding. Such approaches can also reflect “a coalition of regulated firms and those of their customers who receive services below cost as a consequence of regulation” (Posner, 1971:47).

It can be argued that there is an optimum size for effectiveness of such a political coalition to seek gains through regulation, as beyond that size the interest of each member of the coalition becomes too small, and the losses to each opponent become too high (through information and organisation costs), and large coalitions are difficult to organise given the range of issues arising in elections (Peltzman, 1989). These costs can be argued to limit both the size of a coalition and its gains (Peltzman, 1976).

This could lead to a dichotomy where consumers lobby for regulation of monopolistic industries while firms lobby for regulation of competitive industries. Also the number of individuals or firms involved may influence the choice between private cartelisation and seeking regulation through political channels; ie, “it may be cheaper for large-number industries to obtain public regulation than to cartelize privately” (Posner, 1974, p346).

Such a view is consistent with a rent seeking interpretation of political behaviour, with rent seekers wasting resources to obtain regulatory rents, and politicians and bureaucrats capturing a share of the efficiency losses of regulation. It also supports a marginal approach to political allocation of regulatory benefits where more than one group receives those benefits (Peltzman, 1989). The economic theory of regulation still, however, does not address the political mechanism by which regulation is achieved, or how this mechanism itself influences the outcome (den Hertog, 1999).

2.1.3  Pressure groups and public choice

The theory of competition among pressure groups, with their success depending on factors such as their efficiency in producing pressure, the size of the groups involved, and the efficiency of the proposed tax or subsidy was outlined by Becker (1983). Rising deadweight costs of regulation constrain inefficient regulation as the marginal gains to the beneficiaries shrink relative to the losses to others. More efficient replacement regulation, or regulation which addressed costly market failure, could therefore gain support and regulators would tend to focus on potential regulation where the net benefit was greatest. Reductions in the benefits from an existing regulatory regime, due to growing deadweight costs or external factors such as rising costs or growing competition, can therefore prompt deregulation (Peltzman, 1989).

Public choice theory applies a utility maximising approach to political choices and invokes the concept of “politics as exchange”. The rules of such exchange are set through constitutional design, based on consent. Groups then face an incentive to undertake rent-seeking activity within those rules if they can gain at the expense of others, primarily through lobbying for policies that achieve concentrated benefits for themselves in exchange for costs diffused across others (van den Hauwe, 1999). Interest groups themselves can be concentrated geographically, by industry or even by a single issue.

This discussion illustrates that there are many ways in which groups can influence regulation and that regulation has redistributive effects, but does not explain why particular regulations emerge and why in some cases regulation shapes interest groups while in other cases the reverse occurs. Conversely there are also multiple arguments as to why deregulation occurs, including (1) shifts in influence of interest groups, (2) a group deciding it can perform better without regulation, (3) declining profits in the regulated sector, and (4) increasing deadweight costs of regulation (den Hertog, 1999). The predictive value of these theories is therefore limited.


  • [1]Within economic regulation, product quality constitutes an interesting case, with the argument that when consumers care about product quality this can often be dealt with through advertising, so that government regulation is demanded only when it is difficult for consumers to tell when the advertised standard is actually being met. In some cases, third–party certification is accepted (eg, an industry association) where government is unwilling to regulate (Holcombe and Holcombe, 1986).

2.2  Benefits and costs of regulation and fiscal illusion#

Ultimately, for every regulation there will be a net social impact and a range of private impacts, so that the regulation will change both the total welfare of society and the manner in which that welfare is distributed among individuals and groups. It is usually easier to identify the direction of the redistributive effects than to quantify the size of those effects or of the total impact, especially for social regulations.[2]

It is also important to distinguish (1) transfers between groups within society from (2) actual economic costs to society or deadweight losses, which include costs associated with the payment of the transfers.[3]

As far as costs of regulation are concerned, key issues are size and incidence. Beyond those factors, a number of classifications are used. These tend to focus on the initial incidence of the cost, as the ultimate incidence would require a general equilibrium analysis. Table 1 illustrates two common approaches.

Table 1 – Costs of regulation

An economic perspective (den Hertog, 1999)
  1. Costs of formulation and implementation
  2. Costs of maintenance
  3. Costs of compliance
  4. Deadweight costs (associated with distortions from 1-3)
A government perspective
  1. Fiscal (excluding administrative) costs
  2. Administrative costs
  3. Compliance costs
  4. Economic (resulting from 1-3) costs

Government tends to focus on fiscal costs (including administration), while businesses concentrate on those compliance costs which fall directly on them. Economic costs, or deadweight losses, tend to be the hardest to estimate and the least transparent.

Benefits from regulation can take many forms but these can be distilled down to an improvement in the welfare of an individual or group. Whether this occurs through reduced costs or increased income is not as important as whether the source of the gain is wholly or partially from a net increase in society’s welfare, rather than simply a transfer from another individual or group with the inevitable deadweight costs.

The concept of fiscal illusion”[4], is well established as a factor in public policy where undesirable expenditure is undertaken because the true costs are hidden; eg, because neither the revenue nor the expenditure have to be reflected in government accounts. More generally, decision-makers face incentives to undertake expenditure which has a net social cost, because the benefits are concentrated and transparent but the costs are diffused and opaque.

Regulatory illusion includes fiscal illusion, and can be used to explain some of the incentives faced by regulators. Many regulatory decisions do not involve actual revenue collection or expenditure by government, but still impose costs. Regulators tend to focus on the costs that directly affect them. For example Treasuries focus on fiscal costs, reflecting their budget management role and the transparency of expenditure, while departments concentrate on administration costs and implementing policy, and interest groups emphasise compliance costs.

A wider form of regulatory illusion is the view that regulation is necessary to correct market failure on the assumptions that (1) the market will not compensate for factors such as transactions costs, and (2) that government regulation is effective and efficient. In fact, whether these assumptions hold needs to be tested for each case.

