Rayner Thwaites* and Dean R Knight§

Short of simply preventing an activity, supervising, controlling and restraining its performance necessarily involves a degree of participation in the activity. ... The danger in this is not that the supervisor will become the performer … but that the values of the supervisor will come to dominate performance, thus breaking the tension between supervision and performance.[1]

8.1 Introduction

This chapter focuses on the relationship between the supervisor and performer in the review of, or appeal from, administrative decisions which are regulatory in character.[2] In particular, this project will explore whether and when appellate review of such decisions is appropriate and, if it is appropriate, in what form. As a necessary part of that inquiry we also examine the complementary (or, if no provision is made for appellate review, “default”) judicial oversight mechanism: judicial review. We are particularly concerned with the circumstances in which the supervision targets, or should target, the substance or merits of regulatory decisions and any peculiarly New Zealand dimensions to that question. We do so against the background of calls for increased deployment of appeal on the merits, most vividly seen in the Regulatory Standards Bill (the Bill) introduced into Parliament in 2011.[3]

The overarching purpose of this project is to analyse and reflect on the performance–supervision dynamic within New Zealand’s administrative law setting and jurisprudence. Our goal is two-fold. First, we consider whether it is possible to offer generalised conclusions about the most appropriate form for supervision of regulatory decisions, particularly the appropriateness of providing for an appeal on the merits. Secondly, anticipating the need for a contextual approach, we aim to identify and expand upon the key considerations which impact on this question.

Within the broader work of the New Zealand Law Foundation Regulatory Reform Project, the issue of appeal and review of regulatory decisions can be considered a matter of “little policy”.[4] That is, we are concerned with an aspect of how best to translate policy into a legal framework.[5] We are not concerned with the substantive content or direction of policy in any particular area of regulation, but rather the question of how best to implement and give effect to “big policy”. The lodestar for our analysis is, therefore, the need to ensure the framework for judicial supervision, on the one hand, provides adequate checks-and-balances on the exercise of regulatory power and, on the other hand, remains faithful to, and does not unduly interfere with, the policy objectives underlying any particular area of regulation.

Our primary audience is those involved in designing the processes for the supervision of regulatory decisions. This audience can be broken into two broad groups. The first group is composed of those involved in developing the relevant statutory regimes (typically, but not always, through the provision of a statutory appeal). This group includes drafters, policy-makers, legislators and other stakeholders who participate in the development of these regimes.[6] The second group comprises those responsible for interpreting the statutory scheme or developing common law principles which have the effect of defining the nature of the supervisory function. Most obviously this includes the judges responsible for defining the judicial method to be applied in common law judicial review and on appeal, along with the lawyers who argue such cases. Judges are also responsible for defining the appropriate judicial method for statutory appeals where legislation does not definitively prescribe the appropriate method. Both groups together might be able to be described as the designers.

This chapter forms part of the first stage of the Regulatory Reform Project and is therefore an exploratory one which aims to identify and scope the relevant issues. Later stages of the project will contain more detailed analysis of the performance–supervision dynamic and key considerations bearing on the appropriateness of the different forms of appeal and review. In this chapter, we introduce the key concepts within the performance–supervision relationship. We identify the participants in this dynamic and characterise the regulatory decisions subject to supervision. We then identify the different methodologies within both review and appeal, to demonstrate the pluralistic suite of supervision options available as part of the existing status quo and to gauge their respective intensities of review. We close by offering some reflection on the appropriateness of collapsing this diverse and context-sensitive set of supervision methodologies, and instead favouring one particular approach, as is proposed in the Bill.

8.2 Key concepts: what? who? by whom? how?

In New Zealand, there has been a recurrent argument for merits review of regulatory decisions,[7] as exemplified by the provisions of the Bill discussed below. We begin by locating the plea for appeal on the merits from regulatory decisions within its wider context, particularly the New Zealand context.[8] First, it is necessary to appreciate the form and character of regulatory decisions. What do we mean by a “regulatory” decision – the what question. Closely related to this is the who question: who makes these regulatory decisions? Or, framed in the negative, whose decisions are being challenged? The answers to these questions contribute to our understanding of the first limb of the performance-supervision dynamic and allow us to consider whether the request for an appeal on the merits from regulatory decisions is capable of a generalised response.