There are two ways for regulators to deal with costs while progressing a proposal – to modify the proposal to improve the benefit/cost equation, or to shift the costs on to another party. The latter approach is easiest when shifting the costs reduces their transparency. Reduced transparency of costs also reduces barriers to approval of the underlying policy.


  • [2]Quantification requires estimation of demand and supply behaviour and a reliable counterfactual against which to measure the effects (Hahn and Hird, 1991).
  • [3]Economic costs and benefits arise from changes in static and dynamic efficiency. Static efficiency is productive efficiency (from improved use of resources) and allocative efficiency (from allocating resources to the most valuable use). Dynamic efficiency refers to changes over time.
  • [4]Puviani in 1903 in The Theory of Fiscal Illusion addressed fiscal illusion through the question “If the ruling group desires to minimize taxpayer resistance for any given level of revenues collected, how will it set out to organize the fiscal system?” Mechanisms suggested include distorting the link between the total cost of a programme and the individual share, charging taxes at favourable times such as when an inheritance has just been received, charging fees for nominal services such as licences associated with marriage, linking taxes to public opinion such as surcharges on business profits, scare tactics, or taxes where the incidence is unclear (Buchanan, 1987). Similar approaches can of course be taken in the design of regulation. It is also interesting to compare the hypothesised political goal of minimising taxpayer resistance with the economic goal of minimising the deadweight costs of taxation, or regulation. Where the two goals are aligned, so perceived costs equal real costs, presumably the outcome will be the socially optimal one, while otherwise it will be suboptimal.

2.3  Why regulations fail - incentives on regulators#

Regulatory failure can include a situation as wide as when Government intervention fails to achieve its objective, or more narrowly, a situation where the outcome after intervention is worse than it was expected to be without the intervention. The discussion here focuses on the latter type of failure. That failure can be due to fundamental flaws in regulatory policies, such as information gaps, or prescribed to the shortfalls of bureaucracy[5] (see Table 2).

Regulatory failure can arise from the same source as market failure; eg, a cross-border externality may not be reflected in national regulation, or regulation may be designed to capitalise on the externality at the expense of other countries (Neven, 1992). Some of these factors may need to be recognised and taken into account, such as lack of capability to regulate which can drive towards either no regulation or resort to an external regulatory framework. Others, such as flexibility, objectives and interdependence can be identified and addressed in regulatory design.

Table 2 – Regulatory failure
Categories Key Concepts Specific Issues
Capability Administrative, Technical and Informational
  • Information availability.
  • Resources (including funding but also availability of skilled staff).
Design Structural and jurisdictional problems
  • Unclear or mixed regulatory authority (eg, cross-border or cross-agency issues).
  • Contradictory or overlapping objectives.
  • Exclusion of some costs/benefits from analysis.
Regulatory Design problems
  • Lack of flexibility.
  • Inadequate participation by those affected by regulation.
  • Inadequate investment in education, monitoring and enforcement.
  • Affordability of regime in relation to benefits and scale of regulated activity.
  • Interdependence between elements of the regulatory structure.
Incentives “Public Choice” theory
  • Principal-agent (Ministers vs bureaucracy vs public) problems, eg, inefficiency.
  • Information asymmetries.
  • Costing problems (dispersed costs, lack of accounting for economic costs of regulation).
  • Changing political or bureaucratic priorities.
  • Political independence.
  • Regulatory capture (“special interests” etc).

Addressing incentives is the most complex aspect of avoiding or explaining regulatory failure. One approach is to use multi-industry regulators which can be more efficient, face reduced information asymmetries (information disclosure regimes can also help here) and be harder for any one industry to dominate.

Principal/agent theory suggests that agents can favour interest groups at the expense of others in exchange for benefits to themselves. The nature of regulation complicates the principal-agent problem “due to an inability to meaningfully measure performance” (Bailey, 2001). There may be a principal/agent problem among ministers, officials and the public in a Westminster style of government (see Table 3), as ministers fill both roles – that of principal with respect to departments and of agent with respect to the public.

Both ministers and officials tend to be held more accountable for failing to regulate than for regulating at excessive cost, as the former is more transparent and can be held up as the reason for any negative outcomes in the activity to be regulated. Excessive regulation, however, is much harder to detect and the costs will be dispersed among those who are regulated, or those to whom the costs can be passed on. This effect skews regulatory decisions and is particularly prominent when dealing with risks that involve a low probability but very high consequences; eg, air safety or biosecurity.

Table 3 – Incentives on regulators
“Players” Incentives Resources
  • Pressure to “do something”. Risk averse.
  • Lobbyists push for favourable regulations.
  • Ministers with finance roles can focus on fiscal rather than economic costs. Other ministers focus on portfolio interests.
  • Limited time for specific issues.
  • Small office staff limits ability to evaluate advice from officials.
  • Be seen to be taking action and achieve the political objective. Regulatory response becomes institutionalised.
  • Do not present “bad news”. Do not bear costs or see consequences of regulation.
  • Want more resources but investing in policy capability will not benefit current management.
  • Low tolerance of risk so avoid action where failure will be transparent.
  • Focus on fiscal not economic costs.
  • Keep interest groups happy.
  • Limited policy capability and empirical data.
  • Weak cost controls on regulations.
  • Conflicts of interest if funded by cost recovery from those regulated.
  • Limited understanding of economic impact of regulations.
Regulated Parties
  • Favour regimes which limit new entry; eg, quotas and prescriptive process rules.
  • Regulation provides protection against liability.
  • Significant financial interest in outcomes of regulation.
  • Long-term involvement.

One result, in practice, of these incentives is a public sector that is good at providing solutions which are low risk and have a minimal fiscal impact, but which impose high compliance costs on businesses or individuals. Both ministers and officials face strong incentives to “get something done” and weak incentives to “do it well” or take a longer term perspective (eg, develop improved policy capabilities). The popularisation of concepts such as market failure encourage this approach, while the risk of government or regulatory failure is much harder to explain or is seen as an excuse for inaction.

Regulators are also constrained by the extent to which an issue is the subject of public debate (in general low public awareness gives regulators more discretion although a polarised debate can give regulators complete freedom) or of political commitments.

A number of these causes of failure also arise outside government of course, but “the obvious difference is that failed government policies, until corrected, have the force of law, and affect the prosperity and success of all” (Lochner, 2000:838). This is why so much effort is put into improving regulation.


  • [5]However, focusing on small cases can be efficient, staff quality may be high given value of service to future career prospects, efficiency incentives on employees may be similar to the private sector, and there is political scrutiny (Posner, 1974). Agencies are also constrained by the inherent difficulty of much economic regulation and the difficulty of effective scrutiny by the legislature.

3  Methods of regulating#

3.1  Institutions and instruments in regulation#

The two key concepts in regulatory design are institutions and instruments. Regulatory institutions include formal governmental bodies at trans-national, national and local level and behavioural constraints such as constitutional principles and the regulatory quality regime (eg, regulatory impact statements). Regulatory instruments are the tools available for institutions to apply, and include treaties, primary (statute), secondary (regulations) and tertiary (notices and guidelines) legislation, national instruments for directing local government such as national policy statements or national environmental standards, and regulatory takings.

Upstream regulatory design examines how institutions and instruments create incentives for regulators to design an optimal regulatory structure. Downstream regulatory design considers how a regulatory structure creates incentives for individuals or firms to act in a manner that achieves a socially efficient outcome. Although the incentives involved in upstream and downstream design are similar, the tools available to influence them are quite distinct. The downstream issues are significantly more diverse (covering both economic, including competition law, and social regulation), than the upstream issues which are reasonably common across all sectors (including such broad concepts as subsidiarity and regulatory quality). Both provide opportunities for feedback, as a better upstream process should improve downstream outcomes, and downstream experience should provide useful information on designing upstream structures.

The institutional structure is something that can be designed so as to improve the regulatory outcomes.[6] The key purpose of regulatory quality regimes, is therefore in relation to upstream regulatory design; ie, to modify the institutional structure within which regulation is developed and applied to improve the quality of the decisions made at both stages.

3.2  Performance based or prescriptive regulation#

All regulation falls somewhere along a spectrum in terms of how much detail is specified and what is left to interpretation (eg, by users or the Courts) or dependent upon external factors (eg, an occupational licensing regime, market forces, another regulator). Prescriptive regulation and performance-based regulation fall within this spectrum.

Prescriptive regulation, in principle at least, defines how activities are to be undertaken (eg, what techniques or materials to use, what qualifications must be held, where the function may be performed). This approach clearly emphasises a known degree of risk mitigation over innovation or cost management.

Performance-based, or outcome, regulation puts more emphasis on specifying a performance standard for the desired outcome and does not deliberately constrain how compliance is to be achieved. A parallel in environmental regulation is with a requirement to use specific emission control technologies versus the use of market-based instruments to control pollution (Guerin, 2003). This approach is more consistent with innovation and efficiency than prescriptive regulation.

In both cases there is an inevitable dependence on external factors to make the regulation effective. Neither approach will succeed unless the necessary supporting structures are in place, such as quality assurance, and effective monitoring and enforcement.

Performance-based regulation acknowledges this explicitly, which allows for those external factors to be identified and assessed more easily. Prescription, on the other hand, may simply appear to be safe through being a “tried and true” approach which incorporates experience and known techniques and institutions. It may, however, deliver a more certain outcome, although at the possible expense of efficiency and foregone benefits from innovation.

3.3  The components of a regulatory environment#

This section considers how to create an environment that fosters high quality regulatory design, on the assumption that this is the approach most likely to deliver the outcomes that society wants. It does not attempt to define what will constitute high quality regulation or to isolate what the desirable regulatory outcomes might be. A focus on design, capability, incentives and follow-up is required to achieve full integration of best practice and an environment for consistently delivering high quality regulation. These components are addressed below.

Well-designed regulation has a depth of techniques and principles that (1) allows regulators to judge whether a particular proposal meets the appropriate quality standard and (2) covers all significant regulatory activity. Good design therefore requires a commitment to a quality standard that can be readily applied by both those performing the original work and those required to review the results. Those standards typically cover basic principles such as benefits exceeding costs, transparency, fairness, consultation, feasibility of compliance, and whether the regulation is administered at the appropriate level of government. Whatever the nature of the standard however, the way it is framed and the nature of the guidance material available to those applying it will be crucial to whether it is met in practice. Standards that are too open to interpretation, have too many loopholes or are not well communicated will fail.

A crucial issue is how to create an environment in which institutions and individuals involved in regulatory processes are mandated and encouraged to apply quality tests to regulation in practice. The central element is the incentives that are placed on ministers and officials to carry out these processes. It is, however, very difficult to alter those incentives because it is difficult to design outcomes objectively and ascertain responsibility effectively.

The tools most likely to be effective in this situation are therefore indirect. The first involves transparency in the design and implementation of regulation to maximise the probability and reputational effect of identifying problems. If decisions, and their consequences, are opaque then there is unlikely to be any real pressure for good performance. Transparency may not be sufficient to ensure quality, but it is probably necessary.

Another, but even more difficult to apply, tool is consistent and strong commitment from the highest political and administrative levels, to the desired outcome, sending a signal of expectations to both individuals and institutions regarding the type of behaviour that will be rewarded. This must then be coupled to effective performance management systems which are themselves much more easily prescribed than achieved in the public sector, but are also desirable for reasons other than regulatory quality.

The same point applies to capability, which in this context means that regulators have the resources to perform their required tasks and that those responsible for designing a regime and implementing it are aware of its likely impact and of the quality requirements it must meet. Good incentives are worthless without the ability to respond adequately. This involves significant issues of public sector funding and organisation. Government has a crucial role in training and education. This can both involve informing people of the requirements of the regulatory quality regime, and helping them use the tools available to meet those requirements.