Secondly, the supervision limb must be delineated. Answering the by whom question identifies our focus in this project on judicial supervision (or review by independent tribunals which is similar in character). In doing so, we locate this formal judge-based, accountability mechanism amongst the broader suite of accountability mechanisms within administrative law. We then examine the two different forms of judicial supervision – appeal and review – to address the how question. This allows us to chart the contours of appeal and review, providing some comparative context to assess the call for appeal on the merits. In doing so, we focus on the main differences between appeal and review, as well as identify the different judicial methodologies within each form of judicial supervision.

As part of our cartography of the regulatory arena, we have commenced an empirical survey of the appeals and reviews of regulatory decisions. The purpose is to develop a clear profile of the performance-supervision dynamic in New Zealand over the last decade, providing empirical data to test and inform the discursive analysis. So far, the survey is only partial and provisional; it is our intention to present a much fuller study as part of future stages of this project. Nevertheless, while the study is in its formative stages, we have been able to draw on some of the data to help sketch in more general terms the character of the regulatory field.

8.2.1 What: the types of regulatory decisions

8.2.2 Who: the bodies and officials who make regulatory decisions

8.2.3 By whom: who is responsible for supervising the regulators?

This paper focuses on judicial supervision through the mechanism of judicial review and statutory appeals. These are two different court-centred mechanisms for dealing with disputes between government and the governed and ensuring that a decision maker is responsible for his or her decisions. Appeal and review represent the most overt and formal accountability processes for regulatory decisions. They are also the control mechanisms which are increasingly subject to criticism and debate.

We acknowledge that judicial supervision is by no means the only accountability mechanism applying to regulatory decisions. There are also a wide range of other accountability mechanisms, such as internal review, public transparency,[20] and democratic oversight.[21] These other accountability mechanisms also help control the exercise of regulatory power to differing degrees.[22]

8.2.4 How: how is supervision conducted?

(a) Introduction

In this context, when we refer to supervision of a regulatory decision, we are referring to the review by an external body after such a decision has been made. As we noted earlier, appeal and judicial review are two of the main (external) modes of supervision of regulatory decisions.[24]

We consider it is desirable to separate out the development and implementation of a regulatory rule and supervision of the implementation of that rule (or, as Cane describes it, adjudication on disputes arising from implementation).[25] This distinction is founded in the separation of powers. It is based on the idea that the decision-maker implementing the rule (or, in some cases, promulgating the rule) on the one hand, and the body charged with adjudicating that implementation on the other, represent autonomous centres of power with different values.[26] The case for a division between implementation and supervision rests on a constructive tension between these different values. The emphasis of a person charged with implementing a rule or regulation is to give effect to its purpose. The emphasis is on the social objective of the regulation. In contrast, the emphasis of the supervisor is on the legality of the regulation, the process by which it is developed and the protection of individual rights.[27] Similarly, those charged with developing rules and those charged with supervising their promulgation come at those tasks with different values.

A central concern of administrative law always has been, and remains, how to guard against the risk that those supervising an activity will become too involved in that activity, as is punctuated in the opening quote to this chapter.[28] In designing a regulatory framework, the aim is to provide for supervision that does not break the tension between supervision and performance. The nature of both appeal and judicial review vary, as we explore in more detail shortly. Both seek to respect, in different ways, the values underlying the performance-supervision dynamic, particularly in the ways they seek to respect (or not) the judgement of the primary decision-maker on the merits of the regulatory decisions under review.