An effective regulatory quality regime must be practical in terms of both compliance (see discussion above on design and capability) and follow-up. Follow-up involves creating the necessary institutions to monitor the performance of regulators against the quality regime and respond to success or failure. That response can take a number of forms which must reflect political and administrative realities; eg, a veto on progressing low quality regulation may be impractical in urgent situations or may clash with constitutional constraints. Other options include publicity around organisation performance, inclusion of quality comments in documentation within the regulation making process, and formal process reviews. Follow-up does not extend to actual evaluation of policy outcomes, since it is focused on the regulatory process itself, but would include assessing specific policy processes on whether or not appropriate evaluation procedures were built in.


  • [6]Theories based solely on preferences tend to fall short of explaining how those preferences are translated into outcomes, which requires institutional structures. Similarly, institution-focused theories tend not to explain where preferences come from or how institutions arise or are sustained. The discussion here is intended to be at a general level, considering how to shape the incentives on regulators to achieve the optimal outcome independent of the types of policies being considered or the preferences of the parties.

4  Regulatory governance internationally#

A number of countries, including New Zealand, have introduced measures to improve the quality of regulation. This section reviews the overseas experience, outlines New Zealand’s regime and discusses mechanisms for improving the design and implementation of regulation.

4.1  Organisation for Economic Co-operation and Development (OECD)#

4.1.1  Stages of regulatory reform

The OECD identifies three stages of regulatory reform which can run in parallel - deregulation, regulatory reform, regulatory quality management and regulatory policy (OECD, 1997a). Deregulation, which spread in popularity in the 1970s, was a reaction to a loss of faith in the public interest theory of regulation (see Section 2.1) and a perception that high regulatory costs were impeding innovation and growth. However, deregulation, and the concurrent push for privatisation of state run institutions, were in practice re-regulation as a more permissive regime typically requires more sophisticated regulatory institutions than does a prescriptive approach.

Regulatory reform therefore emerged with a focus on a combination of de-regulation, re-regulation and more effective regulation. This approach retained, however, a one-off or episodic approach which proved untenable given the ongoing creation of new regulation and the impossibility of ever achieving an optimum.

The next stage of reform involved regulatory quality management. This sees reform as a dynamic permanent process within government and evolved into regulatory policy that combines quality management of existing regulation with forward looking quality assurance. Strategies for improving regulatory quality therefore include building a regulatory management system, improving the quality of new regulations, and upgrading the quality of existing regulations. A regulatory management system involves driving regulatory reform policy from the highest political levels, establishing explicit quality standards and decision criteria, building capacities for regulatory management and providing oversight of implementation. Improving the quality of new regulations requires adopting regulatory impact analysis, a comprehensive policy on public consultation, systematic consideration of alternatives to regulation and improved regulatory co-ordination.

Regulatory management in turn has evolved into the concept of regulatory governance, recognising the importance of core governance concepts such as “transparency, accountability, efficiency, adaptability and coherence” to high quality regulation (OECD, 2001:5)

The OECD is now focusing on (1) sub-national, supra-national and inter-governmental regulation making, (2) improving controls over “grey” and “quasi” regulation, (3) building the institutions of regulatory reform and (4) promoting understanding of the economic importance of regulation (OECD, 2001).

4.1.2  Mechanisms and tools

In 2001 the OCED reviewed the progress of its regulatory policy agenda (OECD, 2001). The highest level component involves regulatory policies such standardised appraisal systems and guiding principles of good regulation. Practical application involves tools to use in assessing regulatory proposals such as Regulatory Impact Analysis (RIA) and mechanisms for consultation and accountability. There are strong similarities and increasingly rigorous application in the area of regulatory tools across jurisdictions.

The final component is the establishment and embedding of institutions which can develop and promote both policies and tools. The power and resources of such units are highly variable across countries.

The OECD have also reviewed the use of RIA specifically (OECD, 1997b). Most OECD countries now apply it in some form. It has been found to improve the efficiency and effectiveness of governments if properly designed and applied. The OECD has also instituted a series of voluntary country reviews through the Regulatory Reform Programme.

Table 4 – Regulatory impact analysis

What is it?
(RIA) at its most basic level involves asking a set of questions (see a sample below) to determine the robustness of a regulatory proposal. Asking, or even disseminating, the questions can itself improve awareness of regulatory design issues and hence the quality of outcomes. Ultimately, however, the effectiveness of RIA depends on how it is performed and how the results are used.
What does it ask?
  • What is the objective?
  • What is the problem?
  • Why is Government action required?
  • What are the options?
  • What are the costs and benefits of each option, and the associated risks and probability of success?
  • Why is the preferred option the best choice?
  • Have affected parties been consulted?
  • How will the regime affect the incentives of affected parties?
  • What enforcement mechanisms are required?[7]
  • How will the regime be monitored and evaluated?

The OECD notes that regulatory impact analysis is based on elements of public choice theory. Best practice requires maximum political commitment, clear allocation of responsibilities including a central oversight body, training, consistent but flexible analysis, good data collection strategies, targeting of effort, integration with policy making process as early as possible, clear communication of results, public involvement and coverage of both new and existing regulation.

The advantages of RIA stem mainly from the requirement to explicitly demonstrate that there is a real problem, that a range of options have been considered of which regulation is the preferred choice, that the benefits of regulating exceed the costs and that consultation has occurred. Its effectiveness is affected by the nature of external scrutiny of the RIA, and by the relevance of cost-benefit analysis (CBA) to the question at hand. Publicity of the results of the RIA can improve the incentives of those carrying it out. Central oversight also provides an independent perspective, quality control and specialised expertise (Bailey, 2001).

Regulatory review involves the application of RIA and other tools to existing regulations and is often focused on general goals such as reducing red tape and government formalities.

Strategies for regulatory review include (1) a zero-based reinvention at significant cost, (2) a review of a number of regimes against general criteria (such as the Australian National Competition Policy noted in Section 4.2), (3) sunset or automatic review clauses (commonly used for state regulation in Australia and now for local bylaws in New Zealand, see Section 4.3.2) and (4) equivalence rules which allow use of lower cost compliance methods if they are as effective as a prescribed regulatory approach.