The statutory framework for appeals is a first attempt, by the legislature, to provide a working balance between supervision and performance for particular contexts. This is augmented by the courts, as they develop principles to regulate the judicial method for appeals.[29]

In judicial review, the development of doctrine regulating the judicial method of the supervising court remains the task of the courts themselves, the court’s decision shaped by the interaction of doctrine with the legislative scheme and context.[30] Judicial review therefore involves two different discretions: doctrine is intended to “structure judicial discretionary control of exercises of administrative discretionary power”.[31] Judicial review doctrine may be seen as implicitly containing standards of legality (focused on the requirements a primary decision-maker must conform to) and standards of review (focused on the appropriate methodology to be applied by the reviewing court).[32] The latter typically involves an assessment of the nature of the decision and the decision-maker, as a necessary first step toward determining the intrusiveness of review. The development of more intrusive forms of judicial review has been accompanied by an increasing focus on questions of the appropriate “weight”, “latitude”, “respect” or “deference” that should be accorded to decisions under review.[33]

We suggest that doctrinal developments can be mapped on a continuum between vigilance and restraint, recording the degree of intensity applied by the supervising courts to the regulatory decision under review.[34] This allows us to identify the extent to which existing judicial methods presently provide for, or do not provide for, review of the merits.

Our initial mapping of the contours of both appeal and review demonstrate that both mechanisms are pluralistic, with each representing variable forms of judicial methodology and intensity depending on the context and circumstances. Appeals are generally, but not always, more vigilant than judicial review. There is, however, some overlap and convergence between the two.

The relationship between the two modes of review is important. In the absence of statutory provision for appellate review, or of an attempt to restrict the scope of judicial review, common law judicial review is the default mode of supervision. Reciprocally, judicial review is discretionary and relief may be declined where the court considers an appeal the more appropriate remedy. Making provision for an appeal on the merits exposes the regulator to a degree of supervision higher than that involved in other forms of review. In many instances such a commitment is desirable. But any decision to opt for an appeal on the merits should be the outcome of a considered assessment that starts with the nature of the decision under review and considers the form of review appropriate for it.

The present proposal for an appeal on the merits as part of the Regulatory Standards Bill appears to contemplate a strong presumption in favour of one particular type of appeal, namely an “appeal on the merits”.[35] This is not a term of art,[36] and exactly what is intended is unclear, as we explore later. Finally, judges and commentators, rightly or wrongly, view proportionality as an intrusive form of review, going to the merits of a decision.[37] While it is not equivalent to merits review, it may be more intrusive than more longstanding doctrines of judicial review.[38]

[A]n appeal by way of re-hearing is] is more expeditious than a hearing de novo because of its focus on specific alleged errors, but not as restrictive as an appeal stricto sensu. Indeed, an appeal should focus on specific alleged errors. In general, there is no need to provide an opportunity to re-litigate the whole matter, as in a hearing de novo, unless there is good reason not to presume that the first instance decision-maker correctly ascertained the facts. The added cost of a complete rehearing generally counts against this procedure. An appeal should not be by way of case stated unless there is some reason why this option is preferable to an ordinary appeal limited to questions of law.

Provision for such appeals is therefore relatively common. Civil procedure rules adopt appeal by way of rehearing as the default approach, in the absence of specific legislation mandating otherwise.[59] This mode of appeal is also adopted in relation to appeals from some professional bodies.[60]

In summary, the types or forms of appeals outlined above, and the procedural restrictions discussed, demonstrate that judicial supervision by way of appeal varies considerably, with more vigilant and more restrained approaches evident. In some cases, there may be a degree of convergence or analogy with judicial review methodology. However, many (but not all) of the types or forms of review generally contemplate more intensive review than is usually available in judicial review. This is particularly so for those appeals where the appeal provision includes a legislative mandate for the appellate court to consider matters of fact and judgment afresh.

(c) Judicial review

The secondary or supervisory nature of judicial review is emphasised frequently in discussion of the proper approach to review.[64] Long-standing mantras urge the reviewing court not to usurp the role of the public body or official under review:[65]

Judicial review is concerned, not with the decision, but with the decision-making process. Unless that restriction on the power of the court is observed, the court will in my view, under the guise of preventing the abuse of power, be itself guilty of usurping power.

The nature of the alleged error is no longer seen to be as determinative as it once was, as the difficulties of classification are recognised. As Professor Taggart said, “many of the dichotomies upon which administrative law has rested – appeal/review, merits/legality, process/substance, discretion/law, law/policy, and fact/law – are no longer seen as giving as much guidance as they once did”.[70] At the front-line, the recent edition of Judge Over Your Shoulder acknowledges that, on a practical level, it is sometimes “difficult to completely sever” process from the merits.[71] Our courts have also been quite eager to uphold the principle of contextualism – or, as it has been famously put, “in law, context is everything”.[72] The strong emphasis on context means it is difficult to articulate a generalised approach applicable throughout the supervisory jurisdiction.[73] Modern judicial review involves a large dose of (overt and covert) variability of judicial method, driven by contextualism. While our senior judges seem reluctant to explicitly embrace the concepts of variable intensity or deference, variability is inherent is much of the courts’ supervisory method.[74]

The orthodox grounds of review – Lord Cooke’s “fairly, reasonably and in accordance with the law”[75] or Lord Diplock’s “illegality, irrationality, and procedural impropriety”[76] – present a picture of prescribed degrees of intensity. Under the illegality or in accordance with law ground, the courts adopt an extremely vigilant approach, where the supervising court is charged with determining the proper interpretation of law or the determination of the elements required by the legislation or legal context.[77] Similarly, under the procedural impropriety or fairness ground, the courts proclaim their ability to intervene if there has been any defect in the process. In contrast, judicial deference or restraint is generally applied to the merits, whether that be ordinary fact-finding, judgment, discretion, weight, or balancing, under the well-known Wednesbury principle.[78] But, arguably, even this categorical approach to review exhibits elements of variability. In many cases, the tripartite grounds overlap, merge, and tend to break-down.[79] The classification or framing of the error as one of law, process or merits therefore becomes crucial in the determination of the approach adopted by the reviewing court. Such classification is far from straightforward. It involves a degree of judicial judgment and appears to vary according to the context.[80]

Our courts have also overtly departed from these categorical or doctrinal grounds in particular cases or contexts.[81] This is most pronounced in two areas. First, the courts have applied doctrines which enable greater scrutiny of the merits of a decision. Some judges have promoted more vigilant formulations of the unreasonableness ground of review.[82] There is now some acceptance that less deferential forms of the reasonableness ground of review may be appropriate.[83] Others have dabbled with other substantive sub-grounds that involve more intense scrutiny (albeit with differing degrees of acceptance).[84] Substantive legitimate expectation and proportionality are the two sub-grounds which have assumed general legitimacy in some particular contexts. Notably, proportionality is entrenched as the most appropriate methodology for review in cases under the New Zealand Bill of Rights Act 1990,[85] although its application to the balance of administrative law cases has so far been resisted (and continues to be debated).[86]

Secondly, the courts have been willing to restrict or circumscribe the traditional grounds of review in particular circumstances, because of concerns about the suitability of particular questions to judicial adjudication. Non-justiciability might be absolute, with judicial review being entirely unavailable.[87] Review may still be permitted, but on more limited grounds than usual.[88] This form of limited justiciability is commonly seen in supervision of quasi-public bodies like SOEs,[89] review of commercial decisions of public bodies,[90] and other instances of public law adjudication in the commercial sphere.[91]

Given the loose analogy drawn by some between an appeal on the merits and proportionality review, the deployment of proportionality is of some interest in our project.[92] While we recognise that proportionality has been promoted as a doctrine which enables greater consideration of the merits, especially in human rights cases, we are sceptical about the extent to which that claim can be generalised. Proportionality is a methodology which structures the assessment of the necessity, suitability, and appropriateness of an administrative measure.[93] As such, unlike a claim that a decision will be subject to appeal on the merits, to say that a decision will be subject to proportionality review does not directly address the intrusiveness of judicial scrutiny.[94] That said, the distinction between proportionality and appeal on the merits should not be overstated, in terms of the standard or intensity of review they involve. There is a sense in which an invitation to the courts to engage in a proportionality analysis does invite the courts to engage with the merits of a decision, more so than with more traditional administrative law methodologies employed to examine the substance of a decision.[95] But any merits review undertaken under the guise of proportionality usually relates only to one aspect of the decision (namely whether the decision infringes the relevant right or rights disproportionately)[96] and is unlikely to be equivalent to the full merits review of a decision.