Reviews involve a trade-off between the benefits of up-dating current arrangements and the use of resources and expertise which could be used to assess new proposals before sub-standard practices become embedded. Smaller countries have difficulty resourcing a regulatory quality regime on its own and may not be able to undertake significant regulatory review work.


  • [7]A useful checklist for strengths and weaknesses of regulation in terms of compliance is the Dutch Table of Eleven . This can be used to compare measures by assigning a score of 1 to 5 to each element (PC, 2003).

4.2  Other countries’ experience#

New Zealand’s application of RIA through Regulatory Impact Statements (RISs) was closely based on Australia, which itself drew on OECD experience. Australia re-issued its guide to RISs in December 1998 to strengthen the scrutiny of quasi-regulatory measures (Australian Government 1997 and PC, 1999a). Australian practice will be of growing importance for New Zealand with the increasing trends for trans-Tasman regulatory harmonisation (Guerin 2001 and 2002a).[8]

Australia has an Office of Regulation Review whose responsibilities include overseeing RIS compliance and publishing annual reports on departmental performance in this area (PC, 1999b, 2000, 2001 and 2002). Those annual reports cover compliance with RIS requirements in terms of whether statements were prepared and their adequacy for Commonwealth regulation, National Competition Policy Reviews of Commonwealth legislation, and on statements prepared for Ministerial Councils and national standard-setting bodies[9]. They can also address specific regulatory issues. The results continue to show incomplete compliance, including for significant proposals and a lack of integration of RIA into policy advice development. A recent report has suggested scope for improvements (see Table 5).

Table 5 – Options for improving regulatory quality in Australia

  • Integration of RIA into consultation processes
  • Better targeting and clearer guidance on threshold tests
  • More formalised coordination of regulatory review and RIS preparation within regulatory departments & agencies
  • Increased ministerial involvement and accountability
  • More effective sanctions for non-compliance
Other tools
  • Minimum standards for public consultation
  • Integrating preliminary impact assessments into regulatory plans
  • Strong independent regulatory reform advocacy body
  • Improved guidance materials and training on alternatives to prescriptive regulation and improved evaluation and sharing of experience with their use
  • Improved measurement of compliance costs
  • Regular and systematic monitoring and evaluation of the outcomes of regulatory review and reform strategies

Source: PC (2003)

The United States has operated RIA systems for two decades but still has problems such as agencies treating it as the end of rather than the starting point for analysis, lack of quantification of costs and benefits, lack of consideration of alternatives and inadequate transparency (Hahn, 2000).

Whether these experiences reflects the insolubility of such problems even with much greater resources than New Zealand can provide, or differences in political and institutional factors and the resistance of such institutions to change is an open question.


  • [8]Australia is also following international precedent in establishing a Federal Register of Legislative Instruments in which new regulations must be listed before they are enforced. In 1986 Sweden established such a register, required agencies to cull unnecessary or outdated regulations, and automatically cancelled hundreds of regulations not listed by the cut-off date. In 1995 the register was being built up electronically with text on issues such as motives, magnitude of costs and effects. A new Australian Bill proposes to require all legislative instruments in Australia to be registered before enforcement and impose a consistent 10 year time limit. “Legislation to clean up administrative black hole” The Australian Financial Review Friday 11 July 2003 p51.
  • [9]As part of their commitments under National Competition Policy (NCP), all Commonwealth, State and Territory governments undertook to review and change legislation that restricts competition. The National Competition Council was established by all Australian governments in November 1995 to act as a policy advisory body to oversee their implementation of National Competition Policy (NCP).

4.3  New Zealand‘s regulatory quality regime#

New Zealand’s regulatory quality regime is still in an early stage of development, but is already demonstrating the same issues as overseas jurisdictions.

4.3.1  Development

Cabinet originally agreed to Compliance Cost Assessment (CCA) requirements in November 1995 with CCA guidelines released by the Ministry of Commerce in January 1997. The Government agreed in 1997 to a Code of Good Regulatory Practice (the Code), based on the principles of effectiveness, efficiency, equity, transparency, and clarity, and a Generic Policy Development Process (the Process) (MoC, A and MED, B). It also agreed to require a Regulatory Impact Statement (RIS) with any policy proposal to Cabinet that would result in government Bills or statutory regulation unless an exemption applied (MoC, 1999 and Cabinet Office, A). This superseded the Compliance Cost Assessment Framework of 1995.

Table 6 – Regulatory impact statement requirements

When is an RIS required?
An RIS is required for any policy proposal with legislative implications that is submitted to Cabinet. There are a number of exemptions but these are (or are intended to be) for mechanical measures that do not raise significant policy issues. [10]
What does it cover?
As a general rule of thumb, an RIS should not exceed 3 pages (plus a BCCS if it has compliance cost implications for business). It must contain statements of:
  • the nature and magnitude of the problem and the need for government action;
  • the public policy objective(s);
  • feasible options (regulatory and/or non regulatory) that may constitute viable means for achieving the desired objective(s);
  • the net benefit of the proposal, including the total regulatory costs (administrative, compliance, and economic costs) and benefits (including non-quantifiable benefits) of the proposal, and other feasible options; and
  • the consultative programme undertaken.

In 1998 a Regulatory Responsibility Package was developed which covered a Regulatory Task Force; a Regulatory Responsibility Act,[11] and incorporating the Code and the Process into departmental performance assessment with a reference to the Code in the Cabinet Office Manual. This package did not proceed.

In August 1999 the Government announced the Five Steps Initiatives including small business test panels for new laws, removing redundant law (aiming to eliminate 12.5 to 25% of the existing regulatory stock, reducing tax compliance costs; keeping laws up to date (annual compliance cost reduction bill and more use of sunset or review clauses), better information on laws; and facilitating electronic commerce. Some test panels were established but little other progress was made.

In 2000 several initiatives were taken to reinvigorate the regulatory quality regime in New Zealand. The RIS was expanded to an RIS/BCCS from 1 April 2001 with the inclusion of a Business Compliance Cost Statement (BCCS) where appropriate. In 2003 to reduce the length of BCCSs, the Government agreed to include only a summary of key information on compliance costs in BCCSs covering the sources of costs, the parties affected, estimated costs and steps taken to minimise costs.

An RIS does not have to provide details of implementation or monitoring strategies, or refer to implications of the proposal for the Trans-Tasman Mutual Recognition Arrangement (TTMRA). These are potentially significant gaps in the current regime.

4.3.2  Reviews of New Zealand’s regime

In 2001 the Ministerial Panel on Business Compliance Costs was set up to report on measures to reduce compliance costs (MED, A and Cabinet Office, B)[12] and Tasman Economics were commissioned to review the RIS regime (Tasman Economics, 2001).

The issues identified in these reviews (see Table 7) are discussed below in the same groupings used in Section 3.3.

The core of regime design is coverage. The most significant issue here is the exclusion from the RIS/BCCS regime of quasi-regulatory measures such as guidelines and rules. The regime currently covers only regulation promulgated through Acts and Statutory Regulations made by Order in Council.[13] This sort of gap in coverage creates incentives for regulators to use whichever regulatory mechanism is subject to the least procedural controls and scrutiny.

The other coverage issue outstanding is how to ensure regulatory quality in the operation of powers delegated to local government. This was addressed in the Local Government Act 2002[14]. The remaining design issue identified in the review was a need for earlier preparation of RIS/BCCSs and greater use of test panels to evaluate proposed regulations. The government agreed to encourage these moves.

Transparency is one of the strongest incentives for improved regulatory quality as it makes failures more open to scrutiny both by those responsible for assessing bureaucratic performance and by those affected by the regulation. The latter are therefore better placed to challenge low quality regulation.

Table 7 – Proposed extensions to NZ regulatory quality regime
  Business Compliance Cost Panel 2001 Tasman Economics RIS Review 2001
  1. Mandatory use of the Code and the Process.
  2. Earlier preparation of RIS/BCCS and release with any consultation.
  3. Specific business compliance cost objective in the Process.
  4. Apply RIS/BCCS to all Bills.
  5. Longer consultation periods.
  6. Mandatory test panels for legislation needing BCCS.
  7. More focus groups.
  8. Plain English drafting.
  9. Updated and more accessible explanatory notes.
  10. More information and education on new legislation.
  1. Implement Code & Process.
  2. Staged RIS - Initial at start of work; Partial when seeking approval (released with consultation documents), Full for final decisions, Final on introduction of legislation.
  3. Rationalise structure of RIS and Cabinet Submissions.
  4. Allow aggregate RIS where multiple Cabinet papers on the proposed regulation.
  5. Clarify the scope of RIS regime and gradually expand to include quasi-regulation.
  6. Expand content of RIS to include summary, recommended options, implementation and review, and impact particularly on international agreements and business compliance costs.
  1. Key Performance Indicators in chief executives’ performance agreements.
  1. Clarifying accountability for RISs.
  2. Require accountable officials and Ministers to certify RIS meets specified requirements.
  3. Establish Government commitment to responsible regulation.
  4. Make a Minister responsible for quality of regulation.
  1. central unit to monitor and review RIS compliance and quality, and assist with preparation.
  1. Include information and education strategy in RIS/BCCS.
  2. Comment on compliance cost reduction in annual reports.
  3. Greater select committee scrutiny of compliance costs.
  4. Omnibus Compliance Cost Reduction Bill.
  1. Central register of RISs.

Recent measures to increase transparency should therefore improve incentives. Those measures include (1) electronic public access to current versions of legislation and regulations and increased use of plain English drafting, and (2) automatic release of RISs that contain a BCCS and the inclusion of RIS/BCCS issues or a draft RIS/BCCS in consultation documents.

The government has established a compliance cost unit within the Ministry of Economic Development to handle support, monitoring and publication of RISs.[15] This unit’s work includes an education campaign to improve awareness of the requirements and improve the ability of agencies to meet them.

One issue not directly addressed to date is whether there should be centralised approval or “vetting” of an RIS. This can be seen as the ultimate step in a process that begins with imposing the RIS requirement, secondly requires publishing the results, and then involves public assessment of performance before finally introducing formal controls or sanctions on RIS preparation. Whether and how quickly to move along this process is a question of judgement based on the costs and benefits at each stage. There are trade-offs in the short term between widening and deepening of the regime.

Another issue was whether to adopt the approach used in some overseas jurisdictions of having a central regulation register, with inclusion a prerequisite for enforcement and revocation of instruments not listed. This may become less relevant with improved public access to legislation and regulations online, but issues could remain for quasi-regulation.


  • [10]Exemptions apply where the proposal is (a) of a minor or machinery nature, (b) deals with administrative procedures and does not impact on business, consumers, or the public, (c) is required to meet an obligation under an international agreement and primarily repeats or adopts the terms of the agreement, (d) gives effect to a specific Budget decision relating to a tax, fee or charge; entitlement or obligation, or (e) is an Order in Council that provides solely for the commencement of enabling legislation or any provision of enabling legislation.
  • [11]This would have required regulations to be based on sufficient information, realistic alternatives to regulation to be considered, distributional impacts to be assessed, benefits to outweigh costs, routine disclosure of RISs, departmental annual reporting on regulatory activity and certification by Ministers of papers’ consistency with the Act.
  • [12]“Finding the Balance: Maximum Compliance at Minimum Cost”, report of the Ministerial Panel on Business Compliance Costs was released on 11 July 2001, . The Government’s response “Striking the balance: response to Ministerial Panel on Business Compliance Costs” was released on 18 December 2001, . A report back on progress was provided in June 2003, .
  • [13]For Australian work on quasi-regulation see PC (1999c).
  • [14]Under the Local Government Act 2002, where a bylaw is made, amended in more than an editorial or minor way, reviewed or revoked the local authority must follow the special consultative procedure which includes making available a statement of proposal which includes the draft bylaw, releasing a summary of the statement, giving public notice and allowing at least a month for public submissions. The summary must fairly represent the major matters in the proposal, be widely distributed indicate where the statement may be inspected and a copy obtained; and state the period for submissions. The authority must consider whether a bylaw s the most appropriate way of addressing the perceived problem. whether the proposed bylaw is the most appropriate form of bylaw and whether it gives rise to implications under the New Zealand Bill of Rights Act 1990. A 5 year review requirement now also applies for all existing or new bylaws, with a rolling 10 year review requirement from then on. Reviews must apply the same considerations as for new bylaws and use the special consultative procedure. If a review is not carried out, the bylaw expires 2 years after the review deadline.
  • [15]Although there is conflict between the education and enforcement roles, and a unit outside any policy agency might be more desirable in terms of independence of advice and ability to take a “whole of government” view.