(d) Conclusion

The two formal judicial mechanisms for supervising regulatory decisions – appeals and judicial review – are therefore both variable in nature. The degree of intensity brought to the task of supervision differs within each mode of review and, in some cases, overlaps. It is therefore overly simplistic to approach the task of designing supervisory frameworks for regulatory decisions on the basis of a choice between two binary mechanisms: review vs appeal. An approach more alive to the potential impact of different forms of appellate review (or their absence) on the supervisory relationship is required.

The relevant “principles of responsible regulation” are stated as follows:

7 Principles

(1) The principles of responsible regulation are that, except as provided in subsection (2), legislation should –…

(g) if the legislation authorises a Minister, a public entity, or a public official to make decisions that may adversely affect any liberty, freedom, or right of the kind referred to in paragraph (b):

(i) Provide a right of appeal on the merits against those decisions to a court or other independent body; and

(ii) State appropriate criteria for making those decisions. …

(2) Any incompatibility with the principles is justified to the extent that it is reasonable and can be demonstrably justified in a free and democratic society.

The kind of “freedom, liberty or right” that, if adversely affected, triggers the strong presumption of an appeal on the merits is “a person’s liberty, personal security, freedom of choice or action, or rights to own, use and dispose of property”.[103] With the relevant interests so broadly and vaguely defined, there is a strong presumption that a wide swathe of regulatory action will be subject to an appeal on the merits by reason of adversely affecting a relevant interest.

All the principles, including the provision for appeal on the merits, are subject to a limitation requirement, substantially similar to that contained in s 5 of the New Zealand Bill of Rights Act.[104] The onus is placed on the government to justify, to the Parliament and the courts, a failure to provide for an appeal on the merits.[105]

It remains uncertain whether this proposed regime will become law.[106] The Bill has been subjected to stinging criticism from experts and officials,[107] but at the time of writing continues to advance through the legislative process. We remain sceptical about its utility. Our focus in this paper is on its blunt approach to supervision, namely, the favouring of appeal on the merits as the ideal form of supervision. Our concerns are:

• that the problem the Bill is intended to address is not clearly identified, lending a lack of clarity to it;

• that it is informed by an assumption that a simple intervention will result in a simple regulatory system;

• that it further assumes that flexible guidelines can readily be translated into a legislative requirement;

• that there is a general failure to speak to the costs of such a proposal; and

• that it shows little awareness of how it might juridify policy.

A fundamental problem with the Bill is that a diagnosis of the cause of poor quality regulation, the necessary first step in crafting any response, is undeveloped. As Treasury’s Regulatory Impact Statement noted, the Regulatory Standards Bill (and its predecessor Taskforce report) fails to meet this requirement:[108]

The case for new requirements rests in part on a conclusion that changes to existing arrangements won’t substantially improve legislative quality. The Taskforce indicated it was satisfied on this, though it did not really explain why …

One criticism of the Taskforce’s report is that it fails to provide a sound, evidence based explanation of what causes bad law, to justify its recommendations. To be fair to the Taskforce it was not asked to do so …

We agree with the view, implicit in Treasury’s Regulatory Impact Statement, that the Regulatory Standards Bill does not live up to its own ideals for subjecting regulatory interventions to a rigorous vetting process.[109] Gill rightly identifies the underlying problem with the Bill: “Without a clear exposition of the cause of poor quality regulation, it is difficult to prescribe an intervention or package of interventions that will be focussed on the source of the problem.”[110]

The lack of problem diagnosis undermines the blanket endorsement of an appeal on the merits. Promoters of the Bill argue that the mechanisms it contains have the merit of simplicity.[111] We contend that the vagueness of key terms alone makes implementation of the Bill anything but simple. But even if it is conceded that mechanisms contained in the Bill are simple, the presentation of the simplicity point fails to adequately distinguish between a simple intervention and the simplicity of the system resulting from that intervention.