4.3.3  Future paths

Improving the quality of regulatory design and implementation in New Zealand at base depends on the creation of institutional arrangements which are more “incentive compatible”; ie, the creation of rules which align the goals of the political and bureaucratic agents with those of the political entity; thereby encouraging the key players to pursue higher quality regulation (Bailey, 2001). This is, however, much easier to suggest than to achieve, either in theory or practice.

Direct quality control of RISs is difficult due to the subjective nature of the analysis, the concentration of public-sector expertise within the responsible agency, and the conflicts inherent in centralised bureaucratic control of advice to Ministers. Achieving such an effect through rewards or penalties in the public sector therefore requires indirect pressures such as feedback on departmental performance, formal evaluation of outputs (eg, RIS reviews) and greater transparency to the public. Mandatory disclosure of an RIS is particularly useful in this context as the concise and clearly laid out format is usually much more accessible than the underlying policy advice and research.[16]

There are further options for strengthening the central RIS monitoring function, introducing public reporting on regulatory performance by departments as occurs in Australia, publishing all RISs, and requiring staged development of the RIS (with an early version to be included in public consultation on proposals and a final version prepared after legislation or regulation is finalised). Whether these gains are more than marginal and whether the benefits exceed the costs remains to be worked through.

Although greater transparency is easier to achieve than improved capability, the latter is an important tool, particularly as the benefits could flow through to all aspects of the public sector not just regulatory design. Improving state sector capability is a priority of the current government, including encouraging a whole-of-government approach to address the lack of cross-agency co-operation which tends to reflect the lack of (1) a clear mandate[17], (2) a clearly defined focus or purpose which is measurable and linked to overall government goals[18], (3) appropriate membership and leadership of the cross-agency structure[19], and (4) effective governance and accountability[20].

Correcting these problems is an issue of wider public sector governance which may be addressed through initiatives such as the Review of the Centre.[21] One approach suggested overseas is to facilitate budget transfers between agencies (e.g. from police and local authorities to schools for additional classes to address problems caused by disruptive students) to manage the externalities of an agency’s decisions (James, 2000).

Procedures for appointment, monitoring and remuneration of public sector staff is a key element of capability but clearly is much wider than regulatory functions and will accordingly be left for analysis elsewhere. Another aspect of capability that appears likely to offer specific benefits for regulatory quality is better education and training in public policy analysis, including use of the assessment tools offered by the existing regime such as RIA itself and the guidance material available in the New Zealand context (MOC, 1999 and MOC (A), MED (A) and MED(B)).

Another possible tool to improve regulatory quality is regulatory competition through methods such as performance comparisons between local authorities or allowing competing regulatory bodies. The usefulness of this for New Zealand is likely to be limited given scale issues, and the increasing need to harmonise regulatory arrangements with other countries (particularly Australia). However, possibilities do exist particularly for comparisons between local authorities or with Australian regulators, or in narrow fields such as certification authorities where more than one choice exists for those seeking oversight of their activities.

The increasing use of joint institutions to design or administer regulatory regimes on both sides of the Tasman offers new challenges to regulatory quality. Issues of accountability and participation are discussed in previous papers (Guerin 2001 and 2002a), but that leaves the question of whether regulatory impact analysis of the actions of such bodies is appropriate, and if so under what rules it would be carried out and by whom.

There may also be scope for increased use of sunset clauses for regulations, although the frequency of review would need to take into account the resources required.


  • [16]Also appealing, but facing even greater impediments, is the concept of a “regulatory budget” recording the actual costs imposed by regulators and requiring proposed regulations to be prioritised within a budget limit. The practical difficulties would, however, be immense so this option is not addressed further in this paper.
  • [17]A mandate is needed from Ministers and/or Chief Executives (1) authorising and requiring agencies follow a co-operative process, and (2) committing themselves to support that process. Without such a mandate there will always be a risk of agencies not fully committing resources to the joint outcome, and attempting to bypass the agreed process.
  • [18]Unless agreement on this is reached up front, or at least early in the process, the functioning of the grouping is likely to be severely compromised.
  • [19]If the group excludes any agencies which have necessary resources, knowledge or contacts, or whose support is necessary to deliver on the purpose, then chances of success are reduced.
  • [20]Defined reporting procedures, clear accountability for the overall purpose and individual agency objectives, agreed governance arrangements including dispute resolution, agreed duration of the grouping, and resourcing and information requirements have been addressed. Having these details in place tends to reduce friction, whether or not they are used. This area leaves plenty of scope for variation, depending on which the details are specified in advance and which are left up to the grouping to determine for itself.
  • [21]The Review of the Centre is co-ordinated by the State Services Commission.

5  Conclusion#

When government is evaluating regulation, it is crucial that the total costs and benefits to society are considered, as far as possible, rather than focusing on fiscal costs to government or compliance costs to business or consumers and ignoring less explicit costs (ie, falling victim to fiscal illusion). To increase the likelihood of this happening, and of thereby reaching a higher quality decision, we need to consider the incentives on all those involved in the decision-making process, as well as assessing fundamental design issues such as education and enforcement to ensure the desired outcome is feasible.