The Taskforce report suggests some latitude as to what provision for “appeal on the merits” requires. It states that “the principle does not specify the form of the appeal (that is, whether it is by way of rehearing, or de novo), or the burden to be overcome by the appellant”.[112] As to what the term does specify, the intention appears to be that “appeal on the merits” demarcates a form of supervision which is both distinct from, and more intrusive than, judicial review.[113] The Taskforce said this principle is “currently recognised” in the Standing Orders for the Regulatory Review Committee, namely in SO310(2)(d) which states that one basis for the Committee drawing a regulation to the attention of the House is that it “unduly makes the rights or liberties of persons dependent on an administrative decision not subject to review on its merits by a judicial or other independent tribunal”.[114] In its Regulatory Impact Statement on the Bill, however, Treasury voiced the concern that the Bill is based on “too strict a distillation of more flexible guidelines gathered from other sources”.[115] We are similarly doubtful that the principle guiding the Regulations Review Committee's review of subordinate legislation (a principle necessarily framed in contingent form by the adoption of the qualified "unduly") can effectively be transposed into a principle of general application to primary legislation whenever an ill-defined range of interests is adversely effected.

The proposal can be understood to build on reports by the Regulations Review Committee in which judicial review has been held to be an inadequate “appeal mechanism for an administrative decision”, and regulations that rely on judicial review in the absence of fuller appeal rights criticised accordingly.[116] Insofar as the Taskforce was proceeding on the basis that concerns expressed by the Regulations Review Committee in particular cases should be translated into a general presumption that matters currently supervised by judicial review should be the subject of an “appeal on the merits”, the point is seriously under argued. There is no mention of the fact that for a sub-set of cases with which the Taskforce is concerned, involving infringement of existing rights, judicial review is often at its most intrusive.[117]

These comments all issue from our underlying concern with the Bill – that there has been no presentation of a diagnosis of the causes of poor quality regulation to justify the measures proposed. Any proposal for a regulatory intervention of far-reaching scope, such as that contained in the Bill, needs greater analysis and reflection about the underlying mischief it seeks to address and the costs involved. It needs to draw on a fuller accounting of the resources and cost involved, both directly, in making provision for appellate review, and indirectly, in the form of the demands and requirements on others generated by the demands of appellate bodies. Most importantly, in our view, any such proposal needs to accommodate the tension between supervision and performance. In other words, to what extent is it desirable that a set of values appropriate to appellate review should come to dominate the full range of regulatory decision-making subject to review?

While it is not the purpose of this project to directly and fully evaluate the virtues or otherwise of the appeal on the merits provision mandated in the Regulatory Standards Bill, we have sought to locate this proposal within the broader context and to identify key concerns raised for the performance–supervision relationship in prioritising an appeal on the merits over other forms of accountability and control. Future work in the project examining the performance-supervision relationship more generally will also contribute to any evaluation of the desirability of reforms to the supervision of regulatory decisions.

8.4 Conclusion and issues for future examination

Ultimately, in this project we will investigate the context and variability inherent in the performance-supervision dynamic. The underlying factors that, in the regulatory context, influence that variability will be examined. In this chapter we have, on a provisional basis, identified a range of different modes and methodologies of supervision which provide for greater or lesser scrutiny of the merits of regulatory decisions. We have also identified the incongruity between the present pluralistic supervision processes and the favouring of one particular form of supervision – an appeal on merits – proposed in the Regulatory Standards Bill.

In the future stages of this project, we will undertake a principled analysis of the factors underlying these variable approaches. The Legislation Advisory Committee guidelines identify relevant factors in the context of appeals, but consistent with the guidelines’ intended use, do not develop them in much detail. A principled analysis of variability of review involves a critical examination of the factors relied on to justify supervisory restraint (such as relative expertise, administrative efficiency, time and cost, the need for finality, and the dangers of over-legalisation). The arguments advanced in favour of more vigilant supervision (such as the potential for “better” regulatory outcomes, increased trust and confidence in administration, and avoiding adverse economic costs) will also be assessed.

Our approach to these issues, working from scholarship and empirical research, is intended to frame the relevant considerations involved in determining the particular mode and methodology for supervision of regulatory decisions, rather than directing the appropriate form of review in particular circumstances. In this way, a scholarly lens can inform an evaluation of current calls to reform the mode and form of supervision of regulatory decisions.