The standard mechanism for achieving these objectives is through a regulatory quality regime requiring consideration of all of the above factors and providing for processes to ensure such consideration actually occurs and is properly communicated to decision-makers and other stake-holders. This regime needs to be fully integrated into governmental processes and reinforced from the highest levels of government. New Zealand has made progress towards such an outcome, but still has room to improve.

The most significant areas for possible improvement in regulatory design in New Zealand appear to be in enforcement and oversight of the Regulatory Impact Statement regime and carefully designed extensions, such as stronger requirements to build monitoring and evaluation strategies into new regulatory proposals, and extending the scope of the regime to cover types of regulation that do not involve statutes or Orders in Council (the new bylaw impact requirements are an example).

Achieving high quality regulation requires incremental progress on a number of mutually supporting initiatives, with continued reinforcement of the underlying messages and careful building of the necessary institutions and practices. That progress will have to occur in the context of increased interaction with Australian practices through institutions such as the Council of Australian Governments and the obligations of the Trans-Tasman Mutual Recognition Arrangement.

Ultimately the environment must require who “are too willing to believe that there is a legislative and regulatory answer to every problem” to establish both the significance of the problem and the merits of the solution and its alternatives both before they proceed and after the event (Lochner, 2000:841).


Australian Government (1997) “Council of Australian governments - principles and guidelines for national standard setting and regulatory action by ministerial councils and standard-setting bodies.”

Bailey, Steven (2001), “Regulating the regulators: The quest for quality regulation” Unpublished

Becker, Gary S (1983) “A theory of competition among pressure groups for political influence”, XCVIII(3): 371-400 The Quarterly Journal of Economics

Buchanan, James, M. (1987) “Public finance in democratic process: fiscal institutions and individual choice”,

den Hertog, Johan (1999), “General theories of regulation.” In B Bouckaert and G De Geest eds “Encyclopedia of Law and Economics.” Cheltenham: Edward Edgar

Esty, Daniel, C (1999) “Toward optimal environmental governance”, 74: 1495-1574 New York University Law Review

Guerin, Kevin (2001) “Regulatory harmonisation – issues for New Zealand.” Wellington, New Zealand Treasury, Working Paper 01/01>

Guerin, Kevin (2002a) “Subsidiarity: implications for New Zealand” Wellington, New Zealand Treasury, Working Paper 02/03.>

Guerin, Kevin (2002b) “Protection against government takings: compensation for regulation?” Wellington, New Zealand Treasury, Working Paper 02/18>

Guerin, Kevin (2003) “Property rights and environmental policy: a New Zealand perspective” Wellington, New Zealand Treasury, Working Paper 03/02.>

Guesnerie, Roger (1988) “Regulation as an adverse selection problem”, 32: 473-481 European Economic Review

Hahn, Robert (2000) “Assessing regulatory impact analyses: the failures of agencies to comply”, 23(3): 859-885 Harvard Journal of Law & Public Policy

Hahn, Robert and Hird, John (1991) “The costs and benefits of regulation: review and synthesis”, 8: 233-278 Yale Journal on Regulation

Holcombe, Randall and Holcombe, Lora (1986), “The market for regulation”, 142: 684-696 Journal of Institutional and Theoretical Economics

James, Oliver (2000), “Regulation inside government: public interest justifications and regulatory failures”, 78(2): 327-343 Public Administration

Lochner, Philip (2000), “Economic regulation and democratic government”, 25(4): 831-841 Journal of Corporation Law

Neven (1992), “Regulatory reform in the European Community”, 82:2 AEA papers and Proceedings, 98-103

OECD (1997a), “The OECD report on regulatory reform.” Paris, Organisation for Economic Co-operation and Development

OECD (1997b) “Regulatory impact analysis: best practices in OECD countries.” Paris, Organisation for Economic Co-operation and Development

OECD (2001) “Regulatory policies in OECD countries: from intervention to regulatory governance.” PUMA/REG(2001)10, Paris, Organisation for Economic Co-operation and Development

PC (1999a) “A guide to regulation – second edition.” Office of Regulation Review - Productivity Commission, Canberra

PC (1999b) “Regulation and its Review 1998-99.” Office of Regulation Review - Productivity Commission, Canberra

PC (1999c) “Grey Letter Law - Report of the Commonwealth Interdepartmental Committee on Quasi-regulation.” Office of Regulation Review - Productivity Commission, Canberra

PC (2000) “Regulation and its Review 1999-2000.” Office of Regulation Review - Productivity Commission, Canberra

PC (2001) “Regulation and its Review 2000-01.” Office of Regulation Review - Productivity Commission, Canberra

PC (2002) “Regulation and its Review 2001-02.” Office of Regulation Review - Productivity Commission, Canberra

PC (2003) “Mechanisms for improving the quality of regulations: Australia in an international context.” Staff Working paper, Productivity Commission, Canberra

Peltzman, Sam (1976) “Towards a more general theory of regulation”, 19: 211-240 Journal of Law and Economics

Peltzman, Sam (1989) “The economic theory of regulation after a decade of deregulation”, 1-41 Brookings Papers: Microeconomics

Posner, Richard (1974) “Theories of economic regulation”, 5:335-358 Bell Journal of Economics and Management Science

Stigler, George (1971) “The theory of economic regulation”, 2: 3-21 Bell Journal of Economics and Management Science

Tasman Economics (2001) “Regulatory Impact Statement Review.” Tasman Economics, Melbourne

New Zealand - Guidance Documents#

Cabinet Office (A) CO (98) 5 Regulatory impact statements>

Cabinet Office (B) CO (01) 2 Business compliance cost statement>

Cabinet Office (C) “Step-by-step guide”>

MoC (Ministry of Commerce) (1999) “A guide to preparing regulatory impact statements”

MoC (Ministry of Commerce) (A) “The generic policy development process”

MED (Ministry of Economic Development) (A) “Business compliance cost statements – guidelines for departments”

MED (Ministry of Economic Development) (B) “Code of good regulatory practice”